{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1QiCFLer1FIe"
      },
      "source": [
        "**Lab 14 – Gradient boosting and ensemble learning**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vCyq3-8y1FIj"
      },
      "source": [
        "_This notebook contains the sample from https://github.com/ageron/handson-ml3/blob/main/07_ensemble_learning_and_random_forests.ipynb, https://github.com/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn and https://github.com/catboost/tutorials"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9J5g6PDs1FIk"
      },
      "source": [
        "<table align=\"left\">\n",
        "  <td>\n",
        "    <a href=\"https://colab.research.google.com/github/phonchi/nsysu-math604/blob/master/static_files/presentations/14_ensemble.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n",
        "  </td>\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://kaggle.com/kernels/welcome?src=https://github.com/phonchi/nsysu-math604/blob/master/static_files/presentations/14_ensemble.ipynb\"><img src=\"https://kaggle.com/static/images/open-in-kaggle.svg\" /></a>\n",
        "  </td>\n",
        "</table>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install scikeras[tensorflow] -qq\n",
        "!pip install xgboost -U -qq\n",
        "!pip install lightgbm -U -qq\n",
        "!pip install catboost -U\n",
        "!pip install shap"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dpt39LbKyuet",
        "outputId": "d814ae5f-37c3-46a7-b68d-d9148142fd87"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: catboost in /usr/local/lib/python3.7/dist-packages (1.0.6)\n",
            "Requirement already satisfied: numpy>=1.16.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.21.6)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from catboost) (1.15.0)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from catboost) (1.4.1)\n",
            "Requirement already satisfied: pandas>=0.24.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.3.5)\n",
            "Requirement already satisfied: graphviz in /usr/local/lib/python3.7/dist-packages (from catboost) (0.10.1)\n",
            "Requirement already satisfied: plotly in /usr/local/lib/python3.7/dist-packages (from catboost) (5.5.0)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from catboost) (3.2.2)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2022.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (1.4.2)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (0.11.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (3.0.9)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from kiwisolver>=1.0.1->matplotlib->catboost) (4.2.0)\n",
            "Requirement already satisfied: tenacity>=6.2.0 in /usr/local/lib/python3.7/dist-packages (from plotly->catboost) (8.0.1)\n",
            "Collecting shap\n",
            "  Downloading shap-0.40.0-cp37-cp37m-manylinux2010_x86_64.whl (564 kB)\n",
            "\u001b[K     |████████████████████████████████| 564 kB 5.0 MB/s \n",
            "\u001b[?25hRequirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from shap) (1.4.1)\n",
            "Requirement already satisfied: tqdm>4.25.0 in /usr/local/lib/python3.7/dist-packages (from shap) (4.64.0)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.7/dist-packages (from shap) (1.3.5)\n",
            "Requirement already satisfied: numba in /usr/local/lib/python3.7/dist-packages (from shap) (0.51.2)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from shap) (1.21.6)\n",
            "Requirement already satisfied: cloudpickle in /usr/local/lib/python3.7/dist-packages (from shap) (1.3.0)\n",
            "Collecting slicer==0.0.7\n",
            "  Downloading slicer-0.0.7-py3-none-any.whl (14 kB)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from shap) (1.0.2)\n",
            "Requirement already satisfied: packaging>20.9 in /usr/local/lib/python3.7/dist-packages (from shap) (21.3)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>20.9->shap) (3.0.9)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from numba->shap) (57.4.0)\n",
            "Requirement already satisfied: llvmlite<0.35,>=0.34.0.dev0 in /usr/local/lib/python3.7/dist-packages (from numba->shap) (0.34.0)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas->shap) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas->shap) (2022.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas->shap) (1.15.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->shap) (3.1.0)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->shap) (1.1.0)\n",
            "Installing collected packages: slicer, shap\n",
            "Successfully installed shap-0.40.0 slicer-0.0.7\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import time\n",
        "\n",
        "import tensorflow as tf\n",
        "#tf.compat.v1.disable_v2_behavior() #Should be enabled when using DeepSHAP, see https://github.com/slundberg/shap/issues/2189\n",
        "from tensorflow import keras\n",
        "from scikeras.wrappers import KerasClassifier, KerasRegressor\n",
        "\n",
        "from sklearn.datasets import make_moons\n",
        "from sklearn.ensemble import RandomForestClassifier, VotingClassifier\n",
        "from sklearn.ensemble import StackingClassifier\n",
        "from sklearn.ensemble import BaggingClassifier, GradientBoostingRegressor, GradientBoostingClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.model_selection import GridSearchCV, RandomizedSearchCV, StratifiedKFold\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.ensemble import AdaBoostClassifier\n",
        "from sklearn import datasets\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "from xgboost import XGBClassifier\n",
        "from xgboost import XGBRegressor\n",
        "\n",
        "import lightgbm as lgb\n",
        "\n",
        "from catboost import CatBoostClassifier, CatBoostRegressor, Pool\n",
        "\n",
        "import shap\n",
        "shap.initjs() \n",
        "\n",
        "import matplotlib as mpl\n",
        "from matplotlib import pyplot as plt\n",
        "from matplotlib import cm\n",
        "%matplotlib inline"
      ],
      "metadata": {
        "id": "5bV_HvPiH-9i",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 43
        },
        "outputId": "71539579-11a7-4ce4-d6d2-ca247872f37c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<div align='center'><img src='data:image/png;base64,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' /></div><script charset='utf-8'>!function(t){function e(r){if(n[r])return n[r].exports;var i=n[r]={i:r,l:!1,exports:{}};return t[r].call(i.exports,i,i.exports,e),i.l=!0,i.exports}var n={};return e.m=t,e.c=n,e.i=function(t){return t},e.d=function(t,n,r){e.o(t,n)||Object.defineProperty(t,n,{configurable:!1,enumerable:!0,get:r})},e.n=function(t){var n=t&&t.__esModule?function(){return t.default}:function(){return t};return e.d(n,\"a\",n),n},e.o=function(t,e){return Object.prototype.hasOwnProperty.call(t,e)},e.p=\"\",e(e.s=410)}([function(t,e,n){\"use strict\";function r(t,e,n,r,o,a,u,c){if(i(e),!t){var s;if(void 0===e)s=new Error(\"Minified exception occurred; use the non-minified dev environment for the full error message and additional helpful warnings.\");else{var l=[n,r,o,a,u,c],f=0;s=new Error(e.replace(/%s/g,function(){return l[f++]})),s.name=\"Invariant Violation\"}throw s.framesToPop=1,s}}var i=function(t){};t.exports=r},function(t,e,n){\"use strict\";var r=n(8),i=r;t.exports=i},function(t,e,n){\"use strict\";function r(t){for(var e=arguments.length-1,n=\"Minified React error #\"+t+\"; visit http://facebook.github.io/react/docs/error-decoder.html?invariant=\"+t,r=0;r<e;r++)n+=\"&args[]=\"+encodeURIComponent(arguments[r+1]);n+=\" for the full message or use the non-minified dev environment for full errors and additional helpful warnings.\";var i=new Error(n);throw i.name=\"Invariant Violation\",i.framesToPop=1,i}t.exports=r},function(t,e,n){\"use strict\";function r(t){if(null===t||void 0===t)throw new TypeError(\"Object.assign cannot be called with null or undefined\");return Object(t)}function i(){try{if(!Object.assign)return!1;var t=new String(\"abc\");if(t[5]=\"de\",\"5\"===Object.getOwnPropertyNames(t)[0])return!1;for(var e={},n=0;n<10;n++)e[\"_\"+String.fromCharCode(n)]=n;var r=Object.getOwnPropertyNames(e).map(function(t){return e[t]});if(\"0123456789\"!==r.join(\"\"))return!1;var i={};return\"abcdefghijklmnopqrst\".split(\"\").forEach(function(t){i[t]=t}),\"abcdefghijklmnopqrst\"===Object.keys(Object.assign({},i)).join(\"\")}catch(t){return!1}}/*\n",
              "object-assign\n",
              "(c) Sindre Sorhus\n",
              "@license MIT\n",
              "*/\n",
              "var o=Object.getOwnPropertySymbols,a=Object.prototype.hasOwnProperty,u=Object.prototype.propertyIsEnumerable;t.exports=i()?Object.assign:function(t,e){for(var n,i,c=r(t),s=1;s<arguments.length;s++){n=Object(arguments[s]);for(var l in n)a.call(n,l)&&(c[l]=n[l]);if(o){i=o(n);for(var f=0;f<i.length;f++)u.call(n,i[f])&&(c[i[f]]=n[i[f]])}}return c}},function(t,e,n){\"use strict\";function r(t,e){return 1===t.nodeType&&t.getAttribute(d)===String(e)||8===t.nodeType&&t.nodeValue===\" react-text: \"+e+\" \"||8===t.nodeType&&t.nodeValue===\" react-empty: \"+e+\" \"}function i(t){for(var e;e=t._renderedComponent;)t=e;return t}function o(t,e){var n=i(t);n._hostNode=e,e[g]=n}function a(t){var e=t._hostNode;e&&(delete e[g],t._hostNode=null)}function u(t,e){if(!(t._flags&v.hasCachedChildNodes)){var n=t._renderedChildren,a=e.firstChild;t:for(var u in n)if(n.hasOwnProperty(u)){var c=n[u],s=i(c)._domID;if(0!==s){for(;null!==a;a=a.nextSibling)if(r(a,s)){o(c,a);continue t}f(\"32\",s)}}t._flags|=v.hasCachedChildNodes}}function c(t){if(t[g])return t[g];for(var e=[];!t[g];){if(e.push(t),!t.parentNode)return null;t=t.parentNode}for(var n,r;t&&(r=t[g]);t=e.pop())n=r,e.length&&u(r,t);return n}function s(t){var e=c(t);return null!=e&&e._hostNode===t?e:null}function l(t){if(void 0===t._hostNode?f(\"33\"):void 0,t._hostNode)return t._hostNode;for(var e=[];!t._hostNode;)e.push(t),t._hostParent?void 0:f(\"34\"),t=t._hostParent;for(;e.length;t=e.pop())u(t,t._hostNode);return t._hostNode}var f=n(2),p=n(21),h=n(157),d=(n(0),p.ID_ATTRIBUTE_NAME),v=h,g=\"__reactInternalInstance$\"+Math.random().toString(36).slice(2),m={getClosestInstanceFromNode:c,getInstanceFromNode:s,getNodeFromInstance:l,precacheChildNodes:u,precacheNode:o,uncacheNode:a};t.exports=m},function(t,e,n){\"use strict\";function r(t,e,n,a){function u(e){return t(e=new Date(+e)),e}return u.floor=u,u.ceil=function(n){return t(n=new Date(n-1)),e(n,1),t(n),n},u.round=function(t){var e=u(t),n=u.ceil(t);return t-e<n-t?e:n},u.offset=function(t,n){return e(t=new Date(+t),null==n?1:Math.floor(n)),t},u.range=function(n,r,i){var o=[];if(n=u.ceil(n),i=null==i?1:Math.floor(i),!(n<r&&i>0))return o;do o.push(new Date(+n));while(e(n,i),t(n),n<r);return o},u.filter=function(n){return r(function(e){if(e>=e)for(;t(e),!n(e);)e.setTime(e-1)},function(t,r){if(t>=t)for(;--r>=0;)for(;e(t,1),!n(t););})},n&&(u.count=function(e,r){return i.setTime(+e),o.setTime(+r),t(i),t(o),Math.floor(n(i,o))},u.every=function(t){return t=Math.floor(t),isFinite(t)&&t>0?t>1?u.filter(a?function(e){return a(e)%t===0}:function(e){return u.count(0,e)%t===0}):u:null}),u}e.a=r;var i=new Date,o=new Date},function(t,e,n){\"use strict\";var r=!(\"undefined\"==typeof window||!window.document||!window.document.createElement),i={canUseDOM:r,canUseWorkers:\"undefined\"!=typeof Worker,canUseEventListeners:r&&!(!window.addEventListener&&!window.attachEvent),canUseViewport:r&&!!window.screen,isInWorker:!r};t.exports=i},function(t,e,n){\"use strict\";function r(t,e){this._groups=t,this._parents=e}function i(){return new r([[document.documentElement]],D)}var o=n(272),a=n(273),u=n(261),c=n(255),s=n(131),l=n(260),f=n(265),p=n(268),h=n(275),d=n(253),v=n(267),g=n(266),m=n(274),y=n(259),_=n(258),b=n(252),x=n(276),w=n(269),C=n(254),M=n(277),k=n(262),E=n(270),T=n(264),S=n(251),P=n(263),N=n(271),A=n(256),O=n(70),I=n(257);n.d(e,\"c\",function(){return D}),e.b=r;var D=[null];r.prototype=i.prototype={constructor:r,select:o.a,selectAll:a.a,filter:u.a,data:c.a,enter:s.a,exit:l.a,merge:f.a,order:p.a,sort:h.a,call:d.a,nodes:v.a,node:g.a,size:m.a,empty:y.a,each:_.a,attr:b.a,style:x.a,property:w.a,classed:C.a,text:M.a,html:k.a,raise:E.a,lower:T.a,append:S.a,insert:P.a,remove:N.a,datum:A.a,on:O.c,dispatch:I.a},e.a=i},function(t,e,n){\"use strict\";function r(t){return function(){return t}}var i=function(){};i.thatReturns=r,i.thatReturnsFalse=r(!1),i.thatReturnsTrue=r(!0),i.thatReturnsNull=r(null),i.thatReturnsThis=function(){return this},i.thatReturnsArgument=function(t){return t},t.exports=i},function(t,e,n){\"use strict\";var r=null;t.exports={debugTool:r}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(59);n.d(e,\"color\",function(){return r.a}),n.d(e,\"rgb\",function(){return r.b}),n.d(e,\"hsl\",function(){return r.c});var i=n(210);n.d(e,\"lab\",function(){return i.a}),n.d(e,\"hcl\",function(){return i.b});var o=n(209);n.d(e,\"cubehelix\",function(){return o.a})},function(t,e,n){\"use strict\";function r(){T.ReactReconcileTransaction&&x?void 0:l(\"123\")}function i(){this.reinitializeTransaction(),this.dirtyComponentsLength=null,this.callbackQueue=p.getPooled(),this.reconcileTransaction=T.ReactReconcileTransaction.getPooled(!0)}function o(t,e,n,i,o,a){return r(),x.batchedUpdates(t,e,n,i,o,a)}function a(t,e){return t._mountOrder-e._mountOrder}function u(t){var e=t.dirtyComponentsLength;e!==m.length?l(\"124\",e,m.length):void 0,m.sort(a),y++;for(var n=0;n<e;n++){var r=m[n],i=r._pendingCallbacks;r._pendingCallbacks=null;var o;if(d.logTopLevelRenders){var u=r;r._currentElement.type.isReactTopLevelWrapper&&(u=r._renderedComponent),o=\"React update: \"+u.getName(),console.time(o)}if(v.performUpdateIfNecessary(r,t.reconcileTransaction,y),o&&console.timeEnd(o),i)for(var c=0;c<i.length;c++)t.callbackQueue.enqueue(i[c],r.getPublicInstance())}}function c(t){return r(),x.isBatchingUpdates?(m.push(t),void(null==t._updateBatchNumber&&(t._updateBatchNumber=y+1))):void x.batchedUpdates(c,t)}function s(t,e){x.isBatchingUpdates?void 0:l(\"125\"),_.enqueue(t,e),b=!0}var l=n(2),f=n(3),p=n(155),h=n(17),d=n(160),v=n(24),g=n(53),m=(n(0),[]),y=0,_=p.getPooled(),b=!1,x=null,w={initialize:function(){this.dirtyComponentsLength=m.length},close:function(){this.dirtyComponentsLength!==m.length?(m.splice(0,this.dirtyComponentsLength),k()):m.length=0}},C={initialize:function(){this.callbackQueue.reset()},close:function(){this.callbackQueue.notifyAll()}},M=[w,C];f(i.prototype,g,{getTransactionWrappers:function(){return M},destructor:function(){this.dirtyComponentsLength=null,p.release(this.callbackQueue),this.callbackQueue=null,T.ReactReconcileTransaction.release(this.reconcileTransaction),this.reconcileTransaction=null},perform:function(t,e,n){return g.perform.call(this,this.reconcileTransaction.perform,this.reconcileTransaction,t,e,n)}}),h.addPoolingTo(i);var k=function(){for(;m.length||b;){if(m.length){var t=i.getPooled();t.perform(u,null,t),i.release(t)}if(b){b=!1;var e=_;_=p.getPooled(),e.notifyAll(),p.release(e)}}},E={injectReconcileTransaction:function(t){t?void 0:l(\"126\"),T.ReactReconcileTransaction=t},injectBatchingStrategy:function(t){t?void 0:l(\"127\"),\"function\"!=typeof t.batchedUpdates?l(\"128\"):void 0,\"boolean\"!=typeof t.isBatchingUpdates?l(\"129\"):void 0,x=t}},T={ReactReconcileTransaction:null,batchedUpdates:o,enqueueUpdate:c,flushBatchedUpdates:k,injection:E,asap:s};t.exports=T},function(t,e,n){\"use strict\";var r=n(102);n.d(e,\"c\",function(){return r.a});var i=n(18);n.d(e,\"f\",function(){return i.a});var o=n(103);n.d(e,\"d\",function(){return o.a});var a=(n(185),n(104),n(105),n(186),n(197),n(198),n(108),n(188),n(189),n(190),n(191),n(106),n(192),n(193),n(57));n.d(e,\"e\",function(){return a.a});var u=n(107);n.d(e,\"g\",function(){return u.a});var c=(n(194),n(195),n(196),n(109));n.d(e,\"a\",function(){return c.a}),n.d(e,\"b\",function(){return c.b});n(110),n(111),n(199)},function(t,e,n){\"use strict\";n.d(e,\"e\",function(){return r}),n.d(e,\"d\",function(){return i}),n.d(e,\"c\",function(){return o}),n.d(e,\"b\",function(){return a}),n.d(e,\"a\",function(){return u});var r=1e3,i=6e4,o=36e5,a=864e5,u=6048e5},function(t,e,n){\"use strict\";function r(t,e,n,r){this.dispatchConfig=t,this._targetInst=e,this.nativeEvent=n;var i=this.constructor.Interface;for(var o in i)if(i.hasOwnProperty(o)){var u=i[o];u?this[o]=u(n):\"target\"===o?this.target=r:this[o]=n[o]}var c=null!=n.defaultPrevented?n.defaultPrevented:n.returnValue===!1;return c?this.isDefaultPrevented=a.thatReturnsTrue:this.isDefaultPrevented=a.thatReturnsFalse,this.isPropagationStopped=a.thatReturnsFalse,this}var i=n(3),o=n(17),a=n(8),u=(n(1),\"function\"==typeof Proxy,[\"dispatchConfig\",\"_targetInst\",\"nativeEvent\",\"isDefaultPrevented\",\"isPropagationStopped\",\"_dispatchListeners\",\"_dispatchInstances\"]),c={type:null,target:null,currentTarget:a.thatReturnsNull,eventPhase:null,bubbles:null,cancelable:null,timeStamp:function(t){return t.timeStamp||Date.now()},defaultPrevented:null,isTrusted:null};i(r.prototype,{preventDefault:function(){this.defaultPrevented=!0;var t=this.nativeEvent;t&&(t.preventDefault?t.preventDefault():\"unknown\"!=typeof t.returnValue&&(t.returnValue=!1),this.isDefaultPrevented=a.thatReturnsTrue)},stopPropagation:function(){var t=this.nativeEvent;t&&(t.stopPropagation?t.stopPropagation():\"unknown\"!=typeof t.cancelBubble&&(t.cancelBubble=!0),this.isPropagationStopped=a.thatReturnsTrue)},persist:function(){this.isPersistent=a.thatReturnsTrue},isPersistent:a.thatReturnsFalse,destructor:function(){var t=this.constructor.Interface;for(var e in t)this[e]=null;for(var n=0;n<u.length;n++)this[u[n]]=null}}),r.Interface=c,r.augmentClass=function(t,e){var n=this,r=function(){};r.prototype=n.prototype;var a=new r;i(a,t.prototype),t.prototype=a,t.prototype.constructor=t,t.Interface=i({},n.Interface,e),t.augmentClass=n.augmentClass,o.addPoolingTo(t,o.fourArgumentPooler)},o.addPoolingTo(r,o.fourArgumentPooler),t.exports=r},function(t,e,n){\"use strict\";var r={current:null};t.exports=r},function(t,e,n){\"use strict\";n.d(e,\"a\",function(){return i}),n.d(e,\"b\",function(){return o});var r=Array.prototype,i=r.map,o=r.slice},function(t,e,n){\"use strict\";var r=n(2),i=(n(0),function(t){var e=this;if(e.instancePool.length){var n=e.instancePool.pop();return e.call(n,t),n}return new e(t)}),o=function(t,e){var n=this;if(n.instancePool.length){var r=n.instancePool.pop();return n.call(r,t,e),r}return new n(t,e)},a=function(t,e,n){var r=this;if(r.instancePool.length){var i=r.instancePool.pop();return r.call(i,t,e,n),i}return new r(t,e,n)},u=function(t,e,n,r){var i=this;if(i.instancePool.length){var o=i.instancePool.pop();return i.call(o,t,e,n,r),o}return new i(t,e,n,r)},c=function(t){var e=this;t instanceof e?void 0:r(\"25\"),t.destructor(),e.instancePool.length<e.poolSize&&e.instancePool.push(t)},s=10,l=i,f=function(t,e){var n=t;return n.instancePool=[],n.getPooled=e||l,n.poolSize||(n.poolSize=s),n.release=c,n},p={addPoolingTo:f,oneArgumentPooler:i,twoArgumentPooler:o,threeArgumentPooler:a,fourArgumentPooler:u};t.exports=p},function(t,e,n){\"use strict\";e.a=function(t,e){return t<e?-1:t>e?1:t>=e?0:NaN}},function(t,e,n){\"use strict\";e.a=function(t){return function(){return t}}},function(t,e,n){\"use strict\";function r(t){if(g){var e=t.node,n=t.children;if(n.length)for(var r=0;r<n.length;r++)m(e,n[r],null);else null!=t.html?f(e,t.html):null!=t.text&&h(e,t.text)}}function i(t,e){t.parentNode.replaceChild(e.node,t),r(e)}function o(t,e){g?t.children.push(e):t.node.appendChild(e.node)}function a(t,e){g?t.html=e:f(t.node,e)}function u(t,e){g?t.text=e:h(t.node,e)}function c(){return this.node.nodeName}function s(t){return{node:t,children:[],html:null,text:null,toString:c}}var l=n(82),f=n(55),p=n(90),h=n(171),d=1,v=11,g=\"undefined\"!=typeof document&&\"number\"==typeof document.documentMode||\"undefined\"!=typeof navigator&&\"string\"==typeof navigator.userAgent&&/\\bEdge\\/\\d/.test(navigator.userAgent),m=p(function(t,e,n){e.node.nodeType===v||e.node.nodeType===d&&\"object\"===e.node.nodeName.toLowerCase()&&(null==e.node.namespaceURI||e.node.namespaceURI===l.html)?(r(e),t.insertBefore(e.node,n)):(t.insertBefore(e.node,n),r(e))});s.insertTreeBefore=m,s.replaceChildWithTree=i,s.queueChild=o,s.queueHTML=a,s.queueText=u,t.exports=s},function(t,e,n){\"use strict\";function r(t,e){return(t&e)===e}var i=n(2),o=(n(0),{MUST_USE_PROPERTY:1,HAS_BOOLEAN_VALUE:4,HAS_NUMERIC_VALUE:8,HAS_POSITIVE_NUMERIC_VALUE:24,HAS_OVERLOADED_BOOLEAN_VALUE:32,injectDOMPropertyConfig:function(t){var e=o,n=t.Properties||{},a=t.DOMAttributeNamespaces||{},c=t.DOMAttributeNames||{},s=t.DOMPropertyNames||{},l=t.DOMMutationMethods||{};t.isCustomAttribute&&u._isCustomAttributeFunctions.push(t.isCustomAttribute);for(var f in n){u.properties.hasOwnProperty(f)?i(\"48\",f):void 0;var p=f.toLowerCase(),h=n[f],d={attributeName:p,attributeNamespace:null,propertyName:f,mutationMethod:null,mustUseProperty:r(h,e.MUST_USE_PROPERTY),hasBooleanValue:r(h,e.HAS_BOOLEAN_VALUE),hasNumericValue:r(h,e.HAS_NUMERIC_VALUE),hasPositiveNumericValue:r(h,e.HAS_POSITIVE_NUMERIC_VALUE),hasOverloadedBooleanValue:r(h,e.HAS_OVERLOADED_BOOLEAN_VALUE)};if(d.hasBooleanValue+d.hasNumericValue+d.hasOverloadedBooleanValue<=1?void 0:i(\"50\",f),c.hasOwnProperty(f)){var v=c[f];d.attributeName=v}a.hasOwnProperty(f)&&(d.attributeNamespace=a[f]),s.hasOwnProperty(f)&&(d.propertyName=s[f]),l.hasOwnProperty(f)&&(d.mutationMethod=l[f]),u.properties[f]=d}}}),a=\":A-Z_a-z\\\\u00C0-\\\\u00D6\\\\u00D8-\\\\u00F6\\\\u00F8-\\\\u02FF\\\\u0370-\\\\u037D\\\\u037F-\\\\u1FFF\\\\u200C-\\\\u200D\\\\u2070-\\\\u218F\\\\u2C00-\\\\u2FEF\\\\u3001-\\\\uD7FF\\\\uF900-\\\\uFDCF\\\\uFDF0-\\\\uFFFD\",u={ID_ATTRIBUTE_NAME:\"data-reactid\",ROOT_ATTRIBUTE_NAME:\"data-reactroot\",ATTRIBUTE_NAME_START_CHAR:a,ATTRIBUTE_NAME_CHAR:a+\"\\\\-.0-9\\\\u00B7\\\\u0300-\\\\u036F\\\\u203F-\\\\u2040\",properties:{},getPossibleStandardName:null,_isCustomAttributeFunctions:[],isCustomAttribute:function(t){for(var e=0;e<u._isCustomAttributeFunctions.length;e++){var n=u._isCustomAttributeFunctions[e];if(n(t))return!0}return!1},injection:o};t.exports=u},function(t,e,n){\"use strict\";function r(t){return\"button\"===t||\"input\"===t||\"select\"===t||\"textarea\"===t}function i(t,e,n){switch(t){case\"onClick\":case\"onClickCapture\":case\"onDoubleClick\":case\"onDoubleClickCapture\":case\"onMouseDown\":case\"onMouseDownCapture\":case\"onMouseMove\":case\"onMouseMoveCapture\":case\"onMouseUp\":case\"onMouseUpCapture\":return!(!n.disabled||!r(e));default:return!1}}var o=n(2),a=n(83),u=n(50),c=n(87),s=n(165),l=n(166),f=(n(0),{}),p=null,h=function(t,e){t&&(u.executeDispatchesInOrder(t,e),t.isPersistent()||t.constructor.release(t))},d=function(t){return h(t,!0)},v=function(t){return h(t,!1)},g=function(t){return\".\"+t._rootNodeID},m={injection:{injectEventPluginOrder:a.injectEventPluginOrder,injectEventPluginsByName:a.injectEventPluginsByName},putListener:function(t,e,n){\"function\"!=typeof n?o(\"94\",e,typeof n):void 0;var r=g(t),i=f[e]||(f[e]={});i[r]=n;var u=a.registrationNameModules[e];u&&u.didPutListener&&u.didPutListener(t,e,n)},getListener:function(t,e){var n=f[e];if(i(e,t._currentElement.type,t._currentElement.props))return null;var r=g(t);return n&&n[r]},deleteListener:function(t,e){var n=a.registrationNameModules[e];n&&n.willDeleteListener&&n.willDeleteListener(t,e);var r=f[e];if(r){var i=g(t);delete r[i]}},deleteAllListeners:function(t){var e=g(t);for(var n in f)if(f.hasOwnProperty(n)&&f[n][e]){var r=a.registrationNameModules[n];r&&r.willDeleteListener&&r.willDeleteListener(t,n),delete f[n][e]}},extractEvents:function(t,e,n,r){for(var i,o=a.plugins,u=0;u<o.length;u++){var c=o[u];if(c){var l=c.extractEvents(t,e,n,r);l&&(i=s(i,l))}}return i},enqueueEvents:function(t){t&&(p=s(p,t))},processEventQueue:function(t){var e=p;p=null,t?l(e,d):l(e,v),p?o(\"95\"):void 0,c.rethrowCaughtError()},__purge:function(){f={}},__getListenerBank:function(){return f}};t.exports=m},function(t,e,n){\"use strict\";function r(t,e,n){var r=e.dispatchConfig.phasedRegistrationNames[n];return m(t,r)}function i(t,e,n){var i=r(t,n,e);i&&(n._dispatchListeners=v(n._dispatchListeners,i),n._dispatchInstances=v(n._dispatchInstances,t))}function o(t){t&&t.dispatchConfig.phasedRegistrationNames&&d.traverseTwoPhase(t._targetInst,i,t)}function a(t){if(t&&t.dispatchConfig.phasedRegistrationNames){var e=t._targetInst,n=e?d.getParentInstance(e):null;d.traverseTwoPhase(n,i,t)}}function u(t,e,n){if(n&&n.dispatchConfig.registrationName){var r=n.dispatchConfig.registrationName,i=m(t,r);i&&(n._dispatchListeners=v(n._dispatchListeners,i),n._dispatchInstances=v(n._dispatchInstances,t))}}function c(t){t&&t.dispatchConfig.registrationName&&u(t._targetInst,null,t)}function s(t){g(t,o)}function l(t){g(t,a)}function f(t,e,n,r){d.traverseEnterLeave(n,r,u,t,e)}function p(t){g(t,c)}var h=n(22),d=n(50),v=n(165),g=n(166),m=(n(1),h.getListener),y={accumulateTwoPhaseDispatches:s,accumulateTwoPhaseDispatchesSkipTarget:l,accumulateDirectDispatches:p,accumulateEnterLeaveDispatches:f};t.exports=y},function(t,e,n){\"use strict\";function r(){i.attachRefs(this,this._currentElement)}var i=n(368),o=(n(9),n(1),{mountComponent:function(t,e,n,i,o,a){var u=t.mountComponent(e,n,i,o,a);return t._currentElement&&null!=t._currentElement.ref&&e.getReactMountReady().enqueue(r,t),u},getHostNode:function(t){return t.getHostNode()},unmountComponent:function(t,e){i.detachRefs(t,t._currentElement),t.unmountComponent(e)},receiveComponent:function(t,e,n,o){var a=t._currentElement;if(e!==a||o!==t._context){var u=i.shouldUpdateRefs(a,e);u&&i.detachRefs(t,a),t.receiveComponent(e,n,o),u&&t._currentElement&&null!=t._currentElement.ref&&n.getReactMountReady().enqueue(r,t)}},performUpdateIfNecessary:function(t,e,n){t._updateBatchNumber===n&&t.performUpdateIfNecessary(e)}});t.exports=o},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o=n(93),a={view:function(t){if(t.view)return t.view;var e=o(t);if(e.window===e)return e;var n=e.ownerDocument;return n?n.defaultView||n.parentWindow:window},detail:function(t){return t.detail||0}};i.augmentClass(r,a),t.exports=r},function(t,e,n){\"use strict\";var r=n(3),i=n(401),o=n(97),a=n(406),u=n(402),c=n(403),s=n(27),l=n(404),f=n(407),p=n(408),h=(n(1),s.createElement),d=s.createFactory,v=s.cloneElement,g=r,m={Children:{map:i.map,forEach:i.forEach,count:i.count,toArray:i.toArray,only:p},Component:o,PureComponent:a,createElement:h,cloneElement:v,isValidElement:s.isValidElement,PropTypes:l,createClass:u.createClass,createFactory:d,createMixin:function(t){return t},DOM:c,version:f,__spread:g};t.exports=m},function(t,e,n){\"use strict\";function r(t){return void 0!==t.ref}function i(t){return void 0!==t.key}var o=n(3),a=n(15),u=(n(1),n(176),Object.prototype.hasOwnProperty),c=n(174),s={key:!0,ref:!0,__self:!0,__source:!0},l=function(t,e,n,r,i,o,a){var u={$$typeof:c,type:t,key:e,ref:n,props:a,_owner:o};return u};l.createElement=function(t,e,n){var o,c={},f=null,p=null,h=null,d=null;if(null!=e){r(e)&&(p=e.ref),i(e)&&(f=\"\"+e.key),h=void 0===e.__self?null:e.__self,d=void 0===e.__source?null:e.__source;for(o in e)u.call(e,o)&&!s.hasOwnProperty(o)&&(c[o]=e[o])}var v=arguments.length-2;if(1===v)c.children=n;else if(v>1){for(var g=Array(v),m=0;m<v;m++)g[m]=arguments[m+2];c.children=g}if(t&&t.defaultProps){var y=t.defaultProps;for(o in y)void 0===c[o]&&(c[o]=y[o])}return l(t,f,p,h,d,a.current,c)},l.createFactory=function(t){var e=l.createElement.bind(null,t);return e.type=t,e},l.cloneAndReplaceKey=function(t,e){var n=l(t.type,e,t.ref,t._self,t._source,t._owner,t.props);return n},l.cloneElement=function(t,e,n){var c,f=o({},t.props),p=t.key,h=t.ref,d=t._self,v=t._source,g=t._owner;if(null!=e){r(e)&&(h=e.ref,g=a.current),i(e)&&(p=\"\"+e.key);var m;t.type&&t.type.defaultProps&&(m=t.type.defaultProps);for(c in e)u.call(e,c)&&!s.hasOwnProperty(c)&&(void 0===e[c]&&void 0!==m?f[c]=m[c]:f[c]=e[c])}var y=arguments.length-2;if(1===y)f.children=n;else if(y>1){for(var _=Array(y),b=0;b<y;b++)_[b]=arguments[b+2];f.children=_}return l(t.type,p,h,d,v,g,f)},l.isValidElement=function(t){return\"object\"==typeof t&&null!==t&&t.$$typeof===c},t.exports=l},function(t,e,n){\"use strict\";function r(t){for(var e=arguments.length-1,n=\"Minified React error #\"+t+\"; visit http://facebook.github.io/react/docs/error-decoder.html?invariant=\"+t,r=0;r<e;r++)n+=\"&args[]=\"+encodeURIComponent(arguments[r+1]);n+=\" for the full message or use the non-minified dev environment for full errors and additional helpful warnings.\";var i=new Error(n);throw i.name=\"Invariant Violation\",i.framesToPop=1,i}t.exports=r},function(t,e,n){\"use strict\";e.a=function(t){return null===t?NaN:+t}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(211);n.d(e,\"formatDefaultLocale\",function(){return r.a}),n.d(e,\"format\",function(){return r.b}),n.d(e,\"formatPrefix\",function(){return r.c});var i=n(117);n.d(e,\"formatLocale\",function(){return i.a});var o=n(115);n.d(e,\"formatSpecifier\",function(){return o.a});var a=n(215);n.d(e,\"precisionFixed\",function(){return a.a});var u=n(216);n.d(e,\"precisionPrefix\",function(){return u.a});var c=n(217);n.d(e,\"precisionRound\",function(){return c.a})},function(t,e,n){\"use strict\";var r=n(63);n.d(e,\"b\",function(){return r.a});var i=(n(118),n(62),n(119),n(121),n(43));n.d(e,\"a\",function(){return i.a});var o=(n(122),n(223));n.d(e,\"c\",function(){return o.a});var a=(n(124),n(225),n(227),n(123),n(220),n(221),n(219),n(218));n.d(e,\"d\",function(){return a.a});n(222)},function(t,e,n){\"use strict\";function r(t,e){return function(n){return t+n*e}}function i(t,e,n){return t=Math.pow(t,n),e=Math.pow(e,n)-t,n=1/n,function(r){return Math.pow(t+r*e,n)}}function o(t,e){var i=e-t;return i?r(t,i>180||i<-180?i-360*Math.round(i/360):i):n.i(c.a)(isNaN(t)?e:t)}function a(t){return 1===(t=+t)?u:function(e,r){return r-e?i(e,r,t):n.i(c.a)(isNaN(e)?r:e)}}function u(t,e){var i=e-t;return i?r(t,i):n.i(c.a)(isNaN(t)?e:t)}var c=n(120);e.b=o,e.c=a,e.a=u},function(t,e,n){\"use strict\";e.a=function(t){return t.match(/.{6}/g).map(function(t){return\"#\"+t})}},function(t,e,n){\"use strict\";function r(t){var e=t.domain;return t.ticks=function(t){var r=e();return n.i(o.a)(r[0],r[r.length-1],null==t?10:t)},t.tickFormat=function(t,r){return n.i(c.a)(e(),t,r)},t.nice=function(r){var i=e(),a=i.length-1,u=null==r?10:r,c=i[0],s=i[a],l=n.i(o.b)(c,s,u);return l&&(l=n.i(o.b)(Math.floor(c/l)*l,Math.ceil(s/l)*l,u),i[0]=Math.floor(c/l)*l,i[a]=Math.ceil(s/l)*l,e(i)),t},t}function i(){var t=n.i(u.a)(u.b,a.a);return t.copy=function(){return n.i(u.c)(t,i())},r(t)}var o=n(12),a=n(31),u=n(45),c=n(243);e.b=r,e.a=i},function(t,e,n){\"use strict\";n.d(e,\"a\",function(){return r}),n.d(e,\"b\",function(){return i}),n.d(e,\"d\",function(){return o}),n.d(e,\"c\",function(){return a});var r=1e-12,i=Math.PI,o=i/2,a=2*i},function(t,e,n){\"use strict\";e.a=function(t,e){if((r=t.length)>1)for(var n,r,i=1,o=t[e[0]],a=o.length;i<r;++i){n=o,o=t[e[i]];for(var u=0;u<a;++u)o[u][1]+=o[u][0]=isNaN(n[u][1])?n[u][0]:n[u][1]}}},function(t,e,n){\"use strict\";e.a=function(t){for(var e=t.length,n=new Array(e);--e>=0;)n[e]=e;return n}},function(t,e,n){\"use strict\";var r={};t.exports=r},function(t,e,n){(function(t,r){var i;(function(){function o(t,e){return t.set(e[0],e[1]),t}function a(t,e){return t.add(e),t}function u(t,e,n){switch(n.length){case 0:return t.call(e);case 1:return t.call(e,n[0]);case 2:return t.call(e,n[0],n[1]);case 3:return t.call(e,n[0],n[1],n[2])}return t.apply(e,n)}function c(t,e,n,r){for(var i=-1,o=null==t?0:t.length;++i<o;){var a=t[i];e(r,a,n(a),t)}return r}function s(t,e){for(var n=-1,r=null==t?0:t.length;++n<r&&e(t[n],n,t)!==!1;);return t}function l(t,e){for(var n=null==t?0:t.length;n--&&e(t[n],n,t)!==!1;);return t}function f(t,e){for(var n=-1,r=null==t?0:t.length;++n<r;)if(!e(t[n],n,t))return!1;return!0}function p(t,e){for(var n=-1,r=null==t?0:t.length,i=0,o=[];++n<r;){var a=t[n];e(a,n,t)&&(o[i++]=a)}return o}function h(t,e){var n=null==t?0:t.length;return!!n&&M(t,e,0)>-1}function d(t,e,n){for(var r=-1,i=null==t?0:t.length;++r<i;)if(n(e,t[r]))return!0;return!1}function v(t,e){for(var n=-1,r=null==t?0:t.length,i=Array(r);++n<r;)i[n]=e(t[n],n,t);return i}function g(t,e){for(var n=-1,r=e.length,i=t.length;++n<r;)t[i+n]=e[n];return t}function m(t,e,n,r){var i=-1,o=null==t?0:t.length;for(r&&o&&(n=t[++i]);++i<o;)n=e(n,t[i],i,t);return n}function y(t,e,n,r){var i=null==t?0:t.length;for(r&&i&&(n=t[--i]);i--;)n=e(n,t[i],i,t);return n}function _(t,e){for(var n=-1,r=null==t?0:t.length;++n<r;)if(e(t[n],n,t))return!0;return!1}function b(t){return t.split(\"\")}function x(t){return t.match(ze)||[]}function w(t,e,n){var r;return n(t,function(t,n,i){if(e(t,n,i))return r=n,!1}),r}function C(t,e,n,r){for(var i=t.length,o=n+(r?1:-1);r?o--:++o<i;)if(e(t[o],o,t))return o;return-1}function M(t,e,n){return e===e?Z(t,e,n):C(t,E,n)}function k(t,e,n,r){for(var i=n-1,o=t.length;++i<o;)if(r(t[i],e))return i;return-1}function E(t){return t!==t}function T(t,e){var n=null==t?0:t.length;return n?O(t,e)/n:Ut}function S(t){return function(e){return null==e?it:e[t]}}function P(t){return function(e){return null==t?it:t[e]}}function N(t,e,n,r,i){return i(t,function(t,i,o){n=r?(r=!1,t):e(n,t,i,o)}),n}function A(t,e){var n=t.length;for(t.sort(e);n--;)t[n]=t[n].value;return t}function O(t,e){for(var n,r=-1,i=t.length;++r<i;){var o=e(t[r]);o!==it&&(n=n===it?o:n+o)}return n}function I(t,e){for(var n=-1,r=Array(t);++n<t;)r[n]=e(n);return r}function D(t,e){return v(e,function(e){return[e,t[e]]})}function R(t){return function(e){return t(e)}}function L(t,e){return v(e,function(e){return t[e]})}function U(t,e){return t.has(e)}function F(t,e){for(var n=-1,r=t.length;++n<r&&M(e,t[n],0)>-1;);return n}function j(t,e){for(var n=t.length;n--&&M(e,t[n],0)>-1;);return n}function B(t,e){for(var n=t.length,r=0;n--;)t[n]===e&&++r;return r}function W(t){return\"\\\\\"+nr[t]}function V(t,e){return null==t?it:t[e]}function z(t){return Kn.test(t)}function H(t){return Gn.test(t)}function q(t){for(var e,n=[];!(e=t.next()).done;)n.push(e.value);return n}function Y(t){var e=-1,n=Array(t.size);return t.forEach(function(t,r){n[++e]=[r,t]}),n}function K(t,e){return function(n){return t(e(n))}}function G(t,e){for(var n=-1,r=t.length,i=0,o=[];++n<r;){var a=t[n];a!==e&&a!==ft||(t[n]=ft,o[i++]=n)}return o}function $(t){var e=-1,n=Array(t.size);return t.forEach(function(t){n[++e]=t}),n}function X(t){var e=-1,n=Array(t.size);return t.forEach(function(t){n[++e]=[t,t]}),n}function Z(t,e,n){for(var r=n-1,i=t.length;++r<i;)if(t[r]===e)return r;return-1}function Q(t,e,n){for(var r=n+1;r--;)if(t[r]===e)return r;return r}function J(t){return z(t)?et(t):_r(t)}function tt(t){return z(t)?nt(t):b(t)}function et(t){for(var e=qn.lastIndex=0;qn.test(t);)++e;return e}function nt(t){return t.match(qn)||[]}function rt(t){return t.match(Yn)||[]}var it,ot=\"4.17.4\",at=200,ut=\"Unsupported core-js use. Try https://npms.io/search?q=ponyfill.\",ct=\"Expected a function\",st=\"__lodash_hash_undefined__\",lt=500,ft=\"__lodash_placeholder__\",pt=1,ht=2,dt=4,vt=1,gt=2,mt=1,yt=2,_t=4,bt=8,xt=16,wt=32,Ct=64,Mt=128,kt=256,Et=512,Tt=30,St=\"...\",Pt=800,Nt=16,At=1,Ot=2,It=3,Dt=1/0,Rt=9007199254740991,Lt=1.7976931348623157e308,Ut=NaN,Ft=4294967295,jt=Ft-1,Bt=Ft>>>1,Wt=[[\"ary\",Mt],[\"bind\",mt],[\"bindKey\",yt],[\"curry\",bt],[\"curryRight\",xt],[\"flip\",Et],[\"partial\",wt],[\"partialRight\",Ct],[\"rearg\",kt]],Vt=\"[object Arguments]\",zt=\"[object Array]\",Ht=\"[object AsyncFunction]\",qt=\"[object Boolean]\",Yt=\"[object Date]\",Kt=\"[object DOMException]\",Gt=\"[object Error]\",$t=\"[object Function]\",Xt=\"[object GeneratorFunction]\",Zt=\"[object Map]\",Qt=\"[object Number]\",Jt=\"[object Null]\",te=\"[object Object]\",ee=\"[object Promise]\",ne=\"[object Proxy]\",re=\"[object RegExp]\",ie=\"[object Set]\",oe=\"[object String]\",ae=\"[object Symbol]\",ue=\"[object Undefined]\",ce=\"[object WeakMap]\",se=\"[object WeakSet]\",le=\"[object ArrayBuffer]\",fe=\"[object DataView]\",pe=\"[object Float32Array]\",he=\"[object Float64Array]\",de=\"[object Int8Array]\",ve=\"[object Int16Array]\",ge=\"[object Int32Array]\",me=\"[object Uint8Array]\",ye=\"[object Uint8ClampedArray]\",_e=\"[object Uint16Array]\",be=\"[object Uint32Array]\",xe=/\\b__p \\+= '';/g,we=/\\b(__p \\+=) '' \\+/g,Ce=/(__e\\(.*?\\)|\\b__t\\)) \\+\\n'';/g,Me=/&(?:amp|lt|gt|quot|#39);/g,ke=/[&<>\"']/g,Ee=RegExp(Me.source),Te=RegExp(ke.source),Se=/<%-([\\s\\S]+?)%>/g,Pe=/<%([\\s\\S]+?)%>/g,Ne=/<%=([\\s\\S]+?)%>/g,Ae=/\\.|\\[(?:[^[\\]]*|([\"'])(?:(?!\\1)[^\\\\]|\\\\.)*?\\1)\\]/,Oe=/^\\w*$/,Ie=/^\\./,De=/[^.[\\]]+|\\[(?:(-?\\d+(?:\\.\\d+)?)|([\"'])((?:(?!\\2)[^\\\\]|\\\\.)*?)\\2)\\]|(?=(?:\\.|\\[\\])(?:\\.|\\[\\]|$))/g,Re=/[\\\\^$.*+?()[\\]{}|]/g,Le=RegExp(Re.source),Ue=/^\\s+|\\s+$/g,Fe=/^\\s+/,je=/\\s+$/,Be=/\\{(?:\\n\\/\\* \\[wrapped with .+\\] \\*\\/)?\\n?/,We=/\\{\\n\\/\\* \\[wrapped with (.+)\\] \\*/,Ve=/,? & /,ze=/[^\\x00-\\x2f\\x3a-\\x40\\x5b-\\x60\\x7b-\\x7f]+/g,He=/\\\\(\\\\)?/g,qe=/\\$\\{([^\\\\}]*(?:\\\\.[^\\\\}]*)*)\\}/g,Ye=/\\w*$/,Ke=/^[-+]0x[0-9a-f]+$/i,Ge=/^0b[01]+$/i,$e=/^\\[object .+?Constructor\\]$/,Xe=/^0o[0-7]+$/i,Ze=/^(?:0|[1-9]\\d*)$/,Qe=/[\\xc0-\\xd6\\xd8-\\xf6\\xf8-\\xff\\u0100-\\u017f]/g,Je=/($^)/,tn=/['\\n\\r\\u2028\\u2029\\\\]/g,en=\"\\\\ud800-\\\\udfff\",nn=\"\\\\u0300-\\\\u036f\",rn=\"\\\\ufe20-\\\\ufe2f\",on=\"\\\\u20d0-\\\\u20ff\",an=nn+rn+on,un=\"\\\\u2700-\\\\u27bf\",cn=\"a-z\\\\xdf-\\\\xf6\\\\xf8-\\\\xff\",sn=\"\\\\xac\\\\xb1\\\\xd7\\\\xf7\",ln=\"\\\\x00-\\\\x2f\\\\x3a-\\\\x40\\\\x5b-\\\\x60\\\\x7b-\\\\xbf\",fn=\"\\\\u2000-\\\\u206f\",pn=\" \\\\t\\\\x0b\\\\f\\\\xa0\\\\ufeff\\\\n\\\\r\\\\u2028\\\\u2029\\\\u1680\\\\u180e\\\\u2000\\\\u2001\\\\u2002\\\\u2003\\\\u2004\\\\u2005\\\\u2006\\\\u2007\\\\u2008\\\\u2009\\\\u200a\\\\u202f\\\\u205f\\\\u3000\",hn=\"A-Z\\\\xc0-\\\\xd6\\\\xd8-\\\\xde\",dn=\"\\\\ufe0e\\\\ufe0f\",vn=sn+ln+fn+pn,gn=\"['’]\",mn=\"[\"+en+\"]\",yn=\"[\"+vn+\"]\",_n=\"[\"+an+\"]\",bn=\"\\\\d+\",xn=\"[\"+un+\"]\",wn=\"[\"+cn+\"]\",Cn=\"[^\"+en+vn+bn+un+cn+hn+\"]\",Mn=\"\\\\ud83c[\\\\udffb-\\\\udfff]\",kn=\"(?:\"+_n+\"|\"+Mn+\")\",En=\"[^\"+en+\"]\",Tn=\"(?:\\\\ud83c[\\\\udde6-\\\\uddff]){2}\",Sn=\"[\\\\ud800-\\\\udbff][\\\\udc00-\\\\udfff]\",Pn=\"[\"+hn+\"]\",Nn=\"\\\\u200d\",An=\"(?:\"+wn+\"|\"+Cn+\")\",On=\"(?:\"+Pn+\"|\"+Cn+\")\",In=\"(?:\"+gn+\"(?:d|ll|m|re|s|t|ve))?\",Dn=\"(?:\"+gn+\"(?:D|LL|M|RE|S|T|VE))?\",Rn=kn+\"?\",Ln=\"[\"+dn+\"]?\",Un=\"(?:\"+Nn+\"(?:\"+[En,Tn,Sn].join(\"|\")+\")\"+Ln+Rn+\")*\",Fn=\"\\\\d*(?:(?:1st|2nd|3rd|(?![123])\\\\dth)\\\\b)\",jn=\"\\\\d*(?:(?:1ST|2ND|3RD|(?![123])\\\\dTH)\\\\b)\",Bn=Ln+Rn+Un,Wn=\"(?:\"+[xn,Tn,Sn].join(\"|\")+\")\"+Bn,Vn=\"(?:\"+[En+_n+\"?\",_n,Tn,Sn,mn].join(\"|\")+\")\",zn=RegExp(gn,\"g\"),Hn=RegExp(_n,\"g\"),qn=RegExp(Mn+\"(?=\"+Mn+\")|\"+Vn+Bn,\"g\"),Yn=RegExp([Pn+\"?\"+wn+\"+\"+In+\"(?=\"+[yn,Pn,\"$\"].join(\"|\")+\")\",On+\"+\"+Dn+\"(?=\"+[yn,Pn+An,\"$\"].join(\"|\")+\")\",Pn+\"?\"+An+\"+\"+In,Pn+\"+\"+Dn,jn,Fn,bn,Wn].join(\"|\"),\"g\"),Kn=RegExp(\"[\"+Nn+en+an+dn+\"]\"),Gn=/[a-z][A-Z]|[A-Z]{2,}[a-z]|[0-9][a-zA-Z]|[a-zA-Z][0-9]|[^a-zA-Z0-9 ]/,$n=[\"Array\",\"Buffer\",\"DataView\",\"Date\",\"Error\",\"Float32Array\",\"Float64Array\",\"Function\",\"Int8Array\",\"Int16Array\",\"Int32Array\",\"Map\",\"Math\",\"Object\",\"Promise\",\"RegExp\",\"Set\",\"String\",\"Symbol\",\"TypeError\",\"Uint8Array\",\"Uint8ClampedArray\",\"Uint16Array\",\"Uint32Array\",\"WeakMap\",\"_\",\"clearTimeout\",\"isFinite\",\"parseInt\",\"setTimeout\"],Xn=-1,Zn={};Zn[pe]=Zn[he]=Zn[de]=Zn[ve]=Zn[ge]=Zn[me]=Zn[ye]=Zn[_e]=Zn[be]=!0,Zn[Vt]=Zn[zt]=Zn[le]=Zn[qt]=Zn[fe]=Zn[Yt]=Zn[Gt]=Zn[$t]=Zn[Zt]=Zn[Qt]=Zn[te]=Zn[re]=Zn[ie]=Zn[oe]=Zn[ce]=!1;var Qn={};Qn[Vt]=Qn[zt]=Qn[le]=Qn[fe]=Qn[qt]=Qn[Yt]=Qn[pe]=Qn[he]=Qn[de]=Qn[ve]=Qn[ge]=Qn[Zt]=Qn[Qt]=Qn[te]=Qn[re]=Qn[ie]=Qn[oe]=Qn[ae]=Qn[me]=Qn[ye]=Qn[_e]=Qn[be]=!0,Qn[Gt]=Qn[$t]=Qn[ce]=!1;var Jn={\"À\":\"A\",\"Á\":\"A\",\"Â\":\"A\",\"Ã\":\"A\",\"Ä\":\"A\",\"Å\":\"A\",\"à\":\"a\",\"á\":\"a\",\"â\":\"a\",\"ã\":\"a\",\"ä\":\"a\",\"å\":\"a\",\"Ç\":\"C\",\"ç\":\"c\",\"Ð\":\"D\",\"ð\":\"d\",\"È\":\"E\",\"É\":\"E\",\"Ê\":\"E\",\"Ë\":\"E\",\"è\":\"e\",\"é\":\"e\",\"ê\":\"e\",\"ë\":\"e\",\"Ì\":\"I\",\"Í\":\"I\",\"Î\":\"I\",\"Ï\":\"I\",\"ì\":\"i\",\"í\":\"i\",\"î\":\"i\",\"ï\":\"i\",\"Ñ\":\"N\",\"ñ\":\"n\",\"Ò\":\"O\",\"Ó\":\"O\",\"Ô\":\"O\",\"Õ\":\"O\",\"Ö\":\"O\",\"Ø\":\"O\",\"ò\":\"o\",\"ó\":\"o\",\"ô\":\"o\",\"õ\":\"o\",\"ö\":\"o\",\"ø\":\"o\",\"Ù\":\"U\",\"Ú\":\"U\",\"Û\":\"U\",\"Ü\":\"U\",\"ù\":\"u\",\"ú\":\"u\",\"û\":\"u\",\"ü\":\"u\",\"Ý\":\"Y\",\"ý\":\"y\",\"ÿ\":\"y\",\"Æ\":\"Ae\",\"æ\":\"ae\",\"Þ\":\"Th\",\"þ\":\"th\",\"ß\":\"ss\",\"Ā\":\"A\",\"Ă\":\"A\",\"Ą\":\"A\",\"ā\":\"a\",\"ă\":\"a\",\"ą\":\"a\",\"Ć\":\"C\",\"Ĉ\":\"C\",\"Ċ\":\"C\",\"Č\":\"C\",\"ć\":\"c\",\"ĉ\":\"c\",\"ċ\":\"c\",\"č\":\"c\",\"Ď\":\"D\",\"Đ\":\"D\",\"ď\":\"d\",\"đ\":\"d\",\"Ē\":\"E\",\"Ĕ\":\"E\",\"Ė\":\"E\",\"Ę\":\"E\",\"Ě\":\"E\",\"ē\":\"e\",\"ĕ\":\"e\",\"ė\":\"e\",\"ę\":\"e\",\"ě\":\"e\",\"Ĝ\":\"G\",\"Ğ\":\"G\",\"Ġ\":\"G\",\"Ģ\":\"G\",\"ĝ\":\"g\",\"ğ\":\"g\",\"ġ\":\"g\",\"ģ\":\"g\",\"Ĥ\":\"H\",\"Ħ\":\"H\",\"ĥ\":\"h\",\"ħ\":\"h\",\"Ĩ\":\"I\",\"Ī\":\"I\",\"Ĭ\":\"I\",\"Į\":\"I\",\"İ\":\"I\",\"ĩ\":\"i\",\"ī\":\"i\",\"ĭ\":\"i\",\"į\":\"i\",\"ı\":\"i\",\"Ĵ\":\"J\",\"ĵ\":\"j\",\"Ķ\":\"K\",\"ķ\":\"k\",\"ĸ\":\"k\",\"Ĺ\":\"L\",\"Ļ\":\"L\",\"Ľ\":\"L\",\"Ŀ\":\"L\",\"Ł\":\"L\",\"ĺ\":\"l\",\"ļ\":\"l\",\"ľ\":\"l\",\"ŀ\":\"l\",\"ł\":\"l\",\"Ń\":\"N\",\"Ņ\":\"N\",\"Ň\":\"N\",\"Ŋ\":\"N\",\"ń\":\"n\",\"ņ\":\"n\",\"ň\":\"n\",\"ŋ\":\"n\",\"Ō\":\"O\",\"Ŏ\":\"O\",\"Ő\":\"O\",\"ō\":\"o\",\"ŏ\":\"o\",\"ő\":\"o\",\"Ŕ\":\"R\",\"Ŗ\":\"R\",\"Ř\":\"R\",\"ŕ\":\"r\",\"ŗ\":\"r\",\"ř\":\"r\",\"Ś\":\"S\",\"Ŝ\":\"S\",\"Ş\":\"S\",\"Š\":\"S\",\"ś\":\"s\",\"ŝ\":\"s\",\"ş\":\"s\",\"š\":\"s\",\"Ţ\":\"T\",\"Ť\":\"T\",\"Ŧ\":\"T\",\"ţ\":\"t\",\"ť\":\"t\",\"ŧ\":\"t\",\"Ũ\":\"U\",\"Ū\":\"U\",\"Ŭ\":\"U\",\"Ů\":\"U\",\"Ű\":\"U\",\"Ų\":\"U\",\"ũ\":\"u\",\"ū\":\"u\",\"ŭ\":\"u\",\"ů\":\"u\",\"ű\":\"u\",\"ų\":\"u\",\"Ŵ\":\"W\",\"ŵ\":\"w\",\"Ŷ\":\"Y\",\"ŷ\":\"y\",\"Ÿ\":\"Y\",\"Ź\":\"Z\",\"Ż\":\"Z\",\"Ž\":\"Z\",\"ź\":\"z\",\"ż\":\"z\",\"ž\":\"z\",\"IJ\":\"IJ\",\n",
              "\"ij\":\"ij\",\"Œ\":\"Oe\",\"œ\":\"oe\",\"ʼn\":\"'n\",\"ſ\":\"s\"},tr={\"&\":\"&amp;\",\"<\":\"&lt;\",\">\":\"&gt;\",'\"':\"&quot;\",\"'\":\"&#39;\"},er={\"&amp;\":\"&\",\"&lt;\":\"<\",\"&gt;\":\">\",\"&quot;\":'\"',\"&#39;\":\"'\"},nr={\"\\\\\":\"\\\\\",\"'\":\"'\",\"\\n\":\"n\",\"\\r\":\"r\",\"\\u2028\":\"u2028\",\"\\u2029\":\"u2029\"},rr=parseFloat,ir=parseInt,or=\"object\"==typeof t&&t&&t.Object===Object&&t,ar=\"object\"==typeof self&&self&&self.Object===Object&&self,ur=or||ar||Function(\"return this\")(),cr=\"object\"==typeof e&&e&&!e.nodeType&&e,sr=cr&&\"object\"==typeof r&&r&&!r.nodeType&&r,lr=sr&&sr.exports===cr,fr=lr&&or.process,pr=function(){try{return fr&&fr.binding&&fr.binding(\"util\")}catch(t){}}(),hr=pr&&pr.isArrayBuffer,dr=pr&&pr.isDate,vr=pr&&pr.isMap,gr=pr&&pr.isRegExp,mr=pr&&pr.isSet,yr=pr&&pr.isTypedArray,_r=S(\"length\"),br=P(Jn),xr=P(tr),wr=P(er),Cr=function t(e){function n(t){if(sc(t)&&!xp(t)&&!(t instanceof b)){if(t instanceof i)return t;if(bl.call(t,\"__wrapped__\"))return aa(t)}return new i(t)}function r(){}function i(t,e){this.__wrapped__=t,this.__actions__=[],this.__chain__=!!e,this.__index__=0,this.__values__=it}function b(t){this.__wrapped__=t,this.__actions__=[],this.__dir__=1,this.__filtered__=!1,this.__iteratees__=[],this.__takeCount__=Ft,this.__views__=[]}function P(){var t=new b(this.__wrapped__);return t.__actions__=Bi(this.__actions__),t.__dir__=this.__dir__,t.__filtered__=this.__filtered__,t.__iteratees__=Bi(this.__iteratees__),t.__takeCount__=this.__takeCount__,t.__views__=Bi(this.__views__),t}function Z(){if(this.__filtered__){var t=new b(this);t.__dir__=-1,t.__filtered__=!0}else t=this.clone(),t.__dir__*=-1;return t}function et(){var t=this.__wrapped__.value(),e=this.__dir__,n=xp(t),r=e<0,i=n?t.length:0,o=No(0,i,this.__views__),a=o.start,u=o.end,c=u-a,s=r?u:a-1,l=this.__iteratees__,f=l.length,p=0,h=Xl(c,this.__takeCount__);if(!n||!r&&i==c&&h==c)return xi(t,this.__actions__);var d=[];t:for(;c--&&p<h;){s+=e;for(var v=-1,g=t[s];++v<f;){var m=l[v],y=m.iteratee,_=m.type,b=y(g);if(_==Ot)g=b;else if(!b){if(_==At)continue t;break t}}d[p++]=g}return d}function nt(t){var e=-1,n=null==t?0:t.length;for(this.clear();++e<n;){var r=t[e];this.set(r[0],r[1])}}function ze(){this.__data__=uf?uf(null):{},this.size=0}function en(t){var e=this.has(t)&&delete this.__data__[t];return this.size-=e?1:0,e}function nn(t){var e=this.__data__;if(uf){var n=e[t];return n===st?it:n}return bl.call(e,t)?e[t]:it}function rn(t){var e=this.__data__;return uf?e[t]!==it:bl.call(e,t)}function on(t,e){var n=this.__data__;return this.size+=this.has(t)?0:1,n[t]=uf&&e===it?st:e,this}function an(t){var e=-1,n=null==t?0:t.length;for(this.clear();++e<n;){var r=t[e];this.set(r[0],r[1])}}function un(){this.__data__=[],this.size=0}function cn(t){var e=this.__data__,n=In(e,t);if(n<0)return!1;var r=e.length-1;return n==r?e.pop():Dl.call(e,n,1),--this.size,!0}function sn(t){var e=this.__data__,n=In(e,t);return n<0?it:e[n][1]}function ln(t){return In(this.__data__,t)>-1}function fn(t,e){var n=this.__data__,r=In(n,t);return r<0?(++this.size,n.push([t,e])):n[r][1]=e,this}function pn(t){var e=-1,n=null==t?0:t.length;for(this.clear();++e<n;){var r=t[e];this.set(r[0],r[1])}}function hn(){this.size=0,this.__data__={hash:new nt,map:new(nf||an),string:new nt}}function dn(t){var e=Eo(this,t).delete(t);return this.size-=e?1:0,e}function vn(t){return Eo(this,t).get(t)}function gn(t){return Eo(this,t).has(t)}function mn(t,e){var n=Eo(this,t),r=n.size;return n.set(t,e),this.size+=n.size==r?0:1,this}function yn(t){var e=-1,n=null==t?0:t.length;for(this.__data__=new pn;++e<n;)this.add(t[e])}function _n(t){return this.__data__.set(t,st),this}function bn(t){return this.__data__.has(t)}function xn(t){var e=this.__data__=new an(t);this.size=e.size}function wn(){this.__data__=new an,this.size=0}function Cn(t){var e=this.__data__,n=e.delete(t);return this.size=e.size,n}function Mn(t){return this.__data__.get(t)}function kn(t){return this.__data__.has(t)}function En(t,e){var n=this.__data__;if(n instanceof an){var r=n.__data__;if(!nf||r.length<at-1)return r.push([t,e]),this.size=++n.size,this;n=this.__data__=new pn(r)}return n.set(t,e),this.size=n.size,this}function Tn(t,e){var n=xp(t),r=!n&&bp(t),i=!n&&!r&&Cp(t),o=!n&&!r&&!i&&Sp(t),a=n||r||i||o,u=a?I(t.length,hl):[],c=u.length;for(var s in t)!e&&!bl.call(t,s)||a&&(\"length\"==s||i&&(\"offset\"==s||\"parent\"==s)||o&&(\"buffer\"==s||\"byteLength\"==s||\"byteOffset\"==s)||Fo(s,c))||u.push(s);return u}function Sn(t){var e=t.length;return e?t[ni(0,e-1)]:it}function Pn(t,e){return na(Bi(t),jn(e,0,t.length))}function Nn(t){return na(Bi(t))}function An(t,e,n){(n===it||$u(t[e],n))&&(n!==it||e in t)||Un(t,e,n)}function On(t,e,n){var r=t[e];bl.call(t,e)&&$u(r,n)&&(n!==it||e in t)||Un(t,e,n)}function In(t,e){for(var n=t.length;n--;)if($u(t[n][0],e))return n;return-1}function Dn(t,e,n,r){return _f(t,function(t,i,o){e(r,t,n(t),o)}),r}function Rn(t,e){return t&&Wi(e,Hc(e),t)}function Ln(t,e){return t&&Wi(e,qc(e),t)}function Un(t,e,n){\"__proto__\"==e&&Fl?Fl(t,e,{configurable:!0,enumerable:!0,value:n,writable:!0}):t[e]=n}function Fn(t,e){for(var n=-1,r=e.length,i=al(r),o=null==t;++n<r;)i[n]=o?it:Wc(t,e[n]);return i}function jn(t,e,n){return t===t&&(n!==it&&(t=t<=n?t:n),e!==it&&(t=t>=e?t:e)),t}function Bn(t,e,n,r,i,o){var a,u=e&pt,c=e&ht,l=e&dt;if(n&&(a=i?n(t,r,i,o):n(t)),a!==it)return a;if(!cc(t))return t;var f=xp(t);if(f){if(a=Io(t),!u)return Bi(t,a)}else{var p=Af(t),h=p==$t||p==Xt;if(Cp(t))return Si(t,u);if(p==te||p==Vt||h&&!i){if(a=c||h?{}:Do(t),!u)return c?zi(t,Ln(a,t)):Vi(t,Rn(a,t))}else{if(!Qn[p])return i?t:{};a=Ro(t,p,Bn,u)}}o||(o=new xn);var d=o.get(t);if(d)return d;o.set(t,a);var v=l?c?wo:xo:c?qc:Hc,g=f?it:v(t);return s(g||t,function(r,i){g&&(i=r,r=t[i]),On(a,i,Bn(r,e,n,i,t,o))}),a}function Wn(t){var e=Hc(t);return function(n){return Vn(n,t,e)}}function Vn(t,e,n){var r=n.length;if(null==t)return!r;for(t=fl(t);r--;){var i=n[r],o=e[i],a=t[i];if(a===it&&!(i in t)||!o(a))return!1}return!0}function qn(t,e,n){if(\"function\"!=typeof t)throw new dl(ct);return Df(function(){t.apply(it,n)},e)}function Yn(t,e,n,r){var i=-1,o=h,a=!0,u=t.length,c=[],s=e.length;if(!u)return c;n&&(e=v(e,R(n))),r?(o=d,a=!1):e.length>=at&&(o=U,a=!1,e=new yn(e));t:for(;++i<u;){var l=t[i],f=null==n?l:n(l);if(l=r||0!==l?l:0,a&&f===f){for(var p=s;p--;)if(e[p]===f)continue t;c.push(l)}else o(e,f,r)||c.push(l)}return c}function Kn(t,e){var n=!0;return _f(t,function(t,r,i){return n=!!e(t,r,i)}),n}function Gn(t,e,n){for(var r=-1,i=t.length;++r<i;){var o=t[r],a=e(o);if(null!=a&&(u===it?a===a&&!bc(a):n(a,u)))var u=a,c=o}return c}function Jn(t,e,n,r){var i=t.length;for(n=Ec(n),n<0&&(n=-n>i?0:i+n),r=r===it||r>i?i:Ec(r),r<0&&(r+=i),r=n>r?0:Tc(r);n<r;)t[n++]=e;return t}function tr(t,e){var n=[];return _f(t,function(t,r,i){e(t,r,i)&&n.push(t)}),n}function er(t,e,n,r,i){var o=-1,a=t.length;for(n||(n=Uo),i||(i=[]);++o<a;){var u=t[o];e>0&&n(u)?e>1?er(u,e-1,n,r,i):g(i,u):r||(i[i.length]=u)}return i}function nr(t,e){return t&&xf(t,e,Hc)}function or(t,e){return t&&wf(t,e,Hc)}function ar(t,e){return p(e,function(e){return oc(t[e])})}function cr(t,e){e=Ei(e,t);for(var n=0,r=e.length;null!=t&&n<r;)t=t[ra(e[n++])];return n&&n==r?t:it}function sr(t,e,n){var r=e(t);return xp(t)?r:g(r,n(t))}function fr(t){return null==t?t===it?ue:Jt:Ul&&Ul in fl(t)?Po(t):Xo(t)}function pr(t,e){return t>e}function _r(t,e){return null!=t&&bl.call(t,e)}function Cr(t,e){return null!=t&&e in fl(t)}function kr(t,e,n){return t>=Xl(e,n)&&t<$l(e,n)}function Er(t,e,n){for(var r=n?d:h,i=t[0].length,o=t.length,a=o,u=al(o),c=1/0,s=[];a--;){var l=t[a];a&&e&&(l=v(l,R(e))),c=Xl(l.length,c),u[a]=!n&&(e||i>=120&&l.length>=120)?new yn(a&&l):it}l=t[0];var f=-1,p=u[0];t:for(;++f<i&&s.length<c;){var g=l[f],m=e?e(g):g;if(g=n||0!==g?g:0,!(p?U(p,m):r(s,m,n))){for(a=o;--a;){var y=u[a];if(!(y?U(y,m):r(t[a],m,n)))continue t}p&&p.push(m),s.push(g)}}return s}function Tr(t,e,n,r){return nr(t,function(t,i,o){e(r,n(t),i,o)}),r}function Sr(t,e,n){e=Ei(e,t),t=Qo(t,e);var r=null==t?t:t[ra(ka(e))];return null==r?it:u(r,t,n)}function Pr(t){return sc(t)&&fr(t)==Vt}function Nr(t){return sc(t)&&fr(t)==le}function Ar(t){return sc(t)&&fr(t)==Yt}function Or(t,e,n,r,i){return t===e||(null==t||null==e||!sc(t)&&!sc(e)?t!==t&&e!==e:Ir(t,e,n,r,Or,i))}function Ir(t,e,n,r,i,o){var a=xp(t),u=xp(e),c=a?zt:Af(t),s=u?zt:Af(e);c=c==Vt?te:c,s=s==Vt?te:s;var l=c==te,f=s==te,p=c==s;if(p&&Cp(t)){if(!Cp(e))return!1;a=!0,l=!1}if(p&&!l)return o||(o=new xn),a||Sp(t)?mo(t,e,n,r,i,o):yo(t,e,c,n,r,i,o);if(!(n&vt)){var h=l&&bl.call(t,\"__wrapped__\"),d=f&&bl.call(e,\"__wrapped__\");if(h||d){var v=h?t.value():t,g=d?e.value():e;return o||(o=new xn),i(v,g,n,r,o)}}return!!p&&(o||(o=new xn),_o(t,e,n,r,i,o))}function Dr(t){return sc(t)&&Af(t)==Zt}function Rr(t,e,n,r){var i=n.length,o=i,a=!r;if(null==t)return!o;for(t=fl(t);i--;){var u=n[i];if(a&&u[2]?u[1]!==t[u[0]]:!(u[0]in t))return!1}for(;++i<o;){u=n[i];var c=u[0],s=t[c],l=u[1];if(a&&u[2]){if(s===it&&!(c in t))return!1}else{var f=new xn;if(r)var p=r(s,l,c,t,e,f);if(!(p===it?Or(l,s,vt|gt,r,f):p))return!1}}return!0}function Lr(t){if(!cc(t)||zo(t))return!1;var e=oc(t)?El:$e;return e.test(ia(t))}function Ur(t){return sc(t)&&fr(t)==re}function Fr(t){return sc(t)&&Af(t)==ie}function jr(t){return sc(t)&&uc(t.length)&&!!Zn[fr(t)]}function Br(t){return\"function\"==typeof t?t:null==t?Ds:\"object\"==typeof t?xp(t)?Yr(t[0],t[1]):qr(t):Vs(t)}function Wr(t){if(!Ho(t))return Gl(t);var e=[];for(var n in fl(t))bl.call(t,n)&&\"constructor\"!=n&&e.push(n);return e}function Vr(t){if(!cc(t))return $o(t);var e=Ho(t),n=[];for(var r in t)(\"constructor\"!=r||!e&&bl.call(t,r))&&n.push(r);return n}function zr(t,e){return t<e}function Hr(t,e){var n=-1,r=Xu(t)?al(t.length):[];return _f(t,function(t,i,o){r[++n]=e(t,i,o)}),r}function qr(t){var e=To(t);return 1==e.length&&e[0][2]?Yo(e[0][0],e[0][1]):function(n){return n===t||Rr(n,t,e)}}function Yr(t,e){return Bo(t)&&qo(e)?Yo(ra(t),e):function(n){var r=Wc(n,t);return r===it&&r===e?zc(n,t):Or(e,r,vt|gt)}}function Kr(t,e,n,r,i){t!==e&&xf(e,function(o,a){if(cc(o))i||(i=new xn),Gr(t,e,a,n,Kr,r,i);else{var u=r?r(t[a],o,a+\"\",t,e,i):it;u===it&&(u=o),An(t,a,u)}},qc)}function Gr(t,e,n,r,i,o,a){var u=t[n],c=e[n],s=a.get(c);if(s)return void An(t,n,s);var l=o?o(u,c,n+\"\",t,e,a):it,f=l===it;if(f){var p=xp(c),h=!p&&Cp(c),d=!p&&!h&&Sp(c);l=c,p||h||d?xp(u)?l=u:Zu(u)?l=Bi(u):h?(f=!1,l=Si(c,!0)):d?(f=!1,l=Ri(c,!0)):l=[]:mc(c)||bp(c)?(l=u,bp(u)?l=Pc(u):(!cc(u)||r&&oc(u))&&(l=Do(c))):f=!1}f&&(a.set(c,l),i(l,c,r,o,a),a.delete(c)),An(t,n,l)}function $r(t,e){var n=t.length;if(n)return e+=e<0?n:0,Fo(e,n)?t[e]:it}function Xr(t,e,n){var r=-1;e=v(e.length?e:[Ds],R(ko()));var i=Hr(t,function(t,n,i){var o=v(e,function(e){return e(t)});return{criteria:o,index:++r,value:t}});return A(i,function(t,e){return Ui(t,e,n)})}function Zr(t,e){return Qr(t,e,function(e,n){return zc(t,n)})}function Qr(t,e,n){for(var r=-1,i=e.length,o={};++r<i;){var a=e[r],u=cr(t,a);n(u,a)&&ci(o,Ei(a,t),u)}return o}function Jr(t){return function(e){return cr(e,t)}}function ti(t,e,n,r){var i=r?k:M,o=-1,a=e.length,u=t;for(t===e&&(e=Bi(e)),n&&(u=v(t,R(n)));++o<a;)for(var c=0,s=e[o],l=n?n(s):s;(c=i(u,l,c,r))>-1;)u!==t&&Dl.call(u,c,1),Dl.call(t,c,1);return t}function ei(t,e){for(var n=t?e.length:0,r=n-1;n--;){var i=e[n];if(n==r||i!==o){var o=i;Fo(i)?Dl.call(t,i,1):yi(t,i)}}return t}function ni(t,e){return t+zl(Jl()*(e-t+1))}function ri(t,e,n,r){for(var i=-1,o=$l(Vl((e-t)/(n||1)),0),a=al(o);o--;)a[r?o:++i]=t,t+=n;return a}function ii(t,e){var n=\"\";if(!t||e<1||e>Rt)return n;do e%2&&(n+=t),e=zl(e/2),e&&(t+=t);while(e);return n}function oi(t,e){return Rf(Zo(t,e,Ds),t+\"\")}function ai(t){return Sn(rs(t))}function ui(t,e){var n=rs(t);return na(n,jn(e,0,n.length))}function ci(t,e,n,r){if(!cc(t))return t;e=Ei(e,t);for(var i=-1,o=e.length,a=o-1,u=t;null!=u&&++i<o;){var c=ra(e[i]),s=n;if(i!=a){var l=u[c];s=r?r(l,c,u):it,s===it&&(s=cc(l)?l:Fo(e[i+1])?[]:{})}On(u,c,s),u=u[c]}return t}function si(t){return na(rs(t))}function li(t,e,n){var r=-1,i=t.length;e<0&&(e=-e>i?0:i+e),n=n>i?i:n,n<0&&(n+=i),i=e>n?0:n-e>>>0,e>>>=0;for(var o=al(i);++r<i;)o[r]=t[r+e];return o}function fi(t,e){var n;return _f(t,function(t,r,i){return n=e(t,r,i),!n}),!!n}function pi(t,e,n){var r=0,i=null==t?r:t.length;if(\"number\"==typeof e&&e===e&&i<=Bt){for(;r<i;){var o=r+i>>>1,a=t[o];null!==a&&!bc(a)&&(n?a<=e:a<e)?r=o+1:i=o}return i}return hi(t,e,Ds,n)}function hi(t,e,n,r){e=n(e);for(var i=0,o=null==t?0:t.length,a=e!==e,u=null===e,c=bc(e),s=e===it;i<o;){var l=zl((i+o)/2),f=n(t[l]),p=f!==it,h=null===f,d=f===f,v=bc(f);if(a)var g=r||d;else g=s?d&&(r||p):u?d&&p&&(r||!h):c?d&&p&&!h&&(r||!v):!h&&!v&&(r?f<=e:f<e);g?i=l+1:o=l}return Xl(o,jt)}function di(t,e){for(var n=-1,r=t.length,i=0,o=[];++n<r;){var a=t[n],u=e?e(a):a;if(!n||!$u(u,c)){var c=u;o[i++]=0===a?0:a}}return o}function vi(t){return\"number\"==typeof t?t:bc(t)?Ut:+t}function gi(t){if(\"string\"==typeof t)return t;if(xp(t))return v(t,gi)+\"\";if(bc(t))return mf?mf.call(t):\"\";var e=t+\"\";return\"0\"==e&&1/t==-Dt?\"-0\":e}function mi(t,e,n){var r=-1,i=h,o=t.length,a=!0,u=[],c=u;if(n)a=!1,i=d;else if(o>=at){var s=e?null:Tf(t);if(s)return $(s);a=!1,i=U,c=new yn}else c=e?[]:u;t:for(;++r<o;){var l=t[r],f=e?e(l):l;if(l=n||0!==l?l:0,a&&f===f){for(var p=c.length;p--;)if(c[p]===f)continue t;e&&c.push(f),u.push(l)}else i(c,f,n)||(c!==u&&c.push(f),u.push(l))}return u}function yi(t,e){return e=Ei(e,t),t=Qo(t,e),null==t||delete t[ra(ka(e))]}function _i(t,e,n,r){return ci(t,e,n(cr(t,e)),r)}function bi(t,e,n,r){for(var i=t.length,o=r?i:-1;(r?o--:++o<i)&&e(t[o],o,t););return n?li(t,r?0:o,r?o+1:i):li(t,r?o+1:0,r?i:o)}function xi(t,e){var n=t;return n instanceof b&&(n=n.value()),m(e,function(t,e){return e.func.apply(e.thisArg,g([t],e.args))},n)}function wi(t,e,n){var r=t.length;if(r<2)return r?mi(t[0]):[];for(var i=-1,o=al(r);++i<r;)for(var a=t[i],u=-1;++u<r;)u!=i&&(o[i]=Yn(o[i]||a,t[u],e,n));return mi(er(o,1),e,n)}function Ci(t,e,n){for(var r=-1,i=t.length,o=e.length,a={};++r<i;){var u=r<o?e[r]:it;n(a,t[r],u)}return a}function Mi(t){return Zu(t)?t:[]}function ki(t){return\"function\"==typeof t?t:Ds}function Ei(t,e){return xp(t)?t:Bo(t,e)?[t]:Lf(Ac(t))}function Ti(t,e,n){var r=t.length;return n=n===it?r:n,!e&&n>=r?t:li(t,e,n)}function Si(t,e){if(e)return t.slice();var n=t.length,r=Nl?Nl(n):new t.constructor(n);return t.copy(r),r}function Pi(t){var e=new t.constructor(t.byteLength);return new Pl(e).set(new Pl(t)),e}function Ni(t,e){var n=e?Pi(t.buffer):t.buffer;return new t.constructor(n,t.byteOffset,t.byteLength)}function Ai(t,e,n){var r=e?n(Y(t),pt):Y(t);return m(r,o,new t.constructor)}function Oi(t){var e=new t.constructor(t.source,Ye.exec(t));return e.lastIndex=t.lastIndex,e}function Ii(t,e,n){var r=e?n($(t),pt):$(t);return m(r,a,new t.constructor)}function Di(t){return gf?fl(gf.call(t)):{}}function Ri(t,e){var n=e?Pi(t.buffer):t.buffer;return new t.constructor(n,t.byteOffset,t.length)}function Li(t,e){if(t!==e){var n=t!==it,r=null===t,i=t===t,o=bc(t),a=e!==it,u=null===e,c=e===e,s=bc(e);if(!u&&!s&&!o&&t>e||o&&a&&c&&!u&&!s||r&&a&&c||!n&&c||!i)return 1;if(!r&&!o&&!s&&t<e||s&&n&&i&&!r&&!o||u&&n&&i||!a&&i||!c)return-1}return 0}function Ui(t,e,n){for(var r=-1,i=t.criteria,o=e.criteria,a=i.length,u=n.length;++r<a;){var c=Li(i[r],o[r]);if(c){if(r>=u)return c;var s=n[r];return c*(\"desc\"==s?-1:1)}}return t.index-e.index}function Fi(t,e,n,r){for(var i=-1,o=t.length,a=n.length,u=-1,c=e.length,s=$l(o-a,0),l=al(c+s),f=!r;++u<c;)l[u]=e[u];for(;++i<a;)(f||i<o)&&(l[n[i]]=t[i]);for(;s--;)l[u++]=t[i++];return l}function ji(t,e,n,r){for(var i=-1,o=t.length,a=-1,u=n.length,c=-1,s=e.length,l=$l(o-u,0),f=al(l+s),p=!r;++i<l;)f[i]=t[i];for(var h=i;++c<s;)f[h+c]=e[c];for(;++a<u;)(p||i<o)&&(f[h+n[a]]=t[i++]);return f}function Bi(t,e){var n=-1,r=t.length;for(e||(e=al(r));++n<r;)e[n]=t[n];return e}function Wi(t,e,n,r){var i=!n;n||(n={});for(var o=-1,a=e.length;++o<a;){var u=e[o],c=r?r(n[u],t[u],u,n,t):it;c===it&&(c=t[u]),i?Un(n,u,c):On(n,u,c)}return n}function Vi(t,e){return Wi(t,Pf(t),e)}function zi(t,e){return Wi(t,Nf(t),e)}function Hi(t,e){return function(n,r){var i=xp(n)?c:Dn,o=e?e():{};return i(n,t,ko(r,2),o)}}function qi(t){return oi(function(e,n){var r=-1,i=n.length,o=i>1?n[i-1]:it,a=i>2?n[2]:it;for(o=t.length>3&&\"function\"==typeof o?(i--,o):it,a&&jo(n[0],n[1],a)&&(o=i<3?it:o,i=1),e=fl(e);++r<i;){var u=n[r];u&&t(e,u,r,o)}return e})}function Yi(t,e){return function(n,r){if(null==n)return n;if(!Xu(n))return t(n,r);for(var i=n.length,o=e?i:-1,a=fl(n);(e?o--:++o<i)&&r(a[o],o,a)!==!1;);return n}}function Ki(t){return function(e,n,r){for(var i=-1,o=fl(e),a=r(e),u=a.length;u--;){var c=a[t?u:++i];if(n(o[c],c,o)===!1)break}return e}}function Gi(t,e,n){function r(){var e=this&&this!==ur&&this instanceof r?o:t;return e.apply(i?n:this,arguments)}var i=e&mt,o=Zi(t);return r}function $i(t){return function(e){e=Ac(e);var n=z(e)?tt(e):it,r=n?n[0]:e.charAt(0),i=n?Ti(n,1).join(\"\"):e.slice(1);return r[t]()+i}}function Xi(t){return function(e){return m(Ps(ss(e).replace(zn,\"\")),t,\"\")}}function Zi(t){return function(){var e=arguments;switch(e.length){case 0:return new t;case 1:return new t(e[0]);case 2:return new t(e[0],e[1]);case 3:return new t(e[0],e[1],e[2]);case 4:return new t(e[0],e[1],e[2],e[3]);case 5:return new t(e[0],e[1],e[2],e[3],e[4]);case 6:return new t(e[0],e[1],e[2],e[3],e[4],e[5]);case 7:return new t(e[0],e[1],e[2],e[3],e[4],e[5],e[6])}var n=yf(t.prototype),r=t.apply(n,e);return cc(r)?r:n}}function Qi(t,e,n){function r(){for(var o=arguments.length,a=al(o),c=o,s=Mo(r);c--;)a[c]=arguments[c];var l=o<3&&a[0]!==s&&a[o-1]!==s?[]:G(a,s);if(o-=l.length,o<n)return so(t,e,eo,r.placeholder,it,a,l,it,it,n-o);var f=this&&this!==ur&&this instanceof r?i:t;return u(f,this,a)}var i=Zi(t);return r}function Ji(t){return function(e,n,r){var i=fl(e);if(!Xu(e)){var o=ko(n,3);e=Hc(e),n=function(t){return o(i[t],t,i)}}var a=t(e,n,r);return a>-1?i[o?e[a]:a]:it}}function to(t){return bo(function(e){var n=e.length,r=n,o=i.prototype.thru;for(t&&e.reverse();r--;){var a=e[r];if(\"function\"!=typeof a)throw new dl(ct);if(o&&!u&&\"wrapper\"==Co(a))var u=new i([],!0)}for(r=u?r:n;++r<n;){a=e[r];var c=Co(a),s=\"wrapper\"==c?Sf(a):it;u=s&&Vo(s[0])&&s[1]==(Mt|bt|wt|kt)&&!s[4].length&&1==s[9]?u[Co(s[0])].apply(u,s[3]):1==a.length&&Vo(a)?u[c]():u.thru(a)}return function(){var t=arguments,r=t[0];if(u&&1==t.length&&xp(r))return u.plant(r).value();for(var i=0,o=n?e[i].apply(this,t):r;++i<n;)o=e[i].call(this,o);return o}})}function eo(t,e,n,r,i,o,a,u,c,s){function l(){for(var m=arguments.length,y=al(m),_=m;_--;)y[_]=arguments[_];if(d)var b=Mo(l),x=B(y,b);if(r&&(y=Fi(y,r,i,d)),o&&(y=ji(y,o,a,d)),m-=x,d&&m<s){var w=G(y,b);return so(t,e,eo,l.placeholder,n,y,w,u,c,s-m)}var C=p?n:this,M=h?C[t]:t;return m=y.length,u?y=Jo(y,u):v&&m>1&&y.reverse(),f&&c<m&&(y.length=c),this&&this!==ur&&this instanceof l&&(M=g||Zi(M)),M.apply(C,y)}var f=e&Mt,p=e&mt,h=e&yt,d=e&(bt|xt),v=e&Et,g=h?it:Zi(t);return l}function no(t,e){return function(n,r){return Tr(n,t,e(r),{})}}function ro(t,e){return function(n,r){var i;if(n===it&&r===it)return e;if(n!==it&&(i=n),r!==it){if(i===it)return r;\"string\"==typeof n||\"string\"==typeof r?(n=gi(n),r=gi(r)):(n=vi(n),r=vi(r)),i=t(n,r)}return i}}function io(t){return bo(function(e){return e=v(e,R(ko())),oi(function(n){var r=this;return t(e,function(t){return u(t,r,n)})})})}function oo(t,e){e=e===it?\" \":gi(e);var n=e.length;if(n<2)return n?ii(e,t):e;var r=ii(e,Vl(t/J(e)));return z(e)?Ti(tt(r),0,t).join(\"\"):r.slice(0,t)}function ao(t,e,n,r){function i(){for(var e=-1,c=arguments.length,s=-1,l=r.length,f=al(l+c),p=this&&this!==ur&&this instanceof i?a:t;++s<l;)f[s]=r[s];for(;c--;)f[s++]=arguments[++e];return u(p,o?n:this,f)}var o=e&mt,a=Zi(t);return i}function uo(t){return function(e,n,r){return r&&\"number\"!=typeof r&&jo(e,n,r)&&(n=r=it),e=kc(e),n===it?(n=e,e=0):n=kc(n),r=r===it?e<n?1:-1:kc(r),ri(e,n,r,t)}}function co(t){return function(e,n){return\"string\"==typeof e&&\"string\"==typeof n||(e=Sc(e),n=Sc(n)),t(e,n)}}function so(t,e,n,r,i,o,a,u,c,s){var l=e&bt,f=l?a:it,p=l?it:a,h=l?o:it,d=l?it:o;e|=l?wt:Ct,e&=~(l?Ct:wt),e&_t||(e&=~(mt|yt));var v=[t,e,i,h,f,d,p,u,c,s],g=n.apply(it,v);return Vo(t)&&If(g,v),g.placeholder=r,ta(g,t,e)}function lo(t){var e=ll[t];return function(t,n){if(t=Sc(t),n=null==n?0:Xl(Ec(n),292)){var r=(Ac(t)+\"e\").split(\"e\"),i=e(r[0]+\"e\"+(+r[1]+n));return r=(Ac(i)+\"e\").split(\"e\"),+(r[0]+\"e\"+(+r[1]-n))}return e(t)}}function fo(t){return function(e){var n=Af(e);return n==Zt?Y(e):n==ie?X(e):D(e,t(e))}}function po(t,e,n,r,i,o,a,u){var c=e&yt;if(!c&&\"function\"!=typeof t)throw new dl(ct);var s=r?r.length:0;if(s||(e&=~(wt|Ct),r=i=it),a=a===it?a:$l(Ec(a),0),u=u===it?u:Ec(u),s-=i?i.length:0,e&Ct){var l=r,f=i;r=i=it}var p=c?it:Sf(t),h=[t,e,n,r,i,l,f,o,a,u];if(p&&Go(h,p),t=h[0],e=h[1],n=h[2],r=h[3],i=h[4],u=h[9]=h[9]===it?c?0:t.length:$l(h[9]-s,0),!u&&e&(bt|xt)&&(e&=~(bt|xt)),e&&e!=mt)d=e==bt||e==xt?Qi(t,e,u):e!=wt&&e!=(mt|wt)||i.length?eo.apply(it,h):ao(t,e,n,r);else var d=Gi(t,e,n);var v=p?Cf:If;return ta(v(d,h),t,e)}function ho(t,e,n,r){return t===it||$u(t,ml[n])&&!bl.call(r,n)?e:t}function vo(t,e,n,r,i,o){return cc(t)&&cc(e)&&(o.set(e,t),Kr(t,e,it,vo,o),o.delete(e)),t}function go(t){return mc(t)?it:t}function mo(t,e,n,r,i,o){var a=n&vt,u=t.length,c=e.length;if(u!=c&&!(a&&c>u))return!1;var s=o.get(t);if(s&&o.get(e))return s==e;var l=-1,f=!0,p=n&gt?new yn:it;for(o.set(t,e),o.set(e,t);++l<u;){var h=t[l],d=e[l];if(r)var v=a?r(d,h,l,e,t,o):r(h,d,l,t,e,o);if(v!==it){if(v)continue;f=!1;break}if(p){if(!_(e,function(t,e){if(!U(p,e)&&(h===t||i(h,t,n,r,o)))return p.push(e)})){f=!1;break}}else if(h!==d&&!i(h,d,n,r,o)){f=!1;break}}return o.delete(t),o.delete(e),f}function yo(t,e,n,r,i,o,a){switch(n){case fe:if(t.byteLength!=e.byteLength||t.byteOffset!=e.byteOffset)return!1;t=t.buffer,e=e.buffer;case le:return!(t.byteLength!=e.byteLength||!o(new Pl(t),new Pl(e)));case qt:case Yt:case Qt:return $u(+t,+e);case Gt:return t.name==e.name&&t.message==e.message;case re:case oe:return t==e+\"\";case Zt:var u=Y;case ie:var c=r&vt;if(u||(u=$),t.size!=e.size&&!c)return!1;var s=a.get(t);if(s)return s==e;r|=gt,a.set(t,e);var l=mo(u(t),u(e),r,i,o,a);return a.delete(t),l;case ae:if(gf)return gf.call(t)==gf.call(e)}return!1}function _o(t,e,n,r,i,o){var a=n&vt,u=xo(t),c=u.length,s=xo(e),l=s.length;if(c!=l&&!a)return!1;for(var f=c;f--;){var p=u[f];if(!(a?p in e:bl.call(e,p)))return!1}var h=o.get(t);if(h&&o.get(e))return h==e;var d=!0;o.set(t,e),o.set(e,t);for(var v=a;++f<c;){p=u[f];var g=t[p],m=e[p];if(r)var y=a?r(m,g,p,e,t,o):r(g,m,p,t,e,o);if(!(y===it?g===m||i(g,m,n,r,o):y)){d=!1;break}v||(v=\"constructor\"==p)}if(d&&!v){var _=t.constructor,b=e.constructor;_!=b&&\"constructor\"in t&&\"constructor\"in e&&!(\"function\"==typeof _&&_ instanceof _&&\"function\"==typeof b&&b instanceof b)&&(d=!1)}return o.delete(t),o.delete(e),d}function bo(t){return Rf(Zo(t,it,ma),t+\"\")}function xo(t){return sr(t,Hc,Pf)}function wo(t){return sr(t,qc,Nf)}function Co(t){for(var e=t.name+\"\",n=sf[e],r=bl.call(sf,e)?n.length:0;r--;){var i=n[r],o=i.func;if(null==o||o==t)return i.name}return e}function Mo(t){var e=bl.call(n,\"placeholder\")?n:t;return e.placeholder}function ko(){var t=n.iteratee||Rs;return t=t===Rs?Br:t,arguments.length?t(arguments[0],arguments[1]):t}function Eo(t,e){var n=t.__data__;return Wo(e)?n[\"string\"==typeof e?\"string\":\"hash\"]:n.map}function To(t){for(var e=Hc(t),n=e.length;n--;){var r=e[n],i=t[r];e[n]=[r,i,qo(i)]}return e}function So(t,e){var n=V(t,e);return Lr(n)?n:it}function Po(t){var e=bl.call(t,Ul),n=t[Ul];try{t[Ul]=it;var r=!0}catch(t){}var i=Cl.call(t);return r&&(e?t[Ul]=n:delete t[Ul]),i}function No(t,e,n){for(var r=-1,i=n.length;++r<i;){var o=n[r],a=o.size;switch(o.type){case\"drop\":t+=a;break;case\"dropRight\":e-=a;break;case\"take\":e=Xl(e,t+a);break;case\"takeRight\":t=$l(t,e-a)}}return{start:t,end:e}}function Ao(t){var e=t.match(We);return e?e[1].split(Ve):[]}function Oo(t,e,n){e=Ei(e,t);for(var r=-1,i=e.length,o=!1;++r<i;){var a=ra(e[r]);if(!(o=null!=t&&n(t,a)))break;t=t[a]}return o||++r!=i?o:(i=null==t?0:t.length,!!i&&uc(i)&&Fo(a,i)&&(xp(t)||bp(t)))}function Io(t){var e=t.length,n=t.constructor(e);return e&&\"string\"==typeof t[0]&&bl.call(t,\"index\")&&(n.index=t.index,n.input=t.input),n}function Do(t){return\"function\"!=typeof t.constructor||Ho(t)?{}:yf(Al(t))}function Ro(t,e,n,r){var i=t.constructor;switch(e){case le:return Pi(t);case qt:case Yt:return new i(+t);case fe:return Ni(t,r);case pe:case he:case de:case ve:case ge:case me:case ye:case _e:case be:return Ri(t,r);case Zt:return Ai(t,r,n);case Qt:case oe:return new i(t);case re:return Oi(t);case ie:return Ii(t,r,n);case ae:return Di(t)}}function Lo(t,e){var n=e.length;if(!n)return t;var r=n-1;return e[r]=(n>1?\"& \":\"\")+e[r],e=e.join(n>2?\", \":\" \"),t.replace(Be,\"{\\n/* [wrapped with \"+e+\"] */\\n\")}function Uo(t){return xp(t)||bp(t)||!!(Rl&&t&&t[Rl])}function Fo(t,e){return e=null==e?Rt:e,!!e&&(\"number\"==typeof t||Ze.test(t))&&t>-1&&t%1==0&&t<e}function jo(t,e,n){if(!cc(n))return!1;var r=typeof e;return!!(\"number\"==r?Xu(n)&&Fo(e,n.length):\"string\"==r&&e in n)&&$u(n[e],t)}function Bo(t,e){if(xp(t))return!1;var n=typeof t;return!(\"number\"!=n&&\"symbol\"!=n&&\"boolean\"!=n&&null!=t&&!bc(t))||(Oe.test(t)||!Ae.test(t)||null!=e&&t in fl(e))}function Wo(t){var e=typeof t;return\"string\"==e||\"number\"==e||\"symbol\"==e||\"boolean\"==e?\"__proto__\"!==t:null===t}function Vo(t){var e=Co(t),r=n[e];if(\"function\"!=typeof r||!(e in b.prototype))return!1;if(t===r)return!0;var i=Sf(r);return!!i&&t===i[0]}function zo(t){return!!wl&&wl in t}function Ho(t){var e=t&&t.constructor,n=\"function\"==typeof e&&e.prototype||ml;return t===n}function qo(t){return t===t&&!cc(t)}function Yo(t,e){return function(n){return null!=n&&(n[t]===e&&(e!==it||t in fl(n)))}}function Ko(t){var e=Ru(t,function(t){return n.size===lt&&n.clear(),t}),n=e.cache;return e}function Go(t,e){var n=t[1],r=e[1],i=n|r,o=i<(mt|yt|Mt),a=r==Mt&&n==bt||r==Mt&&n==kt&&t[7].length<=e[8]||r==(Mt|kt)&&e[7].length<=e[8]&&n==bt;if(!o&&!a)return t;r&mt&&(t[2]=e[2],i|=n&mt?0:_t);var u=e[3];if(u){var c=t[3];t[3]=c?Fi(c,u,e[4]):u,t[4]=c?G(t[3],ft):e[4]}return u=e[5],u&&(c=t[5],t[5]=c?ji(c,u,e[6]):u,t[6]=c?G(t[5],ft):e[6]),u=e[7],u&&(t[7]=u),r&Mt&&(t[8]=null==t[8]?e[8]:Xl(t[8],e[8])),null==t[9]&&(t[9]=e[9]),t[0]=e[0],t[1]=i,t}function $o(t){var e=[];if(null!=t)for(var n in fl(t))e.push(n);return e}function Xo(t){return Cl.call(t)}function Zo(t,e,n){return e=$l(e===it?t.length-1:e,0),function(){for(var r=arguments,i=-1,o=$l(r.length-e,0),a=al(o);++i<o;)a[i]=r[e+i];i=-1;for(var c=al(e+1);++i<e;)c[i]=r[i];return c[e]=n(a),u(t,this,c)}}function Qo(t,e){return e.length<2?t:cr(t,li(e,0,-1))}function Jo(t,e){for(var n=t.length,r=Xl(e.length,n),i=Bi(t);r--;){var o=e[r];t[r]=Fo(o,n)?i[o]:it}return t}function ta(t,e,n){var r=e+\"\";return Rf(t,Lo(r,oa(Ao(r),n)))}function ea(t){var e=0,n=0;return function(){var r=Zl(),i=Nt-(r-n);if(n=r,i>0){if(++e>=Pt)return arguments[0]}else e=0;return t.apply(it,arguments)}}function na(t,e){var n=-1,r=t.length,i=r-1;for(e=e===it?r:e;++n<e;){var o=ni(n,i),a=t[o];t[o]=t[n],t[n]=a}return t.length=e,t}function ra(t){if(\"string\"==typeof t||bc(t))return t;var e=t+\"\";return\"0\"==e&&1/t==-Dt?\"-0\":e}function ia(t){if(null!=t){try{return _l.call(t)}catch(t){}try{return t+\"\"}catch(t){}}return\"\"}function oa(t,e){return s(Wt,function(n){var r=\"_.\"+n[0];e&n[1]&&!h(t,r)&&t.push(r)}),t.sort()}function aa(t){if(t instanceof b)return t.clone();var e=new i(t.__wrapped__,t.__chain__);return e.__actions__=Bi(t.__actions__),e.__index__=t.__index__,e.__values__=t.__values__,e}function ua(t,e,n){e=(n?jo(t,e,n):e===it)?1:$l(Ec(e),0);var r=null==t?0:t.length;if(!r||e<1)return[];for(var i=0,o=0,a=al(Vl(r/e));i<r;)a[o++]=li(t,i,i+=e);return a}function ca(t){for(var e=-1,n=null==t?0:t.length,r=0,i=[];++e<n;){var o=t[e];o&&(i[r++]=o)}return i}function sa(){var t=arguments.length;if(!t)return[];for(var e=al(t-1),n=arguments[0],r=t;r--;)e[r-1]=arguments[r];return g(xp(n)?Bi(n):[n],er(e,1))}function la(t,e,n){var r=null==t?0:t.length;return r?(e=n||e===it?1:Ec(e),li(t,e<0?0:e,r)):[]}function fa(t,e,n){var r=null==t?0:t.length;return r?(e=n||e===it?1:Ec(e),e=r-e,li(t,0,e<0?0:e)):[]}function pa(t,e){return t&&t.length?bi(t,ko(e,3),!0,!0):[]}function ha(t,e){return t&&t.length?bi(t,ko(e,3),!0):[]}function da(t,e,n,r){var i=null==t?0:t.length;return i?(n&&\"number\"!=typeof n&&jo(t,e,n)&&(n=0,r=i),Jn(t,e,n,r)):[]}function va(t,e,n){var r=null==t?0:t.length;if(!r)return-1;var i=null==n?0:Ec(n);return i<0&&(i=$l(r+i,0)),C(t,ko(e,3),i)}function ga(t,e,n){var r=null==t?0:t.length;if(!r)return-1;var i=r-1;return n!==it&&(i=Ec(n),i=n<0?$l(r+i,0):Xl(i,r-1)),C(t,ko(e,3),i,!0)}function ma(t){var e=null==t?0:t.length;return e?er(t,1):[]}function ya(t){var e=null==t?0:t.length;return e?er(t,Dt):[]}function _a(t,e){var n=null==t?0:t.length;return n?(e=e===it?1:Ec(e),er(t,e)):[]}function ba(t){for(var e=-1,n=null==t?0:t.length,r={};++e<n;){var i=t[e];r[i[0]]=i[1]}return r}function xa(t){return t&&t.length?t[0]:it}function wa(t,e,n){var r=null==t?0:t.length;if(!r)return-1;var i=null==n?0:Ec(n);return i<0&&(i=$l(r+i,0)),M(t,e,i)}function Ca(t){var e=null==t?0:t.length;return e?li(t,0,-1):[]}function Ma(t,e){return null==t?\"\":Kl.call(t,e)}function ka(t){var e=null==t?0:t.length;return e?t[e-1]:it}function Ea(t,e,n){var r=null==t?0:t.length;if(!r)return-1;var i=r;return n!==it&&(i=Ec(n),i=i<0?$l(r+i,0):Xl(i,r-1)),e===e?Q(t,e,i):C(t,E,i,!0)}function Ta(t,e){return t&&t.length?$r(t,Ec(e)):it}function Sa(t,e){return t&&t.length&&e&&e.length?ti(t,e):t}function Pa(t,e,n){return t&&t.length&&e&&e.length?ti(t,e,ko(n,2)):t}function Na(t,e,n){return t&&t.length&&e&&e.length?ti(t,e,it,n):t}function Aa(t,e){var n=[];if(!t||!t.length)return n;var r=-1,i=[],o=t.length;for(e=ko(e,3);++r<o;){var a=t[r];e(a,r,t)&&(n.push(a),i.push(r))}return ei(t,i),n}function Oa(t){return null==t?t:tf.call(t)}function Ia(t,e,n){var r=null==t?0:t.length;return r?(n&&\"number\"!=typeof n&&jo(t,e,n)?(e=0,n=r):(e=null==e?0:Ec(e),n=n===it?r:Ec(n)),li(t,e,n)):[]}function Da(t,e){return pi(t,e)}function Ra(t,e,n){return hi(t,e,ko(n,2))}function La(t,e){var n=null==t?0:t.length;if(n){var r=pi(t,e);if(r<n&&$u(t[r],e))return r}return-1}function Ua(t,e){return pi(t,e,!0)}function Fa(t,e,n){return hi(t,e,ko(n,2),!0)}function ja(t,e){var n=null==t?0:t.length;if(n){var r=pi(t,e,!0)-1;if($u(t[r],e))return r}return-1}function Ba(t){return t&&t.length?di(t):[]}function Wa(t,e){return t&&t.length?di(t,ko(e,2)):[]}function Va(t){var e=null==t?0:t.length;return e?li(t,1,e):[]}function za(t,e,n){return t&&t.length?(e=n||e===it?1:Ec(e),li(t,0,e<0?0:e)):[]}function Ha(t,e,n){var r=null==t?0:t.length;return r?(e=n||e===it?1:Ec(e),e=r-e,li(t,e<0?0:e,r)):[]}function qa(t,e){return t&&t.length?bi(t,ko(e,3),!1,!0):[]}function Ya(t,e){return t&&t.length?bi(t,ko(e,3)):[]}function Ka(t){return t&&t.length?mi(t):[]}function Ga(t,e){return t&&t.length?mi(t,ko(e,2)):[]}function $a(t,e){return e=\"function\"==typeof e?e:it,t&&t.length?mi(t,it,e):[]}function Xa(t){if(!t||!t.length)return[];var e=0;return t=p(t,function(t){if(Zu(t))return e=$l(t.length,e),!0}),I(e,function(e){return v(t,S(e))})}function Za(t,e){if(!t||!t.length)return[];var n=Xa(t);return null==e?n:v(n,function(t){return u(e,it,t)})}function Qa(t,e){return Ci(t||[],e||[],On)}function Ja(t,e){return Ci(t||[],e||[],ci)}function tu(t){var e=n(t);return e.__chain__=!0,e}function eu(t,e){return e(t),t}function nu(t,e){return e(t)}function ru(){return tu(this)}function iu(){return new i(this.value(),this.__chain__)}function ou(){this.__values__===it&&(this.__values__=Mc(this.value()));var t=this.__index__>=this.__values__.length,e=t?it:this.__values__[this.__index__++];return{done:t,value:e}}function au(){return this}function uu(t){for(var e,n=this;n instanceof r;){var i=aa(n);i.__index__=0,i.__values__=it,e?o.__wrapped__=i:e=i;var o=i;n=n.__wrapped__}return o.__wrapped__=t,e}function cu(){var t=this.__wrapped__;if(t instanceof b){var e=t;return this.__actions__.length&&(e=new b(this)),e=e.reverse(),e.__actions__.push({func:nu,args:[Oa],thisArg:it}),new i(e,this.__chain__)}return this.thru(Oa)}function su(){return xi(this.__wrapped__,this.__actions__)}function lu(t,e,n){\n",
              "var r=xp(t)?f:Kn;return n&&jo(t,e,n)&&(e=it),r(t,ko(e,3))}function fu(t,e){var n=xp(t)?p:tr;return n(t,ko(e,3))}function pu(t,e){return er(yu(t,e),1)}function hu(t,e){return er(yu(t,e),Dt)}function du(t,e,n){return n=n===it?1:Ec(n),er(yu(t,e),n)}function vu(t,e){var n=xp(t)?s:_f;return n(t,ko(e,3))}function gu(t,e){var n=xp(t)?l:bf;return n(t,ko(e,3))}function mu(t,e,n,r){t=Xu(t)?t:rs(t),n=n&&!r?Ec(n):0;var i=t.length;return n<0&&(n=$l(i+n,0)),_c(t)?n<=i&&t.indexOf(e,n)>-1:!!i&&M(t,e,n)>-1}function yu(t,e){var n=xp(t)?v:Hr;return n(t,ko(e,3))}function _u(t,e,n,r){return null==t?[]:(xp(e)||(e=null==e?[]:[e]),n=r?it:n,xp(n)||(n=null==n?[]:[n]),Xr(t,e,n))}function bu(t,e,n){var r=xp(t)?m:N,i=arguments.length<3;return r(t,ko(e,4),n,i,_f)}function xu(t,e,n){var r=xp(t)?y:N,i=arguments.length<3;return r(t,ko(e,4),n,i,bf)}function wu(t,e){var n=xp(t)?p:tr;return n(t,Lu(ko(e,3)))}function Cu(t){var e=xp(t)?Sn:ai;return e(t)}function Mu(t,e,n){e=(n?jo(t,e,n):e===it)?1:Ec(e);var r=xp(t)?Pn:ui;return r(t,e)}function ku(t){var e=xp(t)?Nn:si;return e(t)}function Eu(t){if(null==t)return 0;if(Xu(t))return _c(t)?J(t):t.length;var e=Af(t);return e==Zt||e==ie?t.size:Wr(t).length}function Tu(t,e,n){var r=xp(t)?_:fi;return n&&jo(t,e,n)&&(e=it),r(t,ko(e,3))}function Su(t,e){if(\"function\"!=typeof e)throw new dl(ct);return t=Ec(t),function(){if(--t<1)return e.apply(this,arguments)}}function Pu(t,e,n){return e=n?it:e,e=t&&null==e?t.length:e,po(t,Mt,it,it,it,it,e)}function Nu(t,e){var n;if(\"function\"!=typeof e)throw new dl(ct);return t=Ec(t),function(){return--t>0&&(n=e.apply(this,arguments)),t<=1&&(e=it),n}}function Au(t,e,n){e=n?it:e;var r=po(t,bt,it,it,it,it,it,e);return r.placeholder=Au.placeholder,r}function Ou(t,e,n){e=n?it:e;var r=po(t,xt,it,it,it,it,it,e);return r.placeholder=Ou.placeholder,r}function Iu(t,e,n){function r(e){var n=p,r=h;return p=h=it,y=e,v=t.apply(r,n)}function i(t){return y=t,g=Df(u,e),_?r(t):v}function o(t){var n=t-m,r=t-y,i=e-n;return b?Xl(i,d-r):i}function a(t){var n=t-m,r=t-y;return m===it||n>=e||n<0||b&&r>=d}function u(){var t=sp();return a(t)?c(t):void(g=Df(u,o(t)))}function c(t){return g=it,x&&p?r(t):(p=h=it,v)}function s(){g!==it&&Ef(g),y=0,p=m=h=g=it}function l(){return g===it?v:c(sp())}function f(){var t=sp(),n=a(t);if(p=arguments,h=this,m=t,n){if(g===it)return i(m);if(b)return g=Df(u,e),r(m)}return g===it&&(g=Df(u,e)),v}var p,h,d,v,g,m,y=0,_=!1,b=!1,x=!0;if(\"function\"!=typeof t)throw new dl(ct);return e=Sc(e)||0,cc(n)&&(_=!!n.leading,b=\"maxWait\"in n,d=b?$l(Sc(n.maxWait)||0,e):d,x=\"trailing\"in n?!!n.trailing:x),f.cancel=s,f.flush=l,f}function Du(t){return po(t,Et)}function Ru(t,e){if(\"function\"!=typeof t||null!=e&&\"function\"!=typeof e)throw new dl(ct);var n=function(){var r=arguments,i=e?e.apply(this,r):r[0],o=n.cache;if(o.has(i))return o.get(i);var a=t.apply(this,r);return n.cache=o.set(i,a)||o,a};return n.cache=new(Ru.Cache||pn),n}function Lu(t){if(\"function\"!=typeof t)throw new dl(ct);return function(){var e=arguments;switch(e.length){case 0:return!t.call(this);case 1:return!t.call(this,e[0]);case 2:return!t.call(this,e[0],e[1]);case 3:return!t.call(this,e[0],e[1],e[2])}return!t.apply(this,e)}}function Uu(t){return Nu(2,t)}function Fu(t,e){if(\"function\"!=typeof t)throw new dl(ct);return e=e===it?e:Ec(e),oi(t,e)}function ju(t,e){if(\"function\"!=typeof t)throw new dl(ct);return e=null==e?0:$l(Ec(e),0),oi(function(n){var r=n[e],i=Ti(n,0,e);return r&&g(i,r),u(t,this,i)})}function Bu(t,e,n){var r=!0,i=!0;if(\"function\"!=typeof t)throw new dl(ct);return cc(n)&&(r=\"leading\"in n?!!n.leading:r,i=\"trailing\"in n?!!n.trailing:i),Iu(t,e,{leading:r,maxWait:e,trailing:i})}function Wu(t){return Pu(t,1)}function Vu(t,e){return vp(ki(e),t)}function zu(){if(!arguments.length)return[];var t=arguments[0];return xp(t)?t:[t]}function Hu(t){return Bn(t,dt)}function qu(t,e){return e=\"function\"==typeof e?e:it,Bn(t,dt,e)}function Yu(t){return Bn(t,pt|dt)}function Ku(t,e){return e=\"function\"==typeof e?e:it,Bn(t,pt|dt,e)}function Gu(t,e){return null==e||Vn(t,e,Hc(e))}function $u(t,e){return t===e||t!==t&&e!==e}function Xu(t){return null!=t&&uc(t.length)&&!oc(t)}function Zu(t){return sc(t)&&Xu(t)}function Qu(t){return t===!0||t===!1||sc(t)&&fr(t)==qt}function Ju(t){return sc(t)&&1===t.nodeType&&!mc(t)}function tc(t){if(null==t)return!0;if(Xu(t)&&(xp(t)||\"string\"==typeof t||\"function\"==typeof t.splice||Cp(t)||Sp(t)||bp(t)))return!t.length;var e=Af(t);if(e==Zt||e==ie)return!t.size;if(Ho(t))return!Wr(t).length;for(var n in t)if(bl.call(t,n))return!1;return!0}function ec(t,e){return Or(t,e)}function nc(t,e,n){n=\"function\"==typeof n?n:it;var r=n?n(t,e):it;return r===it?Or(t,e,it,n):!!r}function rc(t){if(!sc(t))return!1;var e=fr(t);return e==Gt||e==Kt||\"string\"==typeof t.message&&\"string\"==typeof t.name&&!mc(t)}function ic(t){return\"number\"==typeof t&&Yl(t)}function oc(t){if(!cc(t))return!1;var e=fr(t);return e==$t||e==Xt||e==Ht||e==ne}function ac(t){return\"number\"==typeof t&&t==Ec(t)}function uc(t){return\"number\"==typeof t&&t>-1&&t%1==0&&t<=Rt}function cc(t){var e=typeof t;return null!=t&&(\"object\"==e||\"function\"==e)}function sc(t){return null!=t&&\"object\"==typeof t}function lc(t,e){return t===e||Rr(t,e,To(e))}function fc(t,e,n){return n=\"function\"==typeof n?n:it,Rr(t,e,To(e),n)}function pc(t){return gc(t)&&t!=+t}function hc(t){if(Of(t))throw new cl(ut);return Lr(t)}function dc(t){return null===t}function vc(t){return null==t}function gc(t){return\"number\"==typeof t||sc(t)&&fr(t)==Qt}function mc(t){if(!sc(t)||fr(t)!=te)return!1;var e=Al(t);if(null===e)return!0;var n=bl.call(e,\"constructor\")&&e.constructor;return\"function\"==typeof n&&n instanceof n&&_l.call(n)==Ml}function yc(t){return ac(t)&&t>=-Rt&&t<=Rt}function _c(t){return\"string\"==typeof t||!xp(t)&&sc(t)&&fr(t)==oe}function bc(t){return\"symbol\"==typeof t||sc(t)&&fr(t)==ae}function xc(t){return t===it}function wc(t){return sc(t)&&Af(t)==ce}function Cc(t){return sc(t)&&fr(t)==se}function Mc(t){if(!t)return[];if(Xu(t))return _c(t)?tt(t):Bi(t);if(Ll&&t[Ll])return q(t[Ll]());var e=Af(t),n=e==Zt?Y:e==ie?$:rs;return n(t)}function kc(t){if(!t)return 0===t?t:0;if(t=Sc(t),t===Dt||t===-Dt){var e=t<0?-1:1;return e*Lt}return t===t?t:0}function Ec(t){var e=kc(t),n=e%1;return e===e?n?e-n:e:0}function Tc(t){return t?jn(Ec(t),0,Ft):0}function Sc(t){if(\"number\"==typeof t)return t;if(bc(t))return Ut;if(cc(t)){var e=\"function\"==typeof t.valueOf?t.valueOf():t;t=cc(e)?e+\"\":e}if(\"string\"!=typeof t)return 0===t?t:+t;t=t.replace(Ue,\"\");var n=Ge.test(t);return n||Xe.test(t)?ir(t.slice(2),n?2:8):Ke.test(t)?Ut:+t}function Pc(t){return Wi(t,qc(t))}function Nc(t){return t?jn(Ec(t),-Rt,Rt):0===t?t:0}function Ac(t){return null==t?\"\":gi(t)}function Oc(t,e){var n=yf(t);return null==e?n:Rn(n,e)}function Ic(t,e){return w(t,ko(e,3),nr)}function Dc(t,e){return w(t,ko(e,3),or)}function Rc(t,e){return null==t?t:xf(t,ko(e,3),qc)}function Lc(t,e){return null==t?t:wf(t,ko(e,3),qc)}function Uc(t,e){return t&&nr(t,ko(e,3))}function Fc(t,e){return t&&or(t,ko(e,3))}function jc(t){return null==t?[]:ar(t,Hc(t))}function Bc(t){return null==t?[]:ar(t,qc(t))}function Wc(t,e,n){var r=null==t?it:cr(t,e);return r===it?n:r}function Vc(t,e){return null!=t&&Oo(t,e,_r)}function zc(t,e){return null!=t&&Oo(t,e,Cr)}function Hc(t){return Xu(t)?Tn(t):Wr(t)}function qc(t){return Xu(t)?Tn(t,!0):Vr(t)}function Yc(t,e){var n={};return e=ko(e,3),nr(t,function(t,r,i){Un(n,e(t,r,i),t)}),n}function Kc(t,e){var n={};return e=ko(e,3),nr(t,function(t,r,i){Un(n,r,e(t,r,i))}),n}function Gc(t,e){return $c(t,Lu(ko(e)))}function $c(t,e){if(null==t)return{};var n=v(wo(t),function(t){return[t]});return e=ko(e),Qr(t,n,function(t,n){return e(t,n[0])})}function Xc(t,e,n){e=Ei(e,t);var r=-1,i=e.length;for(i||(i=1,t=it);++r<i;){var o=null==t?it:t[ra(e[r])];o===it&&(r=i,o=n),t=oc(o)?o.call(t):o}return t}function Zc(t,e,n){return null==t?t:ci(t,e,n)}function Qc(t,e,n,r){return r=\"function\"==typeof r?r:it,null==t?t:ci(t,e,n,r)}function Jc(t,e,n){var r=xp(t),i=r||Cp(t)||Sp(t);if(e=ko(e,4),null==n){var o=t&&t.constructor;n=i?r?new o:[]:cc(t)&&oc(o)?yf(Al(t)):{}}return(i?s:nr)(t,function(t,r,i){return e(n,t,r,i)}),n}function ts(t,e){return null==t||yi(t,e)}function es(t,e,n){return null==t?t:_i(t,e,ki(n))}function ns(t,e,n,r){return r=\"function\"==typeof r?r:it,null==t?t:_i(t,e,ki(n),r)}function rs(t){return null==t?[]:L(t,Hc(t))}function is(t){return null==t?[]:L(t,qc(t))}function os(t,e,n){return n===it&&(n=e,e=it),n!==it&&(n=Sc(n),n=n===n?n:0),e!==it&&(e=Sc(e),e=e===e?e:0),jn(Sc(t),e,n)}function as(t,e,n){return e=kc(e),n===it?(n=e,e=0):n=kc(n),t=Sc(t),kr(t,e,n)}function us(t,e,n){if(n&&\"boolean\"!=typeof n&&jo(t,e,n)&&(e=n=it),n===it&&(\"boolean\"==typeof e?(n=e,e=it):\"boolean\"==typeof t&&(n=t,t=it)),t===it&&e===it?(t=0,e=1):(t=kc(t),e===it?(e=t,t=0):e=kc(e)),t>e){var r=t;t=e,e=r}if(n||t%1||e%1){var i=Jl();return Xl(t+i*(e-t+rr(\"1e-\"+((i+\"\").length-1))),e)}return ni(t,e)}function cs(t){return th(Ac(t).toLowerCase())}function ss(t){return t=Ac(t),t&&t.replace(Qe,br).replace(Hn,\"\")}function ls(t,e,n){t=Ac(t),e=gi(e);var r=t.length;n=n===it?r:jn(Ec(n),0,r);var i=n;return n-=e.length,n>=0&&t.slice(n,i)==e}function fs(t){return t=Ac(t),t&&Te.test(t)?t.replace(ke,xr):t}function ps(t){return t=Ac(t),t&&Le.test(t)?t.replace(Re,\"\\\\$&\"):t}function hs(t,e,n){t=Ac(t),e=Ec(e);var r=e?J(t):0;if(!e||r>=e)return t;var i=(e-r)/2;return oo(zl(i),n)+t+oo(Vl(i),n)}function ds(t,e,n){t=Ac(t),e=Ec(e);var r=e?J(t):0;return e&&r<e?t+oo(e-r,n):t}function vs(t,e,n){t=Ac(t),e=Ec(e);var r=e?J(t):0;return e&&r<e?oo(e-r,n)+t:t}function gs(t,e,n){return n||null==e?e=0:e&&(e=+e),Ql(Ac(t).replace(Fe,\"\"),e||0)}function ms(t,e,n){return e=(n?jo(t,e,n):e===it)?1:Ec(e),ii(Ac(t),e)}function ys(){var t=arguments,e=Ac(t[0]);return t.length<3?e:e.replace(t[1],t[2])}function _s(t,e,n){return n&&\"number\"!=typeof n&&jo(t,e,n)&&(e=n=it),(n=n===it?Ft:n>>>0)?(t=Ac(t),t&&(\"string\"==typeof e||null!=e&&!Ep(e))&&(e=gi(e),!e&&z(t))?Ti(tt(t),0,n):t.split(e,n)):[]}function bs(t,e,n){return t=Ac(t),n=null==n?0:jn(Ec(n),0,t.length),e=gi(e),t.slice(n,n+e.length)==e}function xs(t,e,r){var i=n.templateSettings;r&&jo(t,e,r)&&(e=it),t=Ac(t),e=Ip({},e,i,ho);var o,a,u=Ip({},e.imports,i.imports,ho),c=Hc(u),s=L(u,c),l=0,f=e.interpolate||Je,p=\"__p += '\",h=pl((e.escape||Je).source+\"|\"+f.source+\"|\"+(f===Ne?qe:Je).source+\"|\"+(e.evaluate||Je).source+\"|$\",\"g\"),d=\"//# sourceURL=\"+(\"sourceURL\"in e?e.sourceURL:\"lodash.templateSources[\"+ ++Xn+\"]\")+\"\\n\";t.replace(h,function(e,n,r,i,u,c){return r||(r=i),p+=t.slice(l,c).replace(tn,W),n&&(o=!0,p+=\"' +\\n__e(\"+n+\") +\\n'\"),u&&(a=!0,p+=\"';\\n\"+u+\";\\n__p += '\"),r&&(p+=\"' +\\n((__t = (\"+r+\")) == null ? '' : __t) +\\n'\"),l=c+e.length,e}),p+=\"';\\n\";var v=e.variable;v||(p=\"with (obj) {\\n\"+p+\"\\n}\\n\"),p=(a?p.replace(xe,\"\"):p).replace(we,\"$1\").replace(Ce,\"$1;\"),p=\"function(\"+(v||\"obj\")+\") {\\n\"+(v?\"\":\"obj || (obj = {});\\n\")+\"var __t, __p = ''\"+(o?\", __e = _.escape\":\"\")+(a?\", __j = Array.prototype.join;\\nfunction print() { __p += __j.call(arguments, '') }\\n\":\";\\n\")+p+\"return __p\\n}\";var g=eh(function(){return sl(c,d+\"return \"+p).apply(it,s)});if(g.source=p,rc(g))throw g;return g}function ws(t){return Ac(t).toLowerCase()}function Cs(t){return Ac(t).toUpperCase()}function Ms(t,e,n){if(t=Ac(t),t&&(n||e===it))return t.replace(Ue,\"\");if(!t||!(e=gi(e)))return t;var r=tt(t),i=tt(e),o=F(r,i),a=j(r,i)+1;return Ti(r,o,a).join(\"\")}function ks(t,e,n){if(t=Ac(t),t&&(n||e===it))return t.replace(je,\"\");if(!t||!(e=gi(e)))return t;var r=tt(t),i=j(r,tt(e))+1;return Ti(r,0,i).join(\"\")}function Es(t,e,n){if(t=Ac(t),t&&(n||e===it))return t.replace(Fe,\"\");if(!t||!(e=gi(e)))return t;var r=tt(t),i=F(r,tt(e));return Ti(r,i).join(\"\")}function Ts(t,e){var n=Tt,r=St;if(cc(e)){var i=\"separator\"in e?e.separator:i;n=\"length\"in e?Ec(e.length):n,r=\"omission\"in e?gi(e.omission):r}t=Ac(t);var o=t.length;if(z(t)){var a=tt(t);o=a.length}if(n>=o)return t;var u=n-J(r);if(u<1)return r;var c=a?Ti(a,0,u).join(\"\"):t.slice(0,u);if(i===it)return c+r;if(a&&(u+=c.length-u),Ep(i)){if(t.slice(u).search(i)){var s,l=c;for(i.global||(i=pl(i.source,Ac(Ye.exec(i))+\"g\")),i.lastIndex=0;s=i.exec(l);)var f=s.index;c=c.slice(0,f===it?u:f)}}else if(t.indexOf(gi(i),u)!=u){var p=c.lastIndexOf(i);p>-1&&(c=c.slice(0,p))}return c+r}function Ss(t){return t=Ac(t),t&&Ee.test(t)?t.replace(Me,wr):t}function Ps(t,e,n){return t=Ac(t),e=n?it:e,e===it?H(t)?rt(t):x(t):t.match(e)||[]}function Ns(t){var e=null==t?0:t.length,n=ko();return t=e?v(t,function(t){if(\"function\"!=typeof t[1])throw new dl(ct);return[n(t[0]),t[1]]}):[],oi(function(n){for(var r=-1;++r<e;){var i=t[r];if(u(i[0],this,n))return u(i[1],this,n)}})}function As(t){return Wn(Bn(t,pt))}function Os(t){return function(){return t}}function Is(t,e){return null==t||t!==t?e:t}function Ds(t){return t}function Rs(t){return Br(\"function\"==typeof t?t:Bn(t,pt))}function Ls(t){return qr(Bn(t,pt))}function Us(t,e){return Yr(t,Bn(e,pt))}function Fs(t,e,n){var r=Hc(e),i=ar(e,r);null!=n||cc(e)&&(i.length||!r.length)||(n=e,e=t,t=this,i=ar(e,Hc(e)));var o=!(cc(n)&&\"chain\"in n&&!n.chain),a=oc(t);return s(i,function(n){var r=e[n];t[n]=r,a&&(t.prototype[n]=function(){var e=this.__chain__;if(o||e){var n=t(this.__wrapped__),i=n.__actions__=Bi(this.__actions__);return i.push({func:r,args:arguments,thisArg:t}),n.__chain__=e,n}return r.apply(t,g([this.value()],arguments))})}),t}function js(){return ur._===this&&(ur._=kl),this}function Bs(){}function Ws(t){return t=Ec(t),oi(function(e){return $r(e,t)})}function Vs(t){return Bo(t)?S(ra(t)):Jr(t)}function zs(t){return function(e){return null==t?it:cr(t,e)}}function Hs(){return[]}function qs(){return!1}function Ys(){return{}}function Ks(){return\"\"}function Gs(){return!0}function $s(t,e){if(t=Ec(t),t<1||t>Rt)return[];var n=Ft,r=Xl(t,Ft);e=ko(e),t-=Ft;for(var i=I(r,e);++n<t;)e(n);return i}function Xs(t){return xp(t)?v(t,ra):bc(t)?[t]:Bi(Lf(Ac(t)))}function Zs(t){var e=++xl;return Ac(t)+e}function Qs(t){return t&&t.length?Gn(t,Ds,pr):it}function Js(t,e){return t&&t.length?Gn(t,ko(e,2),pr):it}function tl(t){return T(t,Ds)}function el(t,e){return T(t,ko(e,2))}function nl(t){return t&&t.length?Gn(t,Ds,zr):it}function rl(t,e){return t&&t.length?Gn(t,ko(e,2),zr):it}function il(t){return t&&t.length?O(t,Ds):0}function ol(t,e){return t&&t.length?O(t,ko(e,2)):0}e=null==e?ur:Mr.defaults(ur.Object(),e,Mr.pick(ur,$n));var al=e.Array,ul=e.Date,cl=e.Error,sl=e.Function,ll=e.Math,fl=e.Object,pl=e.RegExp,hl=e.String,dl=e.TypeError,vl=al.prototype,gl=sl.prototype,ml=fl.prototype,yl=e[\"__core-js_shared__\"],_l=gl.toString,bl=ml.hasOwnProperty,xl=0,wl=function(){var t=/[^.]+$/.exec(yl&&yl.keys&&yl.keys.IE_PROTO||\"\");return t?\"Symbol(src)_1.\"+t:\"\"}(),Cl=ml.toString,Ml=_l.call(fl),kl=ur._,El=pl(\"^\"+_l.call(bl).replace(Re,\"\\\\$&\").replace(/hasOwnProperty|(function).*?(?=\\\\\\()| for .+?(?=\\\\\\])/g,\"$1.*?\")+\"$\"),Tl=lr?e.Buffer:it,Sl=e.Symbol,Pl=e.Uint8Array,Nl=Tl?Tl.allocUnsafe:it,Al=K(fl.getPrototypeOf,fl),Ol=fl.create,Il=ml.propertyIsEnumerable,Dl=vl.splice,Rl=Sl?Sl.isConcatSpreadable:it,Ll=Sl?Sl.iterator:it,Ul=Sl?Sl.toStringTag:it,Fl=function(){try{var t=So(fl,\"defineProperty\");return t({},\"\",{}),t}catch(t){}}(),jl=e.clearTimeout!==ur.clearTimeout&&e.clearTimeout,Bl=ul&&ul.now!==ur.Date.now&&ul.now,Wl=e.setTimeout!==ur.setTimeout&&e.setTimeout,Vl=ll.ceil,zl=ll.floor,Hl=fl.getOwnPropertySymbols,ql=Tl?Tl.isBuffer:it,Yl=e.isFinite,Kl=vl.join,Gl=K(fl.keys,fl),$l=ll.max,Xl=ll.min,Zl=ul.now,Ql=e.parseInt,Jl=ll.random,tf=vl.reverse,ef=So(e,\"DataView\"),nf=So(e,\"Map\"),rf=So(e,\"Promise\"),of=So(e,\"Set\"),af=So(e,\"WeakMap\"),uf=So(fl,\"create\"),cf=af&&new af,sf={},lf=ia(ef),ff=ia(nf),pf=ia(rf),hf=ia(of),df=ia(af),vf=Sl?Sl.prototype:it,gf=vf?vf.valueOf:it,mf=vf?vf.toString:it,yf=function(){function t(){}return function(e){if(!cc(e))return{};if(Ol)return Ol(e);t.prototype=e;var n=new t;return t.prototype=it,n}}();n.templateSettings={escape:Se,evaluate:Pe,interpolate:Ne,variable:\"\",imports:{_:n}},n.prototype=r.prototype,n.prototype.constructor=n,i.prototype=yf(r.prototype),i.prototype.constructor=i,b.prototype=yf(r.prototype),b.prototype.constructor=b,nt.prototype.clear=ze,nt.prototype.delete=en,nt.prototype.get=nn,nt.prototype.has=rn,nt.prototype.set=on,an.prototype.clear=un,an.prototype.delete=cn,an.prototype.get=sn,an.prototype.has=ln,an.prototype.set=fn,pn.prototype.clear=hn,pn.prototype.delete=dn,pn.prototype.get=vn,pn.prototype.has=gn,pn.prototype.set=mn,yn.prototype.add=yn.prototype.push=_n,yn.prototype.has=bn,xn.prototype.clear=wn,xn.prototype.delete=Cn,xn.prototype.get=Mn,xn.prototype.has=kn,xn.prototype.set=En;var _f=Yi(nr),bf=Yi(or,!0),xf=Ki(),wf=Ki(!0),Cf=cf?function(t,e){return cf.set(t,e),t}:Ds,Mf=Fl?function(t,e){return Fl(t,\"toString\",{configurable:!0,enumerable:!1,value:Os(e),writable:!0})}:Ds,kf=oi,Ef=jl||function(t){return ur.clearTimeout(t)},Tf=of&&1/$(new of([,-0]))[1]==Dt?function(t){return new of(t)}:Bs,Sf=cf?function(t){return cf.get(t)}:Bs,Pf=Hl?function(t){return null==t?[]:(t=fl(t),p(Hl(t),function(e){return Il.call(t,e)}))}:Hs,Nf=Hl?function(t){for(var e=[];t;)g(e,Pf(t)),t=Al(t);return e}:Hs,Af=fr;(ef&&Af(new ef(new ArrayBuffer(1)))!=fe||nf&&Af(new nf)!=Zt||rf&&Af(rf.resolve())!=ee||of&&Af(new of)!=ie||af&&Af(new af)!=ce)&&(Af=function(t){var e=fr(t),n=e==te?t.constructor:it,r=n?ia(n):\"\";if(r)switch(r){case lf:return fe;case ff:return Zt;case pf:return ee;case hf:return ie;case df:return ce}return e});var Of=yl?oc:qs,If=ea(Cf),Df=Wl||function(t,e){return ur.setTimeout(t,e)},Rf=ea(Mf),Lf=Ko(function(t){var e=[];return Ie.test(t)&&e.push(\"\"),t.replace(De,function(t,n,r,i){e.push(r?i.replace(He,\"$1\"):n||t)}),e}),Uf=oi(function(t,e){return Zu(t)?Yn(t,er(e,1,Zu,!0)):[]}),Ff=oi(function(t,e){var n=ka(e);return Zu(n)&&(n=it),Zu(t)?Yn(t,er(e,1,Zu,!0),ko(n,2)):[]}),jf=oi(function(t,e){var n=ka(e);return Zu(n)&&(n=it),Zu(t)?Yn(t,er(e,1,Zu,!0),it,n):[]}),Bf=oi(function(t){var e=v(t,Mi);return e.length&&e[0]===t[0]?Er(e):[]}),Wf=oi(function(t){var e=ka(t),n=v(t,Mi);return e===ka(n)?e=it:n.pop(),n.length&&n[0]===t[0]?Er(n,ko(e,2)):[]}),Vf=oi(function(t){var e=ka(t),n=v(t,Mi);return e=\"function\"==typeof e?e:it,e&&n.pop(),n.length&&n[0]===t[0]?Er(n,it,e):[]}),zf=oi(Sa),Hf=bo(function(t,e){var n=null==t?0:t.length,r=Fn(t,e);return ei(t,v(e,function(t){return Fo(t,n)?+t:t}).sort(Li)),r}),qf=oi(function(t){return mi(er(t,1,Zu,!0))}),Yf=oi(function(t){var e=ka(t);return Zu(e)&&(e=it),mi(er(t,1,Zu,!0),ko(e,2))}),Kf=oi(function(t){var e=ka(t);return e=\"function\"==typeof e?e:it,mi(er(t,1,Zu,!0),it,e)}),Gf=oi(function(t,e){return Zu(t)?Yn(t,e):[]}),$f=oi(function(t){return wi(p(t,Zu))}),Xf=oi(function(t){var e=ka(t);return Zu(e)&&(e=it),wi(p(t,Zu),ko(e,2))}),Zf=oi(function(t){var e=ka(t);return e=\"function\"==typeof e?e:it,wi(p(t,Zu),it,e)}),Qf=oi(Xa),Jf=oi(function(t){var e=t.length,n=e>1?t[e-1]:it;return n=\"function\"==typeof n?(t.pop(),n):it,Za(t,n)}),tp=bo(function(t){var e=t.length,n=e?t[0]:0,r=this.__wrapped__,o=function(e){return Fn(e,t)};return!(e>1||this.__actions__.length)&&r instanceof b&&Fo(n)?(r=r.slice(n,+n+(e?1:0)),r.__actions__.push({func:nu,args:[o],thisArg:it}),new i(r,this.__chain__).thru(function(t){return e&&!t.length&&t.push(it),t})):this.thru(o)}),ep=Hi(function(t,e,n){bl.call(t,n)?++t[n]:Un(t,n,1)}),np=Ji(va),rp=Ji(ga),ip=Hi(function(t,e,n){bl.call(t,n)?t[n].push(e):Un(t,n,[e])}),op=oi(function(t,e,n){var r=-1,i=\"function\"==typeof e,o=Xu(t)?al(t.length):[];return _f(t,function(t){o[++r]=i?u(e,t,n):Sr(t,e,n)}),o}),ap=Hi(function(t,e,n){Un(t,n,e)}),up=Hi(function(t,e,n){t[n?0:1].push(e)},function(){return[[],[]]}),cp=oi(function(t,e){if(null==t)return[];var n=e.length;return n>1&&jo(t,e[0],e[1])?e=[]:n>2&&jo(e[0],e[1],e[2])&&(e=[e[0]]),Xr(t,er(e,1),[])}),sp=Bl||function(){return ur.Date.now()},lp=oi(function(t,e,n){var r=mt;if(n.length){var i=G(n,Mo(lp));r|=wt}return po(t,r,e,n,i)}),fp=oi(function(t,e,n){var r=mt|yt;if(n.length){var i=G(n,Mo(fp));r|=wt}return po(e,r,t,n,i)}),pp=oi(function(t,e){return qn(t,1,e)}),hp=oi(function(t,e,n){return qn(t,Sc(e)||0,n)});Ru.Cache=pn;var dp=kf(function(t,e){e=1==e.length&&xp(e[0])?v(e[0],R(ko())):v(er(e,1),R(ko()));var n=e.length;return oi(function(r){for(var i=-1,o=Xl(r.length,n);++i<o;)r[i]=e[i].call(this,r[i]);return u(t,this,r)})}),vp=oi(function(t,e){var n=G(e,Mo(vp));return po(t,wt,it,e,n)}),gp=oi(function(t,e){var n=G(e,Mo(gp));return po(t,Ct,it,e,n)}),mp=bo(function(t,e){return po(t,kt,it,it,it,e)}),yp=co(pr),_p=co(function(t,e){return t>=e}),bp=Pr(function(){return arguments}())?Pr:function(t){return sc(t)&&bl.call(t,\"callee\")&&!Il.call(t,\"callee\")},xp=al.isArray,wp=hr?R(hr):Nr,Cp=ql||qs,Mp=dr?R(dr):Ar,kp=vr?R(vr):Dr,Ep=gr?R(gr):Ur,Tp=mr?R(mr):Fr,Sp=yr?R(yr):jr,Pp=co(zr),Np=co(function(t,e){return t<=e}),Ap=qi(function(t,e){if(Ho(e)||Xu(e))return void Wi(e,Hc(e),t);for(var n in e)bl.call(e,n)&&On(t,n,e[n])}),Op=qi(function(t,e){Wi(e,qc(e),t)}),Ip=qi(function(t,e,n,r){Wi(e,qc(e),t,r)}),Dp=qi(function(t,e,n,r){Wi(e,Hc(e),t,r)}),Rp=bo(Fn),Lp=oi(function(t){return t.push(it,ho),u(Ip,it,t)}),Up=oi(function(t){return t.push(it,vo),u(Vp,it,t)}),Fp=no(function(t,e,n){t[e]=n},Os(Ds)),jp=no(function(t,e,n){bl.call(t,e)?t[e].push(n):t[e]=[n]},ko),Bp=oi(Sr),Wp=qi(function(t,e,n){Kr(t,e,n)}),Vp=qi(function(t,e,n,r){Kr(t,e,n,r)}),zp=bo(function(t,e){var n={};if(null==t)return n;var r=!1;e=v(e,function(e){return e=Ei(e,t),r||(r=e.length>1),e}),Wi(t,wo(t),n),r&&(n=Bn(n,pt|ht|dt,go));for(var i=e.length;i--;)yi(n,e[i]);return n}),Hp=bo(function(t,e){return null==t?{}:Zr(t,e)}),qp=fo(Hc),Yp=fo(qc),Kp=Xi(function(t,e,n){return e=e.toLowerCase(),t+(n?cs(e):e)}),Gp=Xi(function(t,e,n){return t+(n?\"-\":\"\")+e.toLowerCase()}),$p=Xi(function(t,e,n){return t+(n?\" \":\"\")+e.toLowerCase()}),Xp=$i(\"toLowerCase\"),Zp=Xi(function(t,e,n){return t+(n?\"_\":\"\")+e.toLowerCase()}),Qp=Xi(function(t,e,n){return t+(n?\" \":\"\")+th(e)}),Jp=Xi(function(t,e,n){return t+(n?\" \":\"\")+e.toUpperCase()}),th=$i(\"toUpperCase\"),eh=oi(function(t,e){try{return u(t,it,e)}catch(t){return rc(t)?t:new cl(t)}}),nh=bo(function(t,e){return s(e,function(e){e=ra(e),Un(t,e,lp(t[e],t))}),t}),rh=to(),ih=to(!0),oh=oi(function(t,e){return function(n){return Sr(n,t,e)}}),ah=oi(function(t,e){return function(n){return Sr(t,n,e)}}),uh=io(v),ch=io(f),sh=io(_),lh=uo(),fh=uo(!0),ph=ro(function(t,e){return t+e},0),hh=lo(\"ceil\"),dh=ro(function(t,e){return t/e},1),vh=lo(\"floor\"),gh=ro(function(t,e){return t*e},1),mh=lo(\"round\"),yh=ro(function(t,e){return t-e},0);return n.after=Su,n.ary=Pu,n.assign=Ap,n.assignIn=Op,n.assignInWith=Ip,n.assignWith=Dp,n.at=Rp,n.before=Nu,n.bind=lp,n.bindAll=nh,n.bindKey=fp,n.castArray=zu,n.chain=tu,n.chunk=ua,n.compact=ca,n.concat=sa,n.cond=Ns,n.conforms=As,n.constant=Os,n.countBy=ep,n.create=Oc,n.curry=Au,n.curryRight=Ou,n.debounce=Iu,n.defaults=Lp,n.defaultsDeep=Up,n.defer=pp,n.delay=hp,n.difference=Uf,n.differenceBy=Ff,n.differenceWith=jf,n.drop=la,n.dropRight=fa,n.dropRightWhile=pa,n.dropWhile=ha,n.fill=da,n.filter=fu,n.flatMap=pu,n.flatMapDeep=hu,n.flatMapDepth=du,n.flatten=ma,n.flattenDeep=ya,n.flattenDepth=_a,n.flip=Du,n.flow=rh,n.flowRight=ih,n.fromPairs=ba,n.functions=jc,n.functionsIn=Bc,n.groupBy=ip,n.initial=Ca,n.intersection=Bf,n.intersectionBy=Wf,n.intersectionWith=Vf,n.invert=Fp,n.invertBy=jp,n.invokeMap=op,n.iteratee=Rs,n.keyBy=ap,n.keys=Hc,n.keysIn=qc,n.map=yu,n.mapKeys=Yc,n.mapValues=Kc,n.matches=Ls,n.matchesProperty=Us,n.memoize=Ru,n.merge=Wp,n.mergeWith=Vp,n.method=oh,n.methodOf=ah,n.mixin=Fs,n.negate=Lu,n.nthArg=Ws,n.omit=zp,n.omitBy=Gc,n.once=Uu,n.orderBy=_u,n.over=uh,n.overArgs=dp,n.overEvery=ch,n.overSome=sh,n.partial=vp,n.partialRight=gp,n.partition=up,n.pick=Hp,n.pickBy=$c,n.property=Vs,n.propertyOf=zs,n.pull=zf,n.pullAll=Sa,n.pullAllBy=Pa,n.pullAllWith=Na,n.pullAt=Hf,n.range=lh,n.rangeRight=fh,n.rearg=mp,n.reject=wu,n.remove=Aa,n.rest=Fu,n.reverse=Oa,n.sampleSize=Mu,n.set=Zc,n.setWith=Qc,n.shuffle=ku,n.slice=Ia,n.sortBy=cp,n.sortedUniq=Ba,n.sortedUniqBy=Wa,n.split=_s,n.spread=ju,n.tail=Va,n.take=za,n.takeRight=Ha,n.takeRightWhile=qa,n.takeWhile=Ya,n.tap=eu,n.throttle=Bu,n.thru=nu,n.toArray=Mc,n.toPairs=qp,n.toPairsIn=Yp,n.toPath=Xs,n.toPlainObject=Pc,n.transform=Jc,n.unary=Wu,n.union=qf,n.unionBy=Yf,n.unionWith=Kf,n.uniq=Ka,n.uniqBy=Ga,n.uniqWith=$a,n.unset=ts,n.unzip=Xa,n.unzipWith=Za,n.update=es,n.updateWith=ns,n.values=rs,n.valuesIn=is,n.without=Gf,n.words=Ps,n.wrap=Vu,n.xor=$f,n.xorBy=Xf,n.xorWith=Zf,n.zip=Qf,n.zipObject=Qa,n.zipObjectDeep=Ja,n.zipWith=Jf,n.entries=qp,n.entriesIn=Yp,n.extend=Op,n.extendWith=Ip,Fs(n,n),n.add=ph,n.attempt=eh,n.camelCase=Kp,n.capitalize=cs,n.ceil=hh,n.clamp=os,n.clone=Hu,n.cloneDeep=Yu,n.cloneDeepWith=Ku,n.cloneWith=qu,n.conformsTo=Gu,n.deburr=ss,n.defaultTo=Is,n.divide=dh,n.endsWith=ls,n.eq=$u,n.escape=fs,n.escapeRegExp=ps,n.every=lu,n.find=np,n.findIndex=va,n.findKey=Ic,n.findLast=rp,n.findLastIndex=ga,n.findLastKey=Dc,n.floor=vh,n.forEach=vu,n.forEachRight=gu,n.forIn=Rc,n.forInRight=Lc,n.forOwn=Uc,n.forOwnRight=Fc,n.get=Wc,n.gt=yp,n.gte=_p,n.has=Vc,n.hasIn=zc,n.head=xa,n.identity=Ds,n.includes=mu,n.indexOf=wa,n.inRange=as,n.invoke=Bp,n.isArguments=bp,n.isArray=xp,n.isArrayBuffer=wp,n.isArrayLike=Xu,n.isArrayLikeObject=Zu,n.isBoolean=Qu,n.isBuffer=Cp,n.isDate=Mp,n.isElement=Ju,n.isEmpty=tc,n.isEqual=ec,n.isEqualWith=nc,n.isError=rc,n.isFinite=ic,n.isFunction=oc,n.isInteger=ac,n.isLength=uc,n.isMap=kp,n.isMatch=lc,n.isMatchWith=fc,n.isNaN=pc,n.isNative=hc,n.isNil=vc,n.isNull=dc,n.isNumber=gc,n.isObject=cc,n.isObjectLike=sc,n.isPlainObject=mc,n.isRegExp=Ep,n.isSafeInteger=yc,n.isSet=Tp,n.isString=_c,n.isSymbol=bc,n.isTypedArray=Sp,n.isUndefined=xc,n.isWeakMap=wc,n.isWeakSet=Cc,n.join=Ma,n.kebabCase=Gp,n.last=ka,n.lastIndexOf=Ea,n.lowerCase=$p,n.lowerFirst=Xp,n.lt=Pp,n.lte=Np,n.max=Qs,n.maxBy=Js,n.mean=tl,n.meanBy=el,n.min=nl,n.minBy=rl,n.stubArray=Hs,n.stubFalse=qs,n.stubObject=Ys,n.stubString=Ks,n.stubTrue=Gs,n.multiply=gh,n.nth=Ta,n.noConflict=js,n.noop=Bs,n.now=sp,n.pad=hs,n.padEnd=ds,n.padStart=vs,n.parseInt=gs,n.random=us,n.reduce=bu,n.reduceRight=xu,n.repeat=ms,n.replace=ys,n.result=Xc,n.round=mh,n.runInContext=t,n.sample=Cu,n.size=Eu,n.snakeCase=Zp,n.some=Tu,n.sortedIndex=Da,n.sortedIndexBy=Ra,n.sortedIndexOf=La,n.sortedLastIndex=Ua,n.sortedLastIndexBy=Fa,n.sortedLastIndexOf=ja,n.startCase=Qp,n.startsWith=bs,n.subtract=yh,n.sum=il,n.sumBy=ol,n.template=xs,n.times=$s,n.toFinite=kc,n.toInteger=Ec,n.toLength=Tc,n.toLower=ws,n.toNumber=Sc,n.toSafeInteger=Nc,n.toString=Ac,n.toUpper=Cs,n.trim=Ms,n.trimEnd=ks,n.trimStart=Es,n.truncate=Ts,n.unescape=Ss,n.uniqueId=Zs,n.upperCase=Jp,n.upperFirst=th,n.each=vu,n.eachRight=gu,n.first=xa,Fs(n,function(){var t={};return nr(n,function(e,r){bl.call(n.prototype,r)||(t[r]=e)}),t}(),{chain:!1}),n.VERSION=ot,s([\"bind\",\"bindKey\",\"curry\",\"curryRight\",\"partial\",\"partialRight\"],function(t){n[t].placeholder=n}),s([\"drop\",\"take\"],function(t,e){b.prototype[t]=function(n){n=n===it?1:$l(Ec(n),0);var r=this.__filtered__&&!e?new b(this):this.clone();return r.__filtered__?r.__takeCount__=Xl(n,r.__takeCount__):r.__views__.push({size:Xl(n,Ft),type:t+(r.__dir__<0?\"Right\":\"\")}),r},b.prototype[t+\"Right\"]=function(e){return this.reverse()[t](e).reverse()}}),s([\"filter\",\"map\",\"takeWhile\"],function(t,e){var n=e+1,r=n==At||n==It;b.prototype[t]=function(t){var e=this.clone();return e.__iteratees__.push({iteratee:ko(t,3),type:n}),e.__filtered__=e.__filtered__||r,e}}),s([\"head\",\"last\"],function(t,e){var n=\"take\"+(e?\"Right\":\"\");b.prototype[t]=function(){return this[n](1).value()[0]}}),s([\"initial\",\"tail\"],function(t,e){var n=\"drop\"+(e?\"\":\"Right\");b.prototype[t]=function(){return this.__filtered__?new b(this):this[n](1)}}),b.prototype.compact=function(){return this.filter(Ds)},b.prototype.find=function(t){return this.filter(t).head()},b.prototype.findLast=function(t){return this.reverse().find(t)},b.prototype.invokeMap=oi(function(t,e){return\"function\"==typeof t?new b(this):this.map(function(n){return Sr(n,t,e)})}),b.prototype.reject=function(t){return this.filter(Lu(ko(t)))},b.prototype.slice=function(t,e){t=Ec(t);var n=this;return n.__filtered__&&(t>0||e<0)?new b(n):(t<0?n=n.takeRight(-t):t&&(n=n.drop(t)),e!==it&&(e=Ec(e),n=e<0?n.dropRight(-e):n.take(e-t)),n)},b.prototype.takeRightWhile=function(t){return this.reverse().takeWhile(t).reverse()},b.prototype.toArray=function(){return this.take(Ft)},nr(b.prototype,function(t,e){var r=/^(?:filter|find|map|reject)|While$/.test(e),o=/^(?:head|last)$/.test(e),a=n[o?\"take\"+(\"last\"==e?\"Right\":\"\"):e],u=o||/^find/.test(e);a&&(n.prototype[e]=function(){var e=this.__wrapped__,c=o?[1]:arguments,s=e instanceof b,l=c[0],f=s||xp(e),p=function(t){var e=a.apply(n,g([t],c));return o&&h?e[0]:e};f&&r&&\"function\"==typeof l&&1!=l.length&&(s=f=!1);var h=this.__chain__,d=!!this.__actions__.length,v=u&&!h,m=s&&!d;if(!u&&f){e=m?e:new b(this);var y=t.apply(e,c);return y.__actions__.push({func:nu,args:[p],thisArg:it}),new i(y,h)}return v&&m?t.apply(this,c):(y=this.thru(p),v?o?y.value()[0]:y.value():y)})}),s([\"pop\",\"push\",\"shift\",\"sort\",\"splice\",\"unshift\"],function(t){var e=vl[t],r=/^(?:push|sort|unshift)$/.test(t)?\"tap\":\"thru\",i=/^(?:pop|shift)$/.test(t);n.prototype[t]=function(){var t=arguments;if(i&&!this.__chain__){var n=this.value();return e.apply(xp(n)?n:[],t)}return this[r](function(n){return e.apply(xp(n)?n:[],t)})}}),nr(b.prototype,function(t,e){var r=n[e];if(r){var i=r.name+\"\",o=sf[i]||(sf[i]=[]);o.push({name:e,func:r})}}),sf[eo(it,yt).name]=[{name:\"wrapper\",func:it}],b.prototype.clone=P,b.prototype.reverse=Z,b.prototype.value=et,n.prototype.at=tp,n.prototype.chain=ru,n.prototype.commit=iu,n.prototype.next=ou,n.prototype.plant=uu,n.prototype.reverse=cu,n.prototype.toJSON=n.prototype.valueOf=n.prototype.value=su,n.prototype.first=n.prototype.head,Ll&&(n.prototype[Ll]=au),n},Mr=Cr();ur._=Mr,i=function(){return Mr}.call(e,n,e,r),!(i!==it&&(r.exports=i))}).call(this)}).call(e,n(99),n(100)(t))},function(t,e,n){\"use strict\";var r={remove:function(t){t._reactInternalInstance=void 0},get:function(t){return t._reactInternalInstance},has:function(t){return void 0!==t._reactInternalInstance},set:function(t,e){t._reactInternalInstance=e}};t.exports=r},function(t,e,n){\"use strict\";t.exports=n(26)},function(t,e,n){\"use strict\";var r=n(61);e.a=function(t){return t=n.i(r.a)(Math.abs(t)),t?t[1]:NaN}},function(t,e,n){\"use strict\";e.a=function(t,e){return t=+t,e-=t,function(n){return t+e*n}}},function(t,e,n){\"use strict\";var r=n(228);n.d(e,\"a\",function(){return r.a})},function(t,e,n){\"use strict\";function r(t,e){return(e-=t=+t)?function(n){return(n-t)/e}:n.i(h.a)(e)}function i(t){return function(e,n){var r=t(e=+e,n=+n);return function(t){return t<=e?0:t>=n?1:r(t)}}}function o(t){return function(e,n){var r=t(e=+e,n=+n);return function(t){return t<=0?e:t>=1?n:r(t)}}}function a(t,e,n,r){var i=t[0],o=t[1],a=e[0],u=e[1];return o<i?(i=n(o,i),a=r(u,a)):(i=n(i,o),a=r(a,u)),function(t){return a(i(t))}}function u(t,e,r,i){var o=Math.min(t.length,e.length)-1,a=new Array(o),u=new Array(o),c=-1;for(t[o]<t[0]&&(t=t.slice().reverse(),e=e.slice().reverse());++c<o;)a[c]=r(t[c],t[c+1]),u[c]=i(e[c],e[c+1]);return function(e){var r=n.i(l.c)(t,e,1,o)-1;return u[r](a[r](e))}}function c(t,e){return e.domain(t.domain()).range(t.range()).interpolate(t.interpolate()).clamp(t.clamp())}function s(t,e){function n(){return s=Math.min(g.length,m.length)>2?u:a,l=h=null,c}function c(e){return(l||(l=s(g,m,_?i(t):t,y)))(+e)}var s,l,h,g=v,m=v,y=f.b,_=!1;return c.invert=function(t){return(h||(h=s(m,g,r,_?o(e):e)))(+t)},c.domain=function(t){return arguments.length?(g=p.a.call(t,d.a),n()):g.slice()},c.range=function(t){return arguments.length?(m=p.b.call(t),n()):m.slice()},c.rangeRound=function(t){return m=p.b.call(t),y=f.c,n()},c.clamp=function(t){return arguments.length?(_=!!t,n()):_},c.interpolate=function(t){return arguments.length?(y=t,n()):y},n()}var l=n(12),f=n(31),p=n(16),h=n(65),d=n(126);e.b=r,e.c=c,e.a=s;var v=[0,1]},function(t,e,n){\"use strict\";function r(t,e,n){t._context.bezierCurveTo((2*t._x0+t._x1)/3,(2*t._y0+t._y1)/3,(t._x0+2*t._x1)/3,(t._y0+2*t._y1)/3,(t._x0+4*t._x1+e)/6,(t._y0+4*t._y1+n)/6)}function i(t){this._context=t}e.c=r,e.b=i,i.prototype={\n",
              "areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._y0=this._y1=NaN,this._point=0},lineEnd:function(){switch(this._point){case 3:r(this,this._x1,this._y1);case 2:this._context.lineTo(this._x1,this._y1)}(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2;break;case 2:this._point=3,this._context.lineTo((5*this._x0+this._x1)/6,(5*this._y0+this._y1)/6);default:r(this,t,e)}this._x0=this._x1,this._x1=t,this._y0=this._y1,this._y1=e}},e.a=function(t){return new i(t)}},function(t,e,n){\"use strict\";function r(t,e,n){t._context.bezierCurveTo(t._x1+t._k*(t._x2-t._x0),t._y1+t._k*(t._y2-t._y0),t._x2+t._k*(t._x1-e),t._y2+t._k*(t._y1-n),t._x2,t._y2)}function i(t,e){this._context=t,this._k=(1-e)/6}e.c=r,e.b=i,i.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._x2=this._y0=this._y1=this._y2=NaN,this._point=0},lineEnd:function(){switch(this._point){case 2:this._context.lineTo(this._x2,this._y2);break;case 3:r(this,this._x1,this._y1)}(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2,this._x1=t,this._y1=e;break;case 2:this._point=3;default:r(this,t,e)}this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return new i(t,e)}return n.tension=function(e){return t(+e)},n}(0)},function(t,e,n){\"use strict\";function r(t){this._context=t}r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._point=0},lineEnd:function(){(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2;default:this._context.lineTo(t,e)}}},e.a=function(t){return new r(t)}},function(t,e,n){\"use strict\";e.a=function(){}},function(t,e,n){\"use strict\";function r(t){return\"topMouseUp\"===t||\"topTouchEnd\"===t||\"topTouchCancel\"===t}function i(t){return\"topMouseMove\"===t||\"topTouchMove\"===t}function o(t){return\"topMouseDown\"===t||\"topTouchStart\"===t}function a(t,e,n,r){var i=t.type||\"unknown-event\";t.currentTarget=m.getNodeFromInstance(r),e?v.invokeGuardedCallbackWithCatch(i,n,t):v.invokeGuardedCallback(i,n,t),t.currentTarget=null}function u(t,e){var n=t._dispatchListeners,r=t._dispatchInstances;if(Array.isArray(n))for(var i=0;i<n.length&&!t.isPropagationStopped();i++)a(t,e,n[i],r[i]);else n&&a(t,e,n,r);t._dispatchListeners=null,t._dispatchInstances=null}function c(t){var e=t._dispatchListeners,n=t._dispatchInstances;if(Array.isArray(e)){for(var r=0;r<e.length&&!t.isPropagationStopped();r++)if(e[r](t,n[r]))return n[r]}else if(e&&e(t,n))return n;return null}function s(t){var e=c(t);return t._dispatchInstances=null,t._dispatchListeners=null,e}function l(t){var e=t._dispatchListeners,n=t._dispatchInstances;Array.isArray(e)?d(\"103\"):void 0,t.currentTarget=e?m.getNodeFromInstance(n):null;var r=e?e(t):null;return t.currentTarget=null,t._dispatchListeners=null,t._dispatchInstances=null,r}function f(t){return!!t._dispatchListeners}var p,h,d=n(2),v=n(87),g=(n(0),n(1),{injectComponentTree:function(t){p=t},injectTreeTraversal:function(t){h=t}}),m={isEndish:r,isMoveish:i,isStartish:o,executeDirectDispatch:l,executeDispatchesInOrder:u,executeDispatchesInOrderStopAtTrue:s,hasDispatches:f,getInstanceFromNode:function(t){return p.getInstanceFromNode(t)},getNodeFromInstance:function(t){return p.getNodeFromInstance(t)},isAncestor:function(t,e){return h.isAncestor(t,e)},getLowestCommonAncestor:function(t,e){return h.getLowestCommonAncestor(t,e)},getParentInstance:function(t){return h.getParentInstance(t)},traverseTwoPhase:function(t,e,n){return h.traverseTwoPhase(t,e,n)},traverseEnterLeave:function(t,e,n,r,i){return h.traverseEnterLeave(t,e,n,r,i)},injection:g};t.exports=m},function(t,e,n){\"use strict\";function r(t){return Object.prototype.hasOwnProperty.call(t,v)||(t[v]=h++,f[t[v]]={}),f[t[v]]}var i,o=n(3),a=n(83),u=n(360),c=n(89),s=n(393),l=n(94),f={},p=!1,h=0,d={topAbort:\"abort\",topAnimationEnd:s(\"animationend\")||\"animationend\",topAnimationIteration:s(\"animationiteration\")||\"animationiteration\",topAnimationStart:s(\"animationstart\")||\"animationstart\",topBlur:\"blur\",topCanPlay:\"canplay\",topCanPlayThrough:\"canplaythrough\",topChange:\"change\",topClick:\"click\",topCompositionEnd:\"compositionend\",topCompositionStart:\"compositionstart\",topCompositionUpdate:\"compositionupdate\",topContextMenu:\"contextmenu\",topCopy:\"copy\",topCut:\"cut\",topDoubleClick:\"dblclick\",topDrag:\"drag\",topDragEnd:\"dragend\",topDragEnter:\"dragenter\",topDragExit:\"dragexit\",topDragLeave:\"dragleave\",topDragOver:\"dragover\",topDragStart:\"dragstart\",topDrop:\"drop\",topDurationChange:\"durationchange\",topEmptied:\"emptied\",topEncrypted:\"encrypted\",topEnded:\"ended\",topError:\"error\",topFocus:\"focus\",topInput:\"input\",topKeyDown:\"keydown\",topKeyPress:\"keypress\",topKeyUp:\"keyup\",topLoadedData:\"loadeddata\",topLoadedMetadata:\"loadedmetadata\",topLoadStart:\"loadstart\",topMouseDown:\"mousedown\",topMouseMove:\"mousemove\",topMouseOut:\"mouseout\",topMouseOver:\"mouseover\",topMouseUp:\"mouseup\",topPaste:\"paste\",topPause:\"pause\",topPlay:\"play\",topPlaying:\"playing\",topProgress:\"progress\",topRateChange:\"ratechange\",topScroll:\"scroll\",topSeeked:\"seeked\",topSeeking:\"seeking\",topSelectionChange:\"selectionchange\",topStalled:\"stalled\",topSuspend:\"suspend\",topTextInput:\"textInput\",topTimeUpdate:\"timeupdate\",topTouchCancel:\"touchcancel\",topTouchEnd:\"touchend\",topTouchMove:\"touchmove\",topTouchStart:\"touchstart\",topTransitionEnd:s(\"transitionend\")||\"transitionend\",topVolumeChange:\"volumechange\",topWaiting:\"waiting\",topWheel:\"wheel\"},v=\"_reactListenersID\"+String(Math.random()).slice(2),g=o({},u,{ReactEventListener:null,injection:{injectReactEventListener:function(t){t.setHandleTopLevel(g.handleTopLevel),g.ReactEventListener=t}},setEnabled:function(t){g.ReactEventListener&&g.ReactEventListener.setEnabled(t)},isEnabled:function(){return!(!g.ReactEventListener||!g.ReactEventListener.isEnabled())},listenTo:function(t,e){for(var n=e,i=r(n),o=a.registrationNameDependencies[t],u=0;u<o.length;u++){var c=o[u];i.hasOwnProperty(c)&&i[c]||(\"topWheel\"===c?l(\"wheel\")?g.ReactEventListener.trapBubbledEvent(\"topWheel\",\"wheel\",n):l(\"mousewheel\")?g.ReactEventListener.trapBubbledEvent(\"topWheel\",\"mousewheel\",n):g.ReactEventListener.trapBubbledEvent(\"topWheel\",\"DOMMouseScroll\",n):\"topScroll\"===c?l(\"scroll\",!0)?g.ReactEventListener.trapCapturedEvent(\"topScroll\",\"scroll\",n):g.ReactEventListener.trapBubbledEvent(\"topScroll\",\"scroll\",g.ReactEventListener.WINDOW_HANDLE):\"topFocus\"===c||\"topBlur\"===c?(l(\"focus\",!0)?(g.ReactEventListener.trapCapturedEvent(\"topFocus\",\"focus\",n),g.ReactEventListener.trapCapturedEvent(\"topBlur\",\"blur\",n)):l(\"focusin\")&&(g.ReactEventListener.trapBubbledEvent(\"topFocus\",\"focusin\",n),g.ReactEventListener.trapBubbledEvent(\"topBlur\",\"focusout\",n)),i.topBlur=!0,i.topFocus=!0):d.hasOwnProperty(c)&&g.ReactEventListener.trapBubbledEvent(c,d[c],n),i[c]=!0)}},trapBubbledEvent:function(t,e,n){return g.ReactEventListener.trapBubbledEvent(t,e,n)},trapCapturedEvent:function(t,e,n){return g.ReactEventListener.trapCapturedEvent(t,e,n)},supportsEventPageXY:function(){if(!document.createEvent)return!1;var t=document.createEvent(\"MouseEvent\");return null!=t&&\"pageX\"in t},ensureScrollValueMonitoring:function(){if(void 0===i&&(i=g.supportsEventPageXY()),!i&&!p){var t=c.refreshScrollValues;g.ReactEventListener.monitorScrollValue(t),p=!0}}});t.exports=g},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(25),o=n(89),a=n(92),u={screenX:null,screenY:null,clientX:null,clientY:null,ctrlKey:null,shiftKey:null,altKey:null,metaKey:null,getModifierState:a,button:function(t){var e=t.button;return\"which\"in t?e:2===e?2:4===e?1:0},buttons:null,relatedTarget:function(t){return t.relatedTarget||(t.fromElement===t.srcElement?t.toElement:t.fromElement)},pageX:function(t){return\"pageX\"in t?t.pageX:t.clientX+o.currentScrollLeft},pageY:function(t){return\"pageY\"in t?t.pageY:t.clientY+o.currentScrollTop}};i.augmentClass(r,u),t.exports=r},function(t,e,n){\"use strict\";var r=n(2),i=(n(0),{}),o={reinitializeTransaction:function(){this.transactionWrappers=this.getTransactionWrappers(),this.wrapperInitData?this.wrapperInitData.length=0:this.wrapperInitData=[],this._isInTransaction=!1},_isInTransaction:!1,getTransactionWrappers:null,isInTransaction:function(){return!!this._isInTransaction},perform:function(t,e,n,i,o,a,u,c){this.isInTransaction()?r(\"27\"):void 0;var s,l;try{this._isInTransaction=!0,s=!0,this.initializeAll(0),l=t.call(e,n,i,o,a,u,c),s=!1}finally{try{if(s)try{this.closeAll(0)}catch(t){}else this.closeAll(0)}finally{this._isInTransaction=!1}}return l},initializeAll:function(t){for(var e=this.transactionWrappers,n=t;n<e.length;n++){var r=e[n];try{this.wrapperInitData[n]=i,this.wrapperInitData[n]=r.initialize?r.initialize.call(this):null}finally{if(this.wrapperInitData[n]===i)try{this.initializeAll(n+1)}catch(t){}}}},closeAll:function(t){this.isInTransaction()?void 0:r(\"28\");for(var e=this.transactionWrappers,n=t;n<e.length;n++){var o,a=e[n],u=this.wrapperInitData[n];try{o=!0,u!==i&&a.close&&a.close.call(this,u),o=!1}finally{if(o)try{this.closeAll(n+1)}catch(t){}}}this.wrapperInitData.length=0}};t.exports=o},function(t,e,n){\"use strict\";function r(t){var e=\"\"+t,n=o.exec(e);if(!n)return e;var r,i=\"\",a=0,u=0;for(a=n.index;a<e.length;a++){switch(e.charCodeAt(a)){case 34:r=\"&quot;\";break;case 38:r=\"&amp;\";break;case 39:r=\"&#x27;\";break;case 60:r=\"&lt;\";break;case 62:r=\"&gt;\";break;default:continue}u!==a&&(i+=e.substring(u,a)),u=a+1,i+=r}return u!==a?i+e.substring(u,a):i}function i(t){return\"boolean\"==typeof t||\"number\"==typeof t?\"\"+t:r(t)}var o=/[\"'&<>]/;t.exports=i},function(t,e,n){\"use strict\";var r,i=n(6),o=n(82),a=/^[ \\r\\n\\t\\f]/,u=/<(!--|link|noscript|meta|script|style)[ \\r\\n\\t\\f\\/>]/,c=n(90),s=c(function(t,e){if(t.namespaceURI!==o.svg||\"innerHTML\"in t)t.innerHTML=e;else{r=r||document.createElement(\"div\"),r.innerHTML=\"<svg>\"+e+\"</svg>\";for(var n=r.firstChild;n.firstChild;)t.appendChild(n.firstChild)}});if(i.canUseDOM){var l=document.createElement(\"div\");l.innerHTML=\" \",\"\"===l.innerHTML&&(s=function(t,e){if(t.parentNode&&t.parentNode.replaceChild(t,t),a.test(e)||\"<\"===e[0]&&u.test(e)){t.innerHTML=String.fromCharCode(65279)+e;var n=t.firstChild;1===n.data.length?t.removeChild(n):n.deleteData(0,1)}else t.innerHTML=e}),l=null}t.exports=s},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0}),e.default={colors:{RdBu:[\"rgb(255, 13, 87)\",\"rgb(30, 136, 229)\"],GnPR:[\"rgb(24, 196, 93)\",\"rgb(124, 82, 255)\"],CyPU:[\"#0099C6\",\"#990099\"],PkYg:[\"#DD4477\",\"#66AA00\"],DrDb:[\"#B82E2E\",\"#316395\"],LpLb:[\"#994499\",\"#22AA99\"],YlDp:[\"#AAAA11\",\"#6633CC\"],OrId:[\"#E67300\",\"#3E0099\"]},gray:\"#777\"}},function(t,e,n){\"use strict\";var r=n(29);e.a=function(t,e,n){if(null==n&&(n=r.a),i=t.length){if((e=+e)<=0||i<2)return+n(t[0],0,t);if(e>=1)return+n(t[i-1],i-1,t);var i,o=(i-1)*e,a=Math.floor(o),u=+n(t[a],a,t),c=+n(t[a+1],a+1,t);return u+(c-u)*(o-a)}}},function(t,e,n){\"use strict\";function r(){}function i(t,e){var n=new r;if(t instanceof r)t.each(function(t,e){n.set(e,t)});else if(Array.isArray(t)){var i,o=-1,a=t.length;if(null==e)for(;++o<a;)n.set(o,t[o]);else for(;++o<a;)n.set(e(i=t[o],o,t),i)}else if(t)for(var u in t)n.set(u,t[u]);return n}n.d(e,\"b\",function(){return o});var o=\"$\";r.prototype=i.prototype={constructor:r,has:function(t){return o+t in this},get:function(t){return this[o+t]},set:function(t,e){return this[o+t]=e,this},remove:function(t){var e=o+t;return e in this&&delete this[e]},clear:function(){for(var t in this)t[0]===o&&delete this[t]},keys:function(){var t=[];for(var e in this)e[0]===o&&t.push(e.slice(1));return t},values:function(){var t=[];for(var e in this)e[0]===o&&t.push(this[e]);return t},entries:function(){var t=[];for(var e in this)e[0]===o&&t.push({key:e.slice(1),value:this[e]});return t},size:function(){var t=0;for(var e in this)e[0]===o&&++t;return t},empty:function(){for(var t in this)if(t[0]===o)return!1;return!0},each:function(t){for(var e in this)e[0]===o&&t(this[e],e.slice(1),this)}},e.a=i},function(t,e,n){\"use strict\";function r(){}function i(t){var e;return t=(t+\"\").trim().toLowerCase(),(e=x.exec(t))?(e=parseInt(e[1],16),new s(e>>8&15|e>>4&240,e>>4&15|240&e,(15&e)<<4|15&e,1)):(e=w.exec(t))?o(parseInt(e[1],16)):(e=C.exec(t))?new s(e[1],e[2],e[3],1):(e=M.exec(t))?new s(255*e[1]/100,255*e[2]/100,255*e[3]/100,1):(e=k.exec(t))?a(e[1],e[2],e[3],e[4]):(e=E.exec(t))?a(255*e[1]/100,255*e[2]/100,255*e[3]/100,e[4]):(e=T.exec(t))?l(e[1],e[2]/100,e[3]/100,1):(e=S.exec(t))?l(e[1],e[2]/100,e[3]/100,e[4]):P.hasOwnProperty(t)?o(P[t]):\"transparent\"===t?new s(NaN,NaN,NaN,0):null}function o(t){return new s(t>>16&255,t>>8&255,255&t,1)}function a(t,e,n,r){return r<=0&&(t=e=n=NaN),new s(t,e,n,r)}function u(t){return t instanceof r||(t=i(t)),t?(t=t.rgb(),new s(t.r,t.g,t.b,t.opacity)):new s}function c(t,e,n,r){return 1===arguments.length?u(t):new s(t,e,n,null==r?1:r)}function s(t,e,n,r){this.r=+t,this.g=+e,this.b=+n,this.opacity=+r}function l(t,e,n,r){return r<=0?t=e=n=NaN:n<=0||n>=1?t=e=NaN:e<=0&&(t=NaN),new h(t,e,n,r)}function f(t){if(t instanceof h)return new h(t.h,t.s,t.l,t.opacity);if(t instanceof r||(t=i(t)),!t)return new h;if(t instanceof h)return t;t=t.rgb();var e=t.r/255,n=t.g/255,o=t.b/255,a=Math.min(e,n,o),u=Math.max(e,n,o),c=NaN,s=u-a,l=(u+a)/2;return s?(c=e===u?(n-o)/s+6*(n<o):n===u?(o-e)/s+2:(e-n)/s+4,s/=l<.5?u+a:2-u-a,c*=60):s=l>0&&l<1?0:c,new h(c,s,l,t.opacity)}function p(t,e,n,r){return 1===arguments.length?f(t):new h(t,e,n,null==r?1:r)}function h(t,e,n,r){this.h=+t,this.s=+e,this.l=+n,this.opacity=+r}function d(t,e,n){return 255*(t<60?e+(n-e)*t/60:t<180?n:t<240?e+(n-e)*(240-t)/60:e)}var v=n(60);e.f=r,n.d(e,\"h\",function(){return g}),n.d(e,\"g\",function(){return m}),e.a=i,e.e=u,e.b=c,e.d=s,e.c=p;var g=.7,m=1/g,y=\"\\\\s*([+-]?\\\\d+)\\\\s*\",_=\"\\\\s*([+-]?\\\\d*\\\\.?\\\\d+(?:[eE][+-]?\\\\d+)?)\\\\s*\",b=\"\\\\s*([+-]?\\\\d*\\\\.?\\\\d+(?:[eE][+-]?\\\\d+)?)%\\\\s*\",x=/^#([0-9a-f]{3})$/,w=/^#([0-9a-f]{6})$/,C=new RegExp(\"^rgb\\\\(\"+[y,y,y]+\"\\\\)$\"),M=new RegExp(\"^rgb\\\\(\"+[b,b,b]+\"\\\\)$\"),k=new RegExp(\"^rgba\\\\(\"+[y,y,y,_]+\"\\\\)$\"),E=new RegExp(\"^rgba\\\\(\"+[b,b,b,_]+\"\\\\)$\"),T=new RegExp(\"^hsl\\\\(\"+[_,b,b]+\"\\\\)$\"),S=new RegExp(\"^hsla\\\\(\"+[_,b,b,_]+\"\\\\)$\"),P={aliceblue:15792383,antiquewhite:16444375,aqua:65535,aquamarine:8388564,azure:15794175,beige:16119260,bisque:16770244,black:0,blanchedalmond:16772045,blue:255,blueviolet:9055202,brown:10824234,burlywood:14596231,cadetblue:6266528,chartreuse:8388352,chocolate:13789470,coral:16744272,cornflowerblue:6591981,cornsilk:16775388,crimson:14423100,cyan:65535,darkblue:139,darkcyan:35723,darkgoldenrod:12092939,darkgray:11119017,darkgreen:25600,darkgrey:11119017,darkkhaki:12433259,darkmagenta:9109643,darkolivegreen:5597999,darkorange:16747520,darkorchid:10040012,darkred:9109504,darksalmon:15308410,darkseagreen:9419919,darkslateblue:4734347,darkslategray:3100495,darkslategrey:3100495,darkturquoise:52945,darkviolet:9699539,deeppink:16716947,deepskyblue:49151,dimgray:6908265,dimgrey:6908265,dodgerblue:2003199,firebrick:11674146,floralwhite:16775920,forestgreen:2263842,fuchsia:16711935,gainsboro:14474460,ghostwhite:16316671,gold:16766720,goldenrod:14329120,gray:8421504,green:32768,greenyellow:11403055,grey:8421504,honeydew:15794160,hotpink:16738740,indianred:13458524,indigo:4915330,ivory:16777200,khaki:15787660,lavender:15132410,lavenderblush:16773365,lawngreen:8190976,lemonchiffon:16775885,lightblue:11393254,lightcoral:15761536,lightcyan:14745599,lightgoldenrodyellow:16448210,lightgray:13882323,lightgreen:9498256,lightgrey:13882323,lightpink:16758465,lightsalmon:16752762,lightseagreen:2142890,lightskyblue:8900346,lightslategray:7833753,lightslategrey:7833753,lightsteelblue:11584734,lightyellow:16777184,lime:65280,limegreen:3329330,linen:16445670,magenta:16711935,maroon:8388608,mediumaquamarine:6737322,mediumblue:205,mediumorchid:12211667,mediumpurple:9662683,mediumseagreen:3978097,mediumslateblue:8087790,mediumspringgreen:64154,mediumturquoise:4772300,mediumvioletred:13047173,midnightblue:1644912,mintcream:16121850,mistyrose:16770273,moccasin:16770229,navajowhite:16768685,navy:128,oldlace:16643558,olive:8421376,olivedrab:7048739,orange:16753920,orangered:16729344,orchid:14315734,palegoldenrod:15657130,palegreen:10025880,paleturquoise:11529966,palevioletred:14381203,papayawhip:16773077,peachpuff:16767673,peru:13468991,pink:16761035,plum:14524637,powderblue:11591910,purple:8388736,rebeccapurple:6697881,red:16711680,rosybrown:12357519,royalblue:4286945,saddlebrown:9127187,salmon:16416882,sandybrown:16032864,seagreen:3050327,seashell:16774638,sienna:10506797,silver:12632256,skyblue:8900331,slateblue:6970061,slategray:7372944,slategrey:7372944,snow:16775930,springgreen:65407,steelblue:4620980,tan:13808780,teal:32896,thistle:14204888,tomato:16737095,turquoise:4251856,violet:15631086,wheat:16113331,white:16777215,whitesmoke:16119285,yellow:16776960,yellowgreen:10145074};n.i(v.a)(r,i,{displayable:function(){return this.rgb().displayable()},toString:function(){return this.rgb()+\"\"}}),n.i(v.a)(s,c,n.i(v.b)(r,{brighter:function(t){return t=null==t?m:Math.pow(m,t),new s(this.r*t,this.g*t,this.b*t,this.opacity)},darker:function(t){return t=null==t?g:Math.pow(g,t),new s(this.r*t,this.g*t,this.b*t,this.opacity)},rgb:function(){return this},displayable:function(){return 0<=this.r&&this.r<=255&&0<=this.g&&this.g<=255&&0<=this.b&&this.b<=255&&0<=this.opacity&&this.opacity<=1},toString:function(){var t=this.opacity;return t=isNaN(t)?1:Math.max(0,Math.min(1,t)),(1===t?\"rgb(\":\"rgba(\")+Math.max(0,Math.min(255,Math.round(this.r)||0))+\", \"+Math.max(0,Math.min(255,Math.round(this.g)||0))+\", \"+Math.max(0,Math.min(255,Math.round(this.b)||0))+(1===t?\")\":\", \"+t+\")\")}})),n.i(v.a)(h,p,n.i(v.b)(r,{brighter:function(t){return t=null==t?m:Math.pow(m,t),new h(this.h,this.s,this.l*t,this.opacity)},darker:function(t){return t=null==t?g:Math.pow(g,t),new h(this.h,this.s,this.l*t,this.opacity)},rgb:function(){var t=this.h%360+360*(this.h<0),e=isNaN(t)||isNaN(this.s)?0:this.s,n=this.l,r=n+(n<.5?n:1-n)*e,i=2*n-r;return new s(d(t>=240?t-240:t+120,i,r),d(t,i,r),d(t<120?t+240:t-120,i,r),this.opacity)},displayable:function(){return(0<=this.s&&this.s<=1||isNaN(this.s))&&0<=this.l&&this.l<=1&&0<=this.opacity&&this.opacity<=1}}))},function(t,e,n){\"use strict\";function r(t,e){var n=Object.create(t.prototype);for(var r in e)n[r]=e[r];return n}e.b=r,e.a=function(t,e,n){t.prototype=e.prototype=n,n.constructor=t}},function(t,e,n){\"use strict\";e.a=function(t,e){if((n=(t=e?t.toExponential(e-1):t.toExponential()).indexOf(\"e\"))<0)return null;var n,r=t.slice(0,n);return[r.length>1?r[0]+r.slice(2):r,+t.slice(n+1)]}},function(t,e,n){\"use strict\";function r(t,e,n,r,i){var o=t*t,a=o*t;return((1-3*t+3*o-a)*e+(4-6*o+3*a)*n+(1+3*t+3*o-3*a)*r+a*i)/6}e.b=r,e.a=function(t){var e=t.length-1;return function(n){var i=n<=0?n=0:n>=1?(n=1,e-1):Math.floor(n*e),o=t[i],a=t[i+1],u=i>0?t[i-1]:2*o-a,c=i<e-1?t[i+2]:2*a-o;return r((n-i/e)*e,u,o,a,c)}}},function(t,e,n){\"use strict\";var r=n(10),i=n(123),o=n(118),a=n(121),u=n(43),c=n(122),s=n(124),l=n(120);e.a=function(t,e){var f,p=typeof e;return null==e||\"boolean\"===p?n.i(l.a)(e):(\"number\"===p?u.a:\"string\"===p?(f=n.i(r.color)(e))?(e=f,i.a):s.a:e instanceof r.color?i.a:e instanceof Date?a.a:Array.isArray(e)?o.a:isNaN(e)?c.a:u.a)(t,e)}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(229);n.d(e,\"scaleBand\",function(){return r.a}),n.d(e,\"scalePoint\",function(){return r.b});var i=n(235);n.d(e,\"scaleIdentity\",function(){return i.a});var o=n(34);n.d(e,\"scaleLinear\",function(){return o.a});var a=n(236);n.d(e,\"scaleLog\",function(){return a.a});var u=n(127);n.d(e,\"scaleOrdinal\",function(){return u.a}),n.d(e,\"scaleImplicit\",function(){return u.b});var c=n(237);n.d(e,\"scalePow\",function(){return c.a}),n.d(e,\"scaleSqrt\",function(){return c.b});var s=n(238);n.d(e,\"scaleQuantile\",function(){return s.a});var l=n(239);n.d(e,\"scaleQuantize\",function(){return l.a});var f=n(242);n.d(e,\"scaleThreshold\",function(){return f.a});var p=n(128);n.d(e,\"scaleTime\",function(){return p.a});var h=n(244);n.d(e,\"scaleUtc\",function(){return h.a});var d=n(230);n.d(e,\"schemeCategory10\",function(){return d.a});var v=n(232);n.d(e,\"schemeCategory20b\",function(){return v.a});var g=n(233);n.d(e,\"schemeCategory20c\",function(){return g.a});var m=n(231);n.d(e,\"schemeCategory20\",function(){return m.a});var y=n(234);n.d(e,\"interpolateCubehelixDefault\",function(){return y.a});var _=n(240);n.d(e,\"interpolateRainbow\",function(){return _.a}),n.d(e,\"interpolateWarm\",function(){return _.b}),n.d(e,\"interpolateCool\",function(){return _.c});var b=n(245);n.d(e,\"interpolateViridis\",function(){return b.a}),n.d(e,\"interpolateMagma\",function(){return b.b}),n.d(e,\"interpolateInferno\",function(){return b.c}),n.d(e,\"interpolatePlasma\",function(){return b.d});var x=n(241);n.d(e,\"scaleSequential\",function(){return x.a})},function(t,e,n){\"use strict\";e.a=function(t){return function(){return t}}},function(t,e,n){\"use strict\";function r(t){return function(){var e=this.ownerDocument,n=this.namespaceURI;return n===a.b&&e.documentElement.namespaceURI===a.b?e.createElement(t):e.createElementNS(n,t)}}function i(t){return function(){return this.ownerDocument.createElementNS(t.space,t.local)}}var o=n(67),a=n(68);e.a=function(t){var e=n.i(o.a)(t);return(e.local?i:r)(e)}},function(t,e,n){\"use strict\";var r=n(68);e.a=function(t){var e=t+=\"\",n=e.indexOf(\":\");return n>=0&&\"xmlns\"!==(e=t.slice(0,n))&&(t=t.slice(n+1)),r.a.hasOwnProperty(e)?{space:r.a[e],local:t}:t}},function(t,e,n){\"use strict\";n.d(e,\"b\",function(){return r});var r=\"http://www.w3.org/1999/xhtml\";e.a={svg:\"http://www.w3.org/2000/svg\",xhtml:r,xlink:\"http://www.w3.org/1999/xlink\",xml:\"http://www.w3.org/XML/1998/namespace\",xmlns:\"http://www.w3.org/2000/xmlns/\"}},function(t,e,n){\"use strict\";e.a=function(t,e){var n=t.ownerSVGElement||t;if(n.createSVGPoint){var r=n.createSVGPoint();return r.x=e.clientX,r.y=e.clientY,r=r.matrixTransform(t.getScreenCTM().inverse()),[r.x,r.y]}var i=t.getBoundingClientRect();return[e.clientX-i.left-t.clientLeft,e.clientY-i.top-t.clientTop]}},function(t,e,n){\"use strict\";function r(t,e,n){return t=i(t,e,n),function(e){var n=e.relatedTarget;n&&(n===this||8&n.compareDocumentPosition(this))||t.call(this,e)}}function i(t,e,n){return function(r){var i=l;l=r;try{t.call(this,this.__data__,e,n)}finally{l=i}}}function o(t){return t.trim().split(/^|\\s+/).map(function(t){var e=\"\",n=t.indexOf(\".\");return n>=0&&(e=t.slice(n+1),t=t.slice(0,n)),{type:t,name:e}})}function a(t){return function(){var e=this.__on;if(e){for(var n,r=0,i=-1,o=e.length;r<o;++r)n=e[r],t.type&&n.type!==t.type||n.name!==t.name?e[++i]=n:this.removeEventListener(n.type,n.listener,n.capture);++i?e.length=i:delete this.__on}}}function u(t,e,n){var o=s.hasOwnProperty(t.type)?r:i;return function(r,i,a){var u,c=this.__on,s=o(e,i,a);if(c)for(var l=0,f=c.length;l<f;++l)if((u=c[l]).type===t.type&&u.name===t.name)return this.removeEventListener(u.type,u.listener,u.capture),this.addEventListener(u.type,u.listener=s,u.capture=n),void(u.value=e);this.addEventListener(t.type,s,n),u={type:t.type,name:t.name,value:e,listener:s,capture:n},c?c.push(u):this.__on=[u]}}function c(t,e,n,r){var i=l;t.sourceEvent=l,l=t;try{return e.apply(n,r)}finally{l=i}}n.d(e,\"a\",function(){return l}),e.b=c;var s={},l=null;if(\"undefined\"!=typeof document){var f=document.documentElement;\"onmouseenter\"in f||(s={mouseenter:\"mouseover\",mouseleave:\"mouseout\"})}e.c=function(t,e,n){var r,i,c=o(t+\"\"),s=c.length;{if(!(arguments.length<2)){for(l=e?u:a,null==n&&(n=!1),r=0;r<s;++r)this.each(l(c[r],e,n));return this}var l=this.node().__on;if(l)for(var f,p=0,h=l.length;p<h;++p)for(r=0,f=l[p];r<s;++r)if((i=c[r]).type===f.type&&i.name===f.name)return f.value}}},function(t,e,n){\"use strict\";function r(){}e.a=function(t){return null==t?r:function(){return this.querySelector(t)}}},function(t,e,n){\"use strict\";var r=n(70);e.a=function(){for(var t,e=r.a;t=e.sourceEvent;)e=t;return e}},function(t,e,n){\"use strict\";e.a=function(t){return t.ownerDocument&&t.ownerDocument.defaultView||t.document&&t||t.defaultView}},function(t,e,n){\"use strict\";function r(t,e,n){var r=t._x1,i=t._y1,a=t._x2,u=t._y2;if(t._l01_a>o.a){var c=2*t._l01_2a+3*t._l01_a*t._l12_a+t._l12_2a,s=3*t._l01_a*(t._l01_a+t._l12_a);r=(r*c-t._x0*t._l12_2a+t._x2*t._l01_2a)/s,i=(i*c-t._y0*t._l12_2a+t._y2*t._l01_2a)/s}if(t._l23_a>o.a){var l=2*t._l23_2a+3*t._l23_a*t._l12_a+t._l12_2a,f=3*t._l23_a*(t._l23_a+t._l12_a);a=(a*l+t._x1*t._l23_2a-e*t._l12_2a)/f,u=(u*l+t._y1*t._l23_2a-n*t._l12_2a)/f}t._context.bezierCurveTo(r,i,a,u,t._x2,t._y2)}function i(t,e){this._context=t,this._alpha=e}var o=n(35),a=n(47);e.b=r,i.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._x2=this._y0=this._y1=this._y2=NaN,this._l01_a=this._l12_a=this._l23_a=this._l01_2a=this._l12_2a=this._l23_2a=this._point=0},lineEnd:function(){switch(this._point){case 2:this._context.lineTo(this._x2,this._y2);break;case 3:this.point(this._x2,this._y2)}(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){if(t=+t,e=+e,this._point){var n=this._x2-t,i=this._y2-e;this._l23_a=Math.sqrt(this._l23_2a=Math.pow(n*n+i*i,this._alpha))}switch(this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2;break;case 2:this._point=3;default:r(this,t,e)}this._l01_a=this._l12_a,this._l12_a=this._l23_a,this._l01_2a=this._l12_2a,this._l12_2a=this._l23_2a,this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return e?new i(t,e):new a.b(t,0)}return n.alpha=function(e){return t(+e)},n}(.5)},function(t,e,n){\"use strict\";var r=n(44),i=n(19),o=n(48),a=n(139);e.a=function(){function t(t){var i,o,a,p=t.length,h=!1;for(null==s&&(f=l(a=n.i(r.a)())),i=0;i<=p;++i)!(i<p&&c(o=t[i],i,t))===h&&((h=!h)?f.lineStart():f.lineEnd()),h&&f.point(+e(o,i,t),+u(o,i,t));if(a)return f=null,a+\"\"||null}var e=a.a,u=a.b,c=n.i(i.a)(!0),s=null,l=o.a,f=null;return t.x=function(r){return arguments.length?(e=\"function\"==typeof r?r:n.i(i.a)(+r),t):e},t.y=function(e){return arguments.length?(u=\"function\"==typeof e?e:n.i(i.a)(+e),t):u},t.defined=function(e){return arguments.length?(c=\"function\"==typeof e?e:n.i(i.a)(!!e),t):c},t.curve=function(e){return arguments.length?(l=e,null!=s&&(f=l(s)),t):l},t.context=function(e){return arguments.length?(null==e?s=f=null:f=l(s=e),t):s},t}},function(t,e,n){\"use strict\";function r(t){for(var e,n=0,r=-1,i=t.length;++r<i;)(e=+t[r][1])&&(n+=e);return n}var i=n(37);e.b=r,e.a=function(t){var e=t.map(r);return n.i(i.a)(t).sort(function(t,n){return e[t]-e[n]})}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(78);n.d(e,\"timeFormatDefaultLocale\",function(){return r.a}),n.d(e,\"timeFormat\",function(){return r.b}),n.d(e,\"timeParse\",function(){return r.c}),n.d(e,\"utcFormat\",function(){return r.d}),n.d(e,\"utcParse\",function(){return r.e});var i=n(149);n.d(e,\"timeFormatLocale\",function(){return i.a});var o=n(148);n.d(e,\"isoFormat\",function(){return o.a});var a=n(303);n.d(e,\"isoParse\",function(){return a.a})},function(t,e,n){\"use strict\";function r(t){return o=n.i(i.a)(t),a=o.format,u=o.parse,c=o.utcFormat,s=o.utcParse,o}var i=n(149);n.d(e,\"b\",function(){return a}),n.d(e,\"c\",function(){return u}),n.d(e,\"d\",function(){return c}),n.d(e,\"e\",function(){return s}),e.a=r;var o,a,u,c,s;r({dateTime:\"%x, %X\",date:\"%-m/%-d/%Y\",time:\"%-I:%M:%S %p\",periods:[\"AM\",\"PM\"],days:[\"Sunday\",\"Monday\",\"Tuesday\",\"Wednesday\",\"Thursday\",\"Friday\",\"Saturday\"],shortDays:[\"Sun\",\"Mon\",\"Tue\",\"Wed\",\"Thu\",\"Fri\",\"Sat\"],months:[\"January\",\"February\",\"March\",\"April\",\"May\",\"June\",\"July\",\"August\",\"September\",\"October\",\"November\",\"December\"],shortMonths:[\"Jan\",\"Feb\",\"Mar\",\"Apr\",\"May\",\"Jun\",\"Jul\",\"Aug\",\"Sep\",\"Oct\",\"Nov\",\"Dec\"]})},function(t,e,n){\"use strict\";var r=(n(5),n(306));n.d(e,\"t\",function(){return r.a}),n.d(e,\"n\",function(){return r.a});var i=n(309);n.d(e,\"s\",function(){return i.a}),n.d(e,\"m\",function(){return i.a});var o=n(307);n.d(e,\"r\",function(){return o.a});var a=n(305);n.d(e,\"q\",function(){return a.a});var u=n(304);n.d(e,\"a\",function(){return u.a});var c=n(316);n.d(e,\"p\",function(){return c.a}),n.d(e,\"c\",function(){return c.a}),n.d(e,\"d\",function(){return c.b});var s=n(308);n.d(e,\"o\",function(){return s.a});var l=n(317);n.d(e,\"b\",function(){return l.a});var f=n(312);n.d(e,\"l\",function(){return f.a});var p=n(311);n.d(e,\"k\",function(){return p.a});var h=n(310);n.d(e,\"e\",function(){return h.a});var d=n(314);n.d(e,\"j\",function(){return d.a}),n.d(e,\"g\",function(){return d.a}),n.d(e,\"h\",function(){return d.b});var v=n(313);n.d(e,\"i\",function(){return v.a});var g=n(315);n.d(e,\"f\",function(){return g.a})},function(t,e,n){\"use strict\";function r(t,e){return t===e?0!==t||0!==e||1/t===1/e:t!==t&&e!==e}function i(t,e){if(r(t,e))return!0;if(\"object\"!=typeof t||null===t||\"object\"!=typeof e||null===e)return!1;var n=Object.keys(t),i=Object.keys(e);if(n.length!==i.length)return!1;for(var a=0;a<n.length;a++)if(!o.call(e,n[a])||!r(t[n[a]],e[n[a]]))return!1;return!0}var o=Object.prototype.hasOwnProperty;t.exports=i},function(t,e,n){\"use strict\";function r(t,e){return Array.isArray(e)&&(e=e[1]),e?e.nextSibling:t.firstChild}function i(t,e,n){l.insertTreeBefore(t,e,n)}function o(t,e,n){Array.isArray(e)?u(t,e[0],e[1],n):v(t,e,n)}function a(t,e){if(Array.isArray(e)){var n=e[1];e=e[0],c(t,e,n),t.removeChild(n)}t.removeChild(e)}function u(t,e,n,r){for(var i=e;;){var o=i.nextSibling;if(v(t,i,r),i===n)break;i=o}}function c(t,e,n){for(;;){var r=e.nextSibling;if(r===n)break;t.removeChild(r)}}function s(t,e,n){var r=t.parentNode,i=t.nextSibling;i===e?n&&v(r,document.createTextNode(n),i):n?(d(i,n),c(r,i,e)):c(r,t,e)}var l=n(20),f=n(336),p=(n(4),n(9),n(90)),h=n(55),d=n(171),v=p(function(t,e,n){t.insertBefore(e,n)}),g=f.dangerouslyReplaceNodeWithMarkup,m={dangerouslyReplaceNodeWithMarkup:g,replaceDelimitedText:s,processUpdates:function(t,e){for(var n=0;n<e.length;n++){var u=e[n];switch(u.type){case\"INSERT_MARKUP\":i(t,u.content,r(t,u.afterNode));break;case\"MOVE_EXISTING\":o(t,u.fromNode,r(t,u.afterNode));break;case\"SET_MARKUP\":h(t,u.content);break;case\"TEXT_CONTENT\":d(t,u.content);break;case\"REMOVE_NODE\":a(t,u.fromNode)}}}};t.exports=m},function(t,e,n){\"use strict\";var r={html:\"http://www.w3.org/1999/xhtml\",mathml:\"http://www.w3.org/1998/Math/MathML\",svg:\"http://www.w3.org/2000/svg\"};t.exports=r},function(t,e,n){\"use strict\";function r(){if(u)for(var t in c){var e=c[t],n=u.indexOf(t);if(n>-1?void 0:a(\"96\",t),!s.plugins[n]){e.extractEvents?void 0:a(\"97\",t),s.plugins[n]=e;var r=e.eventTypes;for(var o in r)i(r[o],e,o)?void 0:a(\"98\",o,t)}}}function i(t,e,n){s.eventNameDispatchConfigs.hasOwnProperty(n)?a(\"99\",n):void 0,s.eventNameDispatchConfigs[n]=t;var r=t.phasedRegistrationNames;if(r){for(var i in r)if(r.hasOwnProperty(i)){var u=r[i];o(u,e,n)}return!0}return!!t.registrationName&&(o(t.registrationName,e,n),!0)}function o(t,e,n){s.registrationNameModules[t]?a(\"100\",t):void 0,s.registrationNameModules[t]=e,s.registrationNameDependencies[t]=e.eventTypes[n].dependencies}var a=n(2),u=(n(0),null),c={},s={plugins:[],eventNameDispatchConfigs:{},registrationNameModules:{},registrationNameDependencies:{},possibleRegistrationNames:null,injectEventPluginOrder:function(t){\n",
              "u?a(\"101\"):void 0,u=Array.prototype.slice.call(t),r()},injectEventPluginsByName:function(t){var e=!1;for(var n in t)if(t.hasOwnProperty(n)){var i=t[n];c.hasOwnProperty(n)&&c[n]===i||(c[n]?a(\"102\",n):void 0,c[n]=i,e=!0)}e&&r()},getPluginModuleForEvent:function(t){var e=t.dispatchConfig;if(e.registrationName)return s.registrationNameModules[e.registrationName]||null;if(void 0!==e.phasedRegistrationNames){var n=e.phasedRegistrationNames;for(var r in n)if(n.hasOwnProperty(r)){var i=s.registrationNameModules[n[r]];if(i)return i}}return null},_resetEventPlugins:function(){u=null;for(var t in c)c.hasOwnProperty(t)&&delete c[t];s.plugins.length=0;var e=s.eventNameDispatchConfigs;for(var n in e)e.hasOwnProperty(n)&&delete e[n];var r=s.registrationNameModules;for(var i in r)r.hasOwnProperty(i)&&delete r[i]}};t.exports=s},function(t,e,n){\"use strict\";function r(t){var e=/[=:]/g,n={\"=\":\"=0\",\":\":\"=2\"},r=(\"\"+t).replace(e,function(t){return n[t]});return\"$\"+r}function i(t){var e=/(=0|=2)/g,n={\"=0\":\"=\",\"=2\":\":\"},r=\".\"===t[0]&&\"$\"===t[1]?t.substring(2):t.substring(1);return(\"\"+r).replace(e,function(t){return n[t]})}var o={escape:r,unescape:i};t.exports=o},function(t,e,n){\"use strict\";function r(t){null!=t.checkedLink&&null!=t.valueLink?u(\"87\"):void 0}function i(t){r(t),null!=t.value||null!=t.onChange?u(\"88\"):void 0}function o(t){r(t),null!=t.checked||null!=t.onChange?u(\"89\"):void 0}function a(t){if(t){var e=t.getName();if(e)return\" Check the render method of `\"+e+\"`.\"}return\"\"}var u=n(2),c=n(26),s=n(366),l=(n(0),n(1),{button:!0,checkbox:!0,image:!0,hidden:!0,radio:!0,reset:!0,submit:!0}),f={value:function(t,e,n){return!t[e]||l[t.type]||t.onChange||t.readOnly||t.disabled?null:new Error(\"You provided a `value` prop to a form field without an `onChange` handler. This will render a read-only field. If the field should be mutable use `defaultValue`. Otherwise, set either `onChange` or `readOnly`.\")},checked:function(t,e,n){return!t[e]||t.onChange||t.readOnly||t.disabled?null:new Error(\"You provided a `checked` prop to a form field without an `onChange` handler. This will render a read-only field. If the field should be mutable use `defaultChecked`. Otherwise, set either `onChange` or `readOnly`.\")},onChange:c.PropTypes.func},p={},h={checkPropTypes:function(t,e,n){for(var r in f){if(f.hasOwnProperty(r))var i=f[r](e,r,t,\"prop\",null,s);if(i instanceof Error&&!(i.message in p)){p[i.message]=!0;a(n)}}},getValue:function(t){return t.valueLink?(i(t),t.valueLink.value):t.value},getChecked:function(t){return t.checkedLink?(o(t),t.checkedLink.value):t.checked},executeOnChange:function(t,e){return t.valueLink?(i(t),t.valueLink.requestChange(e.target.value)):t.checkedLink?(o(t),t.checkedLink.requestChange(e.target.checked)):t.onChange?t.onChange.call(void 0,e):void 0}};t.exports=h},function(t,e,n){\"use strict\";var r=n(2),i=(n(0),!1),o={replaceNodeWithMarkup:null,processChildrenUpdates:null,injection:{injectEnvironment:function(t){i?r(\"104\"):void 0,o.replaceNodeWithMarkup=t.replaceNodeWithMarkup,o.processChildrenUpdates=t.processChildrenUpdates,i=!0}}};t.exports=o},function(t,e,n){\"use strict\";function r(t,e,n){try{e(n)}catch(t){null===i&&(i=t)}}var i=null,o={invokeGuardedCallback:r,invokeGuardedCallbackWithCatch:r,rethrowCaughtError:function(){if(i){var t=i;throw i=null,t}}};t.exports=o},function(t,e,n){\"use strict\";function r(t){c.enqueueUpdate(t)}function i(t){var e=typeof t;if(\"object\"!==e)return e;var n=t.constructor&&t.constructor.name||e,r=Object.keys(t);return r.length>0&&r.length<20?n+\" (keys: \"+r.join(\", \")+\")\":n}function o(t,e){var n=u.get(t);if(!n){return null}return n}var a=n(2),u=(n(15),n(40)),c=(n(9),n(11)),s=(n(0),n(1),{isMounted:function(t){var e=u.get(t);return!!e&&!!e._renderedComponent},enqueueCallback:function(t,e,n){s.validateCallback(e,n);var i=o(t);return i?(i._pendingCallbacks?i._pendingCallbacks.push(e):i._pendingCallbacks=[e],void r(i)):null},enqueueCallbackInternal:function(t,e){t._pendingCallbacks?t._pendingCallbacks.push(e):t._pendingCallbacks=[e],r(t)},enqueueForceUpdate:function(t){var e=o(t,\"forceUpdate\");e&&(e._pendingForceUpdate=!0,r(e))},enqueueReplaceState:function(t,e){var n=o(t,\"replaceState\");n&&(n._pendingStateQueue=[e],n._pendingReplaceState=!0,r(n))},enqueueSetState:function(t,e){var n=o(t,\"setState\");if(n){var i=n._pendingStateQueue||(n._pendingStateQueue=[]);i.push(e),r(n)}},enqueueElementInternal:function(t,e,n){t._pendingElement=e,t._context=n,r(t)},validateCallback:function(t,e){t&&\"function\"!=typeof t?a(\"122\",e,i(t)):void 0}});t.exports=s},function(t,e,n){\"use strict\";var r={currentScrollLeft:0,currentScrollTop:0,refreshScrollValues:function(t){r.currentScrollLeft=t.x,r.currentScrollTop=t.y}};t.exports=r},function(t,e,n){\"use strict\";var r=function(t){return\"undefined\"!=typeof MSApp&&MSApp.execUnsafeLocalFunction?function(e,n,r,i){MSApp.execUnsafeLocalFunction(function(){return t(e,n,r,i)})}:t};t.exports=r},function(t,e,n){\"use strict\";function r(t){var e,n=t.keyCode;return\"charCode\"in t?(e=t.charCode,0===e&&13===n&&(e=13)):e=n,e>=32||13===e?e:0}t.exports=r},function(t,e,n){\"use strict\";function r(t){var e=this,n=e.nativeEvent;if(n.getModifierState)return n.getModifierState(t);var r=o[t];return!!r&&!!n[r]}function i(t){return r}var o={Alt:\"altKey\",Control:\"ctrlKey\",Meta:\"metaKey\",Shift:\"shiftKey\"};t.exports=i},function(t,e,n){\"use strict\";function r(t){var e=t.target||t.srcElement||window;return e.correspondingUseElement&&(e=e.correspondingUseElement),3===e.nodeType?e.parentNode:e}t.exports=r},function(t,e,n){\"use strict\";/**\n",
              " * Checks if an event is supported in the current execution environment.\n",
              " *\n",
              " * NOTE: This will not work correctly for non-generic events such as `change`,\n",
              " * `reset`, `load`, `error`, and `select`.\n",
              " *\n",
              " * Borrows from Modernizr.\n",
              " *\n",
              " * @param {string} eventNameSuffix Event name, e.g. \"click\".\n",
              " * @param {?boolean} capture Check if the capture phase is supported.\n",
              " * @return {boolean} True if the event is supported.\n",
              " * @internal\n",
              " * @license Modernizr 3.0.0pre (Custom Build) | MIT\n",
              " */\n",
              "function r(t,e){if(!o.canUseDOM||e&&!(\"addEventListener\"in document))return!1;var n=\"on\"+t,r=n in document;if(!r){var a=document.createElement(\"div\");a.setAttribute(n,\"return;\"),r=\"function\"==typeof a[n]}return!r&&i&&\"wheel\"===t&&(r=document.implementation.hasFeature(\"Events.wheel\",\"3.0\")),r}var i,o=n(6);o.canUseDOM&&(i=document.implementation&&document.implementation.hasFeature&&document.implementation.hasFeature(\"\",\"\")!==!0),t.exports=r},function(t,e,n){\"use strict\";function r(t,e){var n=null===t||t===!1,r=null===e||e===!1;if(n||r)return n===r;var i=typeof t,o=typeof e;return\"string\"===i||\"number\"===i?\"string\"===o||\"number\"===o:\"object\"===o&&t.type===e.type&&t.key===e.key}t.exports=r},function(t,e,n){\"use strict\";var r=(n(3),n(8)),i=(n(1),r);t.exports=i},function(t,e,n){\"use strict\";function r(t,e,n){this.props=t,this.context=e,this.refs=a,this.updater=n||o}var i=n(28),o=n(98),a=(n(176),n(38));n(0),n(1);r.prototype.isReactComponent={},r.prototype.setState=function(t,e){\"object\"!=typeof t&&\"function\"!=typeof t&&null!=t?i(\"85\"):void 0,this.updater.enqueueSetState(this,t),e&&this.updater.enqueueCallback(this,e,\"setState\")},r.prototype.forceUpdate=function(t){this.updater.enqueueForceUpdate(this),t&&this.updater.enqueueCallback(this,t,\"forceUpdate\")};t.exports=r},function(t,e,n){\"use strict\";function r(t,e){}var i=(n(1),{isMounted:function(t){return!1},enqueueCallback:function(t,e){},enqueueForceUpdate:function(t){r(t,\"forceUpdate\")},enqueueReplaceState:function(t,e){r(t,\"replaceState\")},enqueueSetState:function(t,e){r(t,\"setState\")}});t.exports=i},function(t,e){var n;n=function(){return this}();try{n=n||Function(\"return this\")()||(0,eval)(\"this\")}catch(t){\"object\"==typeof window&&(n=window)}t.exports=n},function(t,e){t.exports=function(t){return t.webpackPolyfill||(t.deprecate=function(){},t.paths=[],t.children||(t.children=[]),Object.defineProperty(t,\"loaded\",{enumerable:!0,get:function(){return t.l}}),Object.defineProperty(t,\"id\",{enumerable:!0,get:function(){return t.i}}),t.webpackPolyfill=1),t}},function(t,e,n){\"use strict\";n.d(e,\"b\",function(){return i}),n.d(e,\"a\",function(){return o});var r=Array.prototype,i=r.slice,o=r.map},function(t,e,n){\"use strict\";var r=n(18),i=n(103),o=n.i(i.a)(r.a),a=o.right;o.left;e.a=a},function(t,e,n){\"use strict\";function r(t){return function(e,r){return n.i(i.a)(t(e),r)}}var i=n(18);e.a=function(t){return 1===t.length&&(t=r(t)),{left:function(e,n,r,i){for(null==r&&(r=0),null==i&&(i=e.length);r<i;){var o=r+i>>>1;t(e[o],n)<0?r=o+1:i=o}return r},right:function(e,n,r,i){for(null==r&&(r=0),null==i&&(i=e.length);r<i;){var o=r+i>>>1;t(e[o],n)>0?i=o:r=o+1}return r}}}},function(t,e,n){\"use strict\";var r=n(111);e.a=function(t,e){var i=n.i(r.a)(t,e);return i?Math.sqrt(i):i}},function(t,e,n){\"use strict\";e.a=function(t,e){var n,r,i,o=-1,a=t.length;if(null==e){for(;++o<a;)if(null!=(r=t[o])&&r>=r){n=i=r;break}for(;++o<a;)null!=(r=t[o])&&(n>r&&(n=r),i<r&&(i=r))}else{for(;++o<a;)if(null!=(r=e(t[o],o,t))&&r>=r){n=i=r;break}for(;++o<a;)null!=(r=e(t[o],o,t))&&(n>r&&(n=r),i<r&&(i=r))}return[n,i]}},function(t,e,n){\"use strict\";e.a=function(t,e){var n,r,i=-1,o=t.length;if(null==e){for(;++i<o;)if(null!=(r=t[i])&&r>=r){n=r;break}for(;++i<o;)null!=(r=t[i])&&n>r&&(n=r)}else{for(;++i<o;)if(null!=(r=e(t[i],i,t))&&r>=r){n=r;break}for(;++i<o;)null!=(r=e(t[i],i,t))&&n>r&&(n=r)}return n}},function(t,e,n){\"use strict\";e.a=function(t,e,n){t=+t,e=+e,n=(i=arguments.length)<2?(e=t,t=0,1):i<3?1:+n;for(var r=-1,i=0|Math.max(0,Math.ceil((e-t)/n)),o=new Array(i);++r<i;)o[r]=t+r*n;return o}},function(t,e,n){\"use strict\";e.a=function(t){return Math.ceil(Math.log(t.length)/Math.LN2)+1}},function(t,e,n){\"use strict\";function r(t,e,n){var r=Math.abs(e-t)/Math.max(0,n),i=Math.pow(10,Math.floor(Math.log(r)/Math.LN10)),c=r/i;return c>=o?i*=10:c>=a?i*=5:c>=u&&(i*=2),e<t?-i:i}var i=n(107);e.b=r;var o=Math.sqrt(50),a=Math.sqrt(10),u=Math.sqrt(2);e.a=function(t,e,o){var a=r(t,e,o);return n.i(i.a)(Math.ceil(t/a)*a,Math.floor(e/a)*a+a/2,a)}},function(t,e,n){\"use strict\";function r(t){return t.length}var i=n(106);e.a=function(t){if(!(u=t.length))return[];for(var e=-1,o=n.i(i.a)(t,r),a=new Array(o);++e<o;)for(var u,c=-1,s=a[e]=new Array(u);++c<u;)s[c]=t[c][e];return a}},function(t,e,n){\"use strict\";var r=n(29);e.a=function(t,e){var i,o,a=t.length,u=0,c=0,s=-1,l=0;if(null==e)for(;++s<a;)isNaN(i=n.i(r.a)(t[s]))||(o=i-u,u+=o/++l,c+=o*(i-u));else for(;++s<a;)isNaN(i=n.i(r.a)(e(t[s],s,t)))||(o=i-u,u+=o/++l,c+=o*(i-u));if(l>1)return c/(l-1)}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(201);n.d(e,\"axisTop\",function(){return r.a}),n.d(e,\"axisRight\",function(){return r.b}),n.d(e,\"axisBottom\",function(){return r.c}),n.d(e,\"axisLeft\",function(){return r.d})},function(t,e,n){\"use strict\";n.d(e,\"b\",function(){return r}),n.d(e,\"a\",function(){return i});var r=Math.PI/180,i=180/Math.PI},function(t,e,n){\"use strict\";var r=n(61);n.d(e,\"b\",function(){return i});var i;e.a=function(t,e){var o=n.i(r.a)(t,e);if(!o)return t+\"\";var a=o[0],u=o[1],c=u-(i=3*Math.max(-8,Math.min(8,Math.floor(u/3))))+1,s=a.length;return c===s?a:c>s?a+new Array(c-s+1).join(\"0\"):c>0?a.slice(0,c)+\".\"+a.slice(c):\"0.\"+new Array(1-c).join(\"0\")+n.i(r.a)(t,Math.max(0,e+c-1))[0]}},function(t,e,n){\"use strict\";function r(t){if(!(e=o.exec(t)))throw new Error(\"invalid format: \"+t);var e,n=e[1]||\" \",r=e[2]||\">\",a=e[3]||\"-\",u=e[4]||\"\",c=!!e[5],s=e[6]&&+e[6],l=!!e[7],f=e[8]&&+e[8].slice(1),p=e[9]||\"\";\"n\"===p?(l=!0,p=\"g\"):i.a[p]||(p=\"\"),(c||\"0\"===n&&\"=\"===r)&&(c=!0,n=\"0\",r=\"=\"),this.fill=n,this.align=r,this.sign=a,this.symbol=u,this.zero=c,this.width=s,this.comma=l,this.precision=f,this.type=p}var i=n(116),o=/^(?:(.)?([<>=^]))?([+\\-\\( ])?([$#])?(0)?(\\d+)?(,)?(\\.\\d+)?([a-z%])?$/i;e.a=function(t){return new r(t)},r.prototype.toString=function(){return this.fill+this.align+this.sign+this.symbol+(this.zero?\"0\":\"\")+(null==this.width?\"\":Math.max(1,0|this.width))+(this.comma?\",\":\"\")+(null==this.precision?\"\":\".\"+Math.max(0,0|this.precision))+this.type}},function(t,e,n){\"use strict\";var r=n(212),i=n(114),o=n(214);e.a={\"\":r.a,\"%\":function(t,e){return(100*t).toFixed(e)},b:function(t){return Math.round(t).toString(2)},c:function(t){return t+\"\"},d:function(t){return Math.round(t).toString(10)},e:function(t,e){return t.toExponential(e)},f:function(t,e){return t.toFixed(e)},g:function(t,e){return t.toPrecision(e)},o:function(t){return Math.round(t).toString(8)},p:function(t,e){return n.i(o.a)(100*t,e)},r:o.a,s:i.a,X:function(t){return Math.round(t).toString(16).toUpperCase()},x:function(t){return Math.round(t).toString(16)}}},function(t,e,n){\"use strict\";function r(t){return t}var i=n(42),o=n(213),a=n(115),u=n(116),c=n(114),s=[\"y\",\"z\",\"a\",\"f\",\"p\",\"n\",\"µ\",\"m\",\"\",\"k\",\"M\",\"G\",\"T\",\"P\",\"E\",\"Z\",\"Y\"];e.a=function(t){function e(t){function e(t){var e,n,a,u=_,l=b;if(\"c\"===y)l=x(t)+l,t=\"\";else{t=+t;var p=(t<0||1/t<0)&&(t*=-1,!0);if(t=x(t,m),p)for(e=-1,n=t.length,p=!1;++e<n;)if(a=t.charCodeAt(e),48<a&&a<58||\"x\"===y&&96<a&&a<103||\"X\"===y&&64<a&&a<71){p=!0;break}if(u=(p?\"(\"===o?o:\"-\":\"-\"===o||\"(\"===o?\"\":o)+u,l=l+(\"s\"===y?s[8+c.b/3]:\"\")+(p&&\"(\"===o?\")\":\"\"),w)for(e=-1,n=t.length;++e<n;)if(a=t.charCodeAt(e),48>a||a>57){l=(46===a?h+t.slice(e+1):t.slice(e))+l,t=t.slice(0,e);break}}g&&!d&&(t=f(t,1/0));var C=u.length+t.length+l.length,M=C<v?new Array(v-C+1).join(r):\"\";switch(g&&d&&(t=f(M+t,M.length?v-l.length:1/0),M=\"\"),i){case\"<\":return u+t+l+M;case\"=\":return u+M+t+l;case\"^\":return M.slice(0,C=M.length>>1)+u+t+l+M.slice(C)}return M+u+t+l}t=n.i(a.a)(t);var r=t.fill,i=t.align,o=t.sign,l=t.symbol,d=t.zero,v=t.width,g=t.comma,m=t.precision,y=t.type,_=\"$\"===l?p[0]:\"#\"===l&&/[boxX]/.test(y)?\"0\"+y.toLowerCase():\"\",b=\"$\"===l?p[1]:/[%p]/.test(y)?\"%\":\"\",x=u.a[y],w=!y||/[defgprs%]/.test(y);return m=null==m?y?6:12:/[gprs]/.test(y)?Math.max(1,Math.min(21,m)):Math.max(0,Math.min(20,m)),e.toString=function(){return t+\"\"},e}function l(t,r){var o=e((t=n.i(a.a)(t),t.type=\"f\",t)),u=3*Math.max(-8,Math.min(8,Math.floor(n.i(i.a)(r)/3))),c=Math.pow(10,-u),l=s[8+u/3];return function(t){return o(c*t)+l}}var f=t.grouping&&t.thousands?n.i(o.a)(t.grouping,t.thousands):r,p=t.currency,h=t.decimal;return{format:e,formatPrefix:l}}},function(t,e,n){\"use strict\";var r=n(63);e.a=function(t,e){var i,o=e?e.length:0,a=t?Math.min(o,t.length):0,u=new Array(o),c=new Array(o);for(i=0;i<a;++i)u[i]=n.i(r.a)(t[i],e[i]);for(;i<o;++i)c[i]=e[i];return function(t){for(i=0;i<a;++i)c[i]=u[i](t);return c}}},function(t,e,n){\"use strict\";var r=n(62);e.a=function(t){var e=t.length;return function(i){var o=Math.floor(((i%=1)<0?++i:i)*e),a=t[(o+e-1)%e],u=t[o%e],c=t[(o+1)%e],s=t[(o+2)%e];return n.i(r.b)((i-o/e)*e,a,u,c,s)}}},function(t,e,n){\"use strict\";e.a=function(t){return function(){return t}}},function(t,e,n){\"use strict\";e.a=function(t,e){var n=new Date;return t=+t,e-=t,function(r){return n.setTime(t+e*r),n}}},function(t,e,n){\"use strict\";var r=n(63);e.a=function(t,e){var i,o={},a={};null!==t&&\"object\"==typeof t||(t={}),null!==e&&\"object\"==typeof e||(e={});for(i in e)i in t?o[i]=n.i(r.a)(t[i],e[i]):a[i]=e[i];return function(t){for(i in o)a[i]=o[i](t);return a}}},function(t,e,n){\"use strict\";function r(t){return function(e){var r,o,a=e.length,u=new Array(a),c=new Array(a),s=new Array(a);for(r=0;r<a;++r)o=n.i(i.rgb)(e[r]),u[r]=o.r||0,c[r]=o.g||0,s[r]=o.b||0;return u=t(u),c=t(c),s=t(s),o.opacity=1,function(t){return o.r=u(t),o.g=c(t),o.b=s(t),o+\"\"}}}var i=n(10),o=n(62),a=n(119),u=n(32);e.a=function t(e){function r(t,e){var r=o((t=n.i(i.rgb)(t)).r,(e=n.i(i.rgb)(e)).r),a=o(t.g,e.g),c=o(t.b,e.b),s=n.i(u.a)(t.opacity,e.opacity);return function(e){return t.r=r(e),t.g=a(e),t.b=c(e),t.opacity=s(e),t+\"\"}}var o=n.i(u.c)(e);return r.gamma=t,r}(1);r(o.a),r(a.a)},function(t,e,n){\"use strict\";function r(t){return function(){return t}}function i(t){return function(e){return t(e)+\"\"}}var o=n(43),a=/[-+]?(?:\\d+\\.?\\d*|\\.?\\d+)(?:[eE][-+]?\\d+)?/g,u=new RegExp(a.source,\"g\");e.a=function(t,e){var c,s,l,f=a.lastIndex=u.lastIndex=0,p=-1,h=[],d=[];for(t+=\"\",e+=\"\";(c=a.exec(t))&&(s=u.exec(e));)(l=s.index)>f&&(l=e.slice(f,l),h[p]?h[p]+=l:h[++p]=l),(c=c[0])===(s=s[0])?h[p]?h[p]+=s:h[++p]=s:(h[++p]=null,d.push({i:p,x:n.i(o.a)(c,s)})),f=u.lastIndex;return f<e.length&&(l=e.slice(f),h[p]?h[p]+=l:h[++p]=l),h.length<2?d[0]?i(d[0].x):r(e):(e=d.length,function(t){for(var n,r=0;r<e;++r)h[(n=d[r]).i]=n.x(t);return h.join(\"\")})}},function(t,e,n){\"use strict\";e.a=function(t,e){t=t.slice();var n,r=0,i=t.length-1,o=t[r],a=t[i];return a<o&&(n=r,r=i,i=n,n=o,o=a,a=n),t[r]=e.floor(o),t[i]=e.ceil(a),t}},function(t,e,n){\"use strict\";e.a=function(t){return+t}},function(t,e,n){\"use strict\";function r(t){function e(e){var n=e+\"\",r=u.get(n);if(!r){if(s!==a)return s;u.set(n,r=c.push(e))}return t[(r-1)%t.length]}var u=n.i(i.a)(),c=[],s=a;return t=null==t?[]:o.b.call(t),e.domain=function(t){if(!arguments.length)return c.slice();c=[],u=n.i(i.a)();for(var r,o,a=-1,s=t.length;++a<s;)u.has(o=(r=t[a])+\"\")||u.set(o,c.push(r));return e},e.range=function(n){return arguments.length?(t=o.b.call(n),e):t.slice()},e.unknown=function(t){return arguments.length?(s=t,e):s},e.copy=function(){return r().domain(c).range(t).unknown(s)},e}var i=n(203),o=n(16);n.d(e,\"b\",function(){return a}),e.a=r;var a={name:\"implicit\"}},function(t,e,n){\"use strict\";function r(t){return new Date(t)}function i(t){return t instanceof Date?+t:+new Date(+t)}function o(t,e,c,s,b,x,w,C,M){function k(n){return(w(n)<n?N:x(n)<n?A:b(n)<n?O:s(n)<n?I:e(n)<n?c(n)<n?D:R:t(n)<n?L:U)(n)}function E(e,r,i,o){if(null==e&&(e=10),\"number\"==typeof e){var u=Math.abs(i-r)/e,c=n.i(a.d)(function(t){return t[2]}).right(F,u);c===F.length?(o=n.i(a.b)(r/_,i/_,e),e=t):c?(c=F[u/F[c-1][2]<F[c][2]/u?c-1:c],o=c[1],e=c[0]):(o=n.i(a.b)(r,i,e),e=C)}return null==o?e:e.every(o)}var T=n.i(f.a)(f.b,u.a),S=T.invert,P=T.domain,N=M(\".%L\"),A=M(\":%S\"),O=M(\"%I:%M\"),I=M(\"%I %p\"),D=M(\"%a %d\"),R=M(\"%b %d\"),L=M(\"%B\"),U=M(\"%Y\"),F=[[w,1,h],[w,5,5*h],[w,15,15*h],[w,30,30*h],[x,1,d],[x,5,5*d],[x,15,15*d],[x,30,30*d],[b,1,v],[b,3,3*v],[b,6,6*v],[b,12,12*v],[s,1,g],[s,2,2*g],[c,1,m],[e,1,y],[e,3,3*y],[t,1,_]];return T.invert=function(t){return new Date(S(t))},T.domain=function(t){return arguments.length?P(l.a.call(t,i)):P().map(r)},T.ticks=function(t,e){var n,r=P(),i=r[0],o=r[r.length-1],a=o<i;return a&&(n=i,i=o,o=n),n=E(t,i,o,e),n=n?n.range(i,o+1):[],a?n.reverse():n},T.tickFormat=function(t,e){return null==e?k:M(e)},T.nice=function(t,e){var r=P();return(t=E(t,r[0],r[r.length-1],e))?P(n.i(p.a)(r,t)):T},T.copy=function(){return n.i(f.c)(T,o(t,e,c,s,b,x,w,C,M))},T}var a=n(12),u=n(31),c=n(79),s=n(77),l=n(16),f=n(45),p=n(125);e.b=o;var h=1e3,d=60*h,v=60*d,g=24*v,m=7*g,y=30*g,_=365*g;e.a=function(){return o(c.b,c.o,c.p,c.a,c.q,c.r,c.s,c.t,s.timeFormat).domain([new Date(2e3,0,1),new Date(2e3,0,2)])}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(66);n.d(e,\"creator\",function(){return r.a});var i=n(247);n.d(e,\"local\",function(){return i.a});var o=n(130);n.d(e,\"matcher\",function(){return o.a});var a=n(248);n.d(e,\"mouse\",function(){return a.a});var u=n(67);n.d(e,\"namespace\",function(){return u.a});var c=n(68);n.d(e,\"namespaces\",function(){return c.a});var s=n(249);n.d(e,\"select\",function(){return s.a});var l=n(250);n.d(e,\"selectAll\",function(){return l.a});var f=n(7);n.d(e,\"selection\",function(){return f.a});var p=n(71);n.d(e,\"selector\",function(){return p.a});var h=n(133);n.d(e,\"selectorAll\",function(){return h.a});var d=n(278);n.d(e,\"touch\",function(){return d.a});var v=n(279);n.d(e,\"touches\",function(){return v.a});var g=n(73);n.d(e,\"window\",function(){return g.a});var m=n(70);n.d(e,\"event\",function(){return m.a}),n.d(e,\"customEvent\",function(){return m.b})},function(t,e,n){\"use strict\";var r=function(t){return function(){return this.matches(t)}};if(\"undefined\"!=typeof document){var i=document.documentElement;if(!i.matches){var o=i.webkitMatchesSelector||i.msMatchesSelector||i.mozMatchesSelector||i.oMatchesSelector;r=function(t){return function(){return o.call(this,t)}}}}e.a=r},function(t,e,n){\"use strict\";function r(t,e){this.ownerDocument=t.ownerDocument,this.namespaceURI=t.namespaceURI,this._next=null,this._parent=t,this.__data__=e}var i=n(132),o=n(7);e.b=r,e.a=function(){return new o.b(this._enter||this._groups.map(i.a),this._parents)},r.prototype={constructor:r,appendChild:function(t){return this._parent.insertBefore(t,this._next)},insertBefore:function(t,e){return this._parent.insertBefore(t,e)},querySelector:function(t){return this._parent.querySelector(t)},querySelectorAll:function(t){return this._parent.querySelectorAll(t)}}},function(t,e,n){\"use strict\";e.a=function(t){return new Array(t.length)}},function(t,e,n){\"use strict\";function r(){return[]}e.a=function(t){return null==t?r:function(){return this.querySelectorAll(t)}}},function(t,e,n){\"use strict\";Object.defineProperty(e,\"__esModule\",{value:!0});var r=n(280);n.d(e,\"arc\",function(){return r.a});var i=n(135);n.d(e,\"area\",function(){return i.a});var o=n(75);n.d(e,\"line\",function(){return o.a});var a=n(299);n.d(e,\"pie\",function(){return a.a});var u=n(300);n.d(e,\"radialArea\",function(){return u.a});var c=n(140);n.d(e,\"radialLine\",function(){return c.a});var s=n(302);n.d(e,\"symbol\",function(){return s.a}),n.d(e,\"symbols\",function(){return s.b});var l=n(141);n.d(e,\"symbolCircle\",function(){return l.a});var f=n(142);n.d(e,\"symbolCross\",function(){return f.a});var p=n(143);n.d(e,\"symbolDiamond\",function(){return p.a});var h=n(144);n.d(e,\"symbolSquare\",function(){return h.a});var d=n(145);n.d(e,\"symbolStar\",function(){return d.a});var v=n(146);n.d(e,\"symbolTriangle\",function(){return v.a});var g=n(147);n.d(e,\"symbolWye\",function(){return g.a});var m=n(282);n.d(e,\"curveBasisClosed\",function(){return m.a});var y=n(283);n.d(e,\"curveBasisOpen\",function(){return y.a});var _=n(46);n.d(e,\"curveBasis\",function(){return _.a});var b=n(284);n.d(e,\"curveBundle\",function(){return b.a});var x=n(136);n.d(e,\"curveCardinalClosed\",function(){return x.a});var w=n(137);n.d(e,\"curveCardinalOpen\",function(){return w.a});var C=n(47);n.d(e,\"curveCardinal\",function(){return C.a});var M=n(285);n.d(e,\"curveCatmullRomClosed\",function(){return M.a});var k=n(286);n.d(e,\"curveCatmullRomOpen\",function(){return k.a});var E=n(74);n.d(e,\"curveCatmullRom\",function(){return E.a});var T=n(287);n.d(e,\"curveLinearClosed\",function(){return T.a});var S=n(48);n.d(e,\"curveLinear\",function(){return S.a});var P=n(288);n.d(e,\"curveMonotoneX\",function(){return P.a}),n.d(e,\"curveMonotoneY\",function(){return P.b});var N=n(289);n.d(e,\"curveNatural\",function(){return N.a});var A=n(290);n.d(e,\"curveStep\",function(){return A.a}),n.d(e,\"curveStepAfter\",function(){return A.b}),n.d(e,\"curveStepBefore\",function(){return A.c});var O=n(301);n.d(e,\"stack\",function(){return O.a});var I=n(293);n.d(e,\"stackOffsetExpand\",function(){return I.a});var D=n(36);n.d(e,\"stackOffsetNone\",function(){return D.a});var R=n(294);n.d(e,\"stackOffsetSilhouette\",function(){return R.a});var L=n(295);n.d(e,\"stackOffsetWiggle\",function(){return L.a});var U=n(76);n.d(e,\"stackOrderAscending\",function(){return U.a});var F=n(296);n.d(e,\"stackOrderDescending\",function(){return F.a});var j=n(297);n.d(e,\"stackOrderInsideOut\",function(){return j.a});var B=n(37);n.d(e,\"stackOrderNone\",function(){return B.a});var W=n(298);n.d(e,\"stackOrderReverse\",function(){return W.a})},function(t,e,n){\"use strict\";var r=n(44),i=n(19),o=n(48),a=n(75),u=n(139);e.a=function(){function t(t){var e,i,o,a,u,g=t.length,m=!1,y=new Array(g),_=new Array(g);for(null==h&&(v=d(u=n.i(r.a)())),e=0;e<=g;++e){if(!(e<g&&p(a=t[e],e,t))===m)if(m=!m)i=e,v.areaStart(),v.lineStart();else{for(v.lineEnd(),v.lineStart(),o=e-1;o>=i;--o)v.point(y[o],_[o]);v.lineEnd(),v.areaEnd()}m&&(y[e]=+c(a,e,t),_[e]=+l(a,e,t),v.point(s?+s(a,e,t):y[e],f?+f(a,e,t):_[e]))}if(u)return v=null,u+\"\"||null}function e(){return n.i(a.a)().defined(p).curve(d).context(h)}var c=u.a,s=null,l=n.i(i.a)(0),f=u.b,p=n.i(i.a)(!0),h=null,d=o.a,v=null;return t.x=function(e){return arguments.length?(c=\"function\"==typeof e?e:n.i(i.a)(+e),s=null,t):c},t.x0=function(e){return arguments.length?(c=\"function\"==typeof e?e:n.i(i.a)(+e),t):c},t.x1=function(e){return arguments.length?(s=null==e?null:\"function\"==typeof e?e:n.i(i.a)(+e),t):s},t.y=function(e){return arguments.length?(l=\"function\"==typeof e?e:n.i(i.a)(+e),f=null,t):l},t.y0=function(e){return arguments.length?(l=\"function\"==typeof e?e:n.i(i.a)(+e),t):l},t.y1=function(e){return arguments.length?(f=null==e?null:\"function\"==typeof e?e:n.i(i.a)(+e),t):f},t.lineX0=t.lineY0=function(){return e().x(c).y(l)},t.lineY1=function(){return e().x(c).y(f)},t.lineX1=function(){return e().x(s).y(l)},t.defined=function(e){return arguments.length?(p=\"function\"==typeof e?e:n.i(i.a)(!!e),t):p},t.curve=function(e){return arguments.length?(d=e,null!=h&&(v=d(h)),t):d},t.context=function(e){return arguments.length?(null==e?h=v=null:v=d(h=e),t):h},t}},function(t,e,n){\"use strict\";function r(t,e){this._context=t,this._k=(1-e)/6}var i=n(49),o=n(47);e.b=r,r.prototype={areaStart:i.a,areaEnd:i.a,lineStart:function(){this._x0=this._x1=this._x2=this._x3=this._x4=this._x5=this._y0=this._y1=this._y2=this._y3=this._y4=this._y5=NaN,this._point=0},lineEnd:function(){switch(this._point){case 1:this._context.moveTo(this._x3,this._y3),this._context.closePath();break;case 2:this._context.lineTo(this._x3,this._y3),this._context.closePath();break;case 3:this.point(this._x3,this._y3),this.point(this._x4,this._y4),this.point(this._x5,this._y5)}},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._x3=t,this._y3=e;break;case 1:this._point=2,this._context.moveTo(this._x4=t,this._y4=e);break;case 2:this._point=3,this._x5=t,this._y5=e;break;default:n.i(o.c)(this,t,e)}this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return new r(t,e)}return n.tension=function(e){return t(+e)},n}(0)},function(t,e,n){\"use strict\";function r(t,e){this._context=t,this._k=(1-e)/6}var i=n(47);e.b=r,r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._x2=this._y0=this._y1=this._y2=NaN,this._point=0},lineEnd:function(){(this._line||0!==this._line&&3===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1;break;case 1:this._point=2;break;case 2:this._point=3,this._line?this._context.lineTo(this._x2,this._y2):this._context.moveTo(this._x2,this._y2);break;case 3:this._point=4;default:n.i(i.c)(this,t,e)}this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return new r(t,e)}return n.tension=function(e){return t(+e)},n}(0)},function(t,e,n){\"use strict\";function r(t){this._curve=t}function i(t){function e(e){return new r(t(e))}return e._curve=t,e}var o=n(48);n.d(e,\"b\",function(){return a}),e.a=i;var a=i(o.a);r.prototype={areaStart:function(){this._curve.areaStart()},areaEnd:function(){this._curve.areaEnd()},lineStart:function(){this._curve.lineStart()},lineEnd:function(){this._curve.lineEnd()},point:function(t,e){this._curve.point(e*Math.sin(t),e*-Math.cos(t))}}},function(t,e,n){\"use strict\";function r(t){return t[0]}function i(t){return t[1]}e.a=r,e.b=i},function(t,e,n){\"use strict\";function r(t){var e=t.curve;return t.angle=t.x,delete t.x,t.radius=t.y,delete t.y,t.curve=function(t){return arguments.length?e(n.i(i.a)(t)):e()._curve},t}var i=n(138),o=n(75);e.b=r,e.a=function(){return r(n.i(o.a)().curve(i.b))}},function(t,e,n){\"use strict\";var r=n(35);e.a={draw:function(t,e){var n=Math.sqrt(e/r.b);t.moveTo(n,0),t.arc(0,0,n,0,r.c)}}},function(t,e,n){\"use strict\";e.a={draw:function(t,e){var n=Math.sqrt(e/5)/2;t.moveTo(-3*n,-n),t.lineTo(-n,-n),t.lineTo(-n,-3*n),t.lineTo(n,-3*n),t.lineTo(n,-n),t.lineTo(3*n,-n),t.lineTo(3*n,n),t.lineTo(n,n),t.lineTo(n,3*n),t.lineTo(-n,3*n),t.lineTo(-n,n),t.lineTo(-3*n,n),t.closePath()}}},function(t,e,n){\"use strict\";var r=Math.sqrt(1/3),i=2*r;e.a={draw:function(t,e){var n=Math.sqrt(e/i),o=n*r;t.moveTo(0,-n),t.lineTo(o,0),t.lineTo(0,n),t.lineTo(-o,0),t.closePath()}}},function(t,e,n){\"use strict\";e.a={draw:function(t,e){var n=Math.sqrt(e),r=-n/2;t.rect(r,r,n,n)}}},function(t,e,n){\"use strict\";var r=n(35),i=.8908130915292852,o=Math.sin(r.b/10)/Math.sin(7*r.b/10),a=Math.sin(r.c/10)*o,u=-Math.cos(r.c/10)*o;e.a={draw:function(t,e){var n=Math.sqrt(e*i),o=a*n,c=u*n;t.moveTo(0,-n),t.lineTo(o,c);for(var s=1;s<5;++s){var l=r.c*s/5,f=Math.cos(l),p=Math.sin(l);t.lineTo(p*n,-f*n),t.lineTo(f*o-p*c,p*o+f*c)}t.closePath()}}},function(t,e,n){\"use strict\";var r=Math.sqrt(3);e.a={draw:function(t,e){var n=-Math.sqrt(e/(3*r));t.moveTo(0,2*n),t.lineTo(-r*n,-n),t.lineTo(r*n,-n),t.closePath()}}},function(t,e,n){\"use strict\";var r=-.5,i=Math.sqrt(3)/2,o=1/Math.sqrt(12),a=3*(o/2+1);e.a={draw:function(t,e){var n=Math.sqrt(e/a),u=n/2,c=n*o,s=u,l=n*o+n,f=-s,p=l;t.moveTo(u,c),t.lineTo(s,l),t.lineTo(f,p),t.lineTo(r*u-i*c,i*u+r*c),t.lineTo(r*s-i*l,i*s+r*l),t.lineTo(r*f-i*p,i*f+r*p),t.lineTo(r*u+i*c,r*c-i*u),t.lineTo(r*s+i*l,r*l-i*s),t.lineTo(r*f+i*p,r*p-i*f),t.closePath()}}},function(t,e,n){\"use strict\";function r(t){return t.toISOString()}var i=n(78);n.d(e,\"b\",function(){return o});var o=\"%Y-%m-%dT%H:%M:%S.%LZ\",a=Date.prototype.toISOString?r:n.i(i.d)(o);e.a=a},function(t,e,n){\"use strict\";function r(t){if(0<=t.y&&t.y<100){var e=new Date(-1,t.m,t.d,t.H,t.M,t.S,t.L);return e.setFullYear(t.y),e}return new Date(t.y,t.m,t.d,t.H,t.M,t.S,t.L)}function i(t){if(0<=t.y&&t.y<100){var e=new Date(Date.UTC(-1,t.m,t.d,t.H,t.M,t.S,t.L));return e.setUTCFullYear(t.y),e}return new Date(Date.UTC(t.y,t.m,t.d,t.H,t.M,t.S,t.L))}function o(t){return{y:t,m:0,d:1,H:0,M:0,S:0,L:0}}function a(t){function e(t,e){return function(n){var r,i,o,a=[],u=-1,c=0,s=t.length;for(n instanceof Date||(n=new Date(+n));++u<s;)37===t.charCodeAt(u)&&(a.push(t.slice(c,u)),null!=(i=et[r=t.charAt(++u)])?r=t.charAt(++u):i=\"e\"===r?\" \":\"0\",(o=e[r])&&(r=o(n,i)),a.push(r),c=u+1);return a.push(t.slice(c,u)),a.join(\"\")}}function n(t,e){return function(n){var r=o(1900),u=a(r,t,n+=\"\",0);if(u!=n.length)return null;if(\"p\"in r&&(r.H=r.H%12+12*r.p),\"W\"in r||\"U\"in r){\"w\"in r||(r.w=\"W\"in r?1:0);var c=\"Z\"in r?i(o(r.y)).getUTCDay():e(o(r.y)).getDay();r.m=0,r.d=\"W\"in r?(r.w+6)%7+7*r.W-(c+5)%7:r.w+7*r.U-(c+6)%7}return\"Z\"in r?(r.H+=r.Z/100|0,r.M+=r.Z%100,i(r)):e(r)}}function a(t,e,n,r){for(var i,o,a=0,u=e.length,c=n.length;a<u;){if(r>=c)return-1;if(i=e.charCodeAt(a++),37===i){if(i=e.charAt(a++),o=Ut[i in et?e.charAt(a++):i],!o||(r=o(t,n,r))<0)return-1}else if(i!=n.charCodeAt(r++))return-1}return r}function u(t,e,n){var r=kt.exec(e.slice(n));return r?(t.p=Et[r[0].toLowerCase()],n+r[0].length):-1}function c(t,e,n){var r=Pt.exec(e.slice(n));return r?(t.w=Nt[r[0].toLowerCase()],n+r[0].length):-1}function tt(t,e,n){var r=Tt.exec(e.slice(n));return r?(t.w=St[r[0].toLowerCase()],n+r[0].length):-1}function nt(t,e,n){var r=It.exec(e.slice(n));return r?(t.m=Dt[r[0].toLowerCase()],n+r[0].length):-1}function rt(t,e,n){var r=At.exec(e.slice(n));return r?(t.m=Ot[r[0].toLowerCase()],n+r[0].length):-1}function it(t,e,n){return a(t,mt,e,n)}function ot(t,e,n){return a(t,yt,e,n)}function at(t,e,n){return a(t,_t,e,n)}function ut(t){return wt[t.getDay()]}function ct(t){return xt[t.getDay()]}function st(t){return Mt[t.getMonth()]}function lt(t){return Ct[t.getMonth()]}function ft(t){return bt[+(t.getHours()>=12)]}function pt(t){return wt[t.getUTCDay()]}function ht(t){return xt[t.getUTCDay()]}function dt(t){return Mt[t.getUTCMonth()]}function vt(t){return Ct[t.getUTCMonth()]}function gt(t){return bt[+(t.getUTCHours()>=12)]}var mt=t.dateTime,yt=t.date,_t=t.time,bt=t.periods,xt=t.days,wt=t.shortDays,Ct=t.months,Mt=t.shortMonths,kt=s(bt),Et=l(bt),Tt=s(xt),St=l(xt),Pt=s(wt),Nt=l(wt),At=s(Ct),Ot=l(Ct),It=s(Mt),Dt=l(Mt),Rt={a:ut,A:ct,b:st,B:lt,c:null,d:k,e:k,H:E,I:T,j:S,L:P,m:N,M:A,p:ft,S:O,U:I,w:D,W:R,x:null,X:null,y:L,Y:U,Z:F,\"%\":J},Lt={a:pt,A:ht,b:dt,B:vt,c:null,d:j,e:j,H:B,I:W,j:V,L:z,m:H,M:q,p:gt,S:Y,U:K,w:G,W:$,x:null,X:null,y:X,Y:Z,Z:Q,\"%\":J},Ut={a:c,A:tt,b:nt,B:rt,c:it,d:y,e:y,H:b,I:b,j:_,L:C,m:m,M:x,p:u,S:w,U:p,w:f,W:h,x:ot,X:at,y:v,Y:d,Z:g,\"%\":M};return Rt.x=e(yt,Rt),Rt.X=e(_t,Rt),Rt.c=e(mt,Rt),Lt.x=e(yt,Lt),Lt.X=e(_t,Lt),Lt.c=e(mt,Lt),{format:function(t){var n=e(t+=\"\",Rt);return n.toString=function(){return t},n},parse:function(t){var e=n(t+=\"\",r);return e.toString=function(){return t},e},utcFormat:function(t){var n=e(t+=\"\",Lt);return n.toString=function(){return t},n},utcParse:function(t){var e=n(t,i);return e.toString=function(){return t},e}}}function u(t,e,n){var r=t<0?\"-\":\"\",i=(r?-t:t)+\"\",o=i.length;return r+(o<n?new Array(n-o+1).join(e)+i:i)}function c(t){return t.replace(it,\"\\\\$&\")}function s(t){return new RegExp(\"^(?:\"+t.map(c).join(\"|\")+\")\",\"i\")}function l(t){for(var e={},n=-1,r=t.length;++n<r;)e[t[n].toLowerCase()]=n;return e}function f(t,e,n){var r=nt.exec(e.slice(n,n+1));return r?(t.w=+r[0],n+r[0].length):-1}function p(t,e,n){var r=nt.exec(e.slice(n));return r?(t.U=+r[0],n+r[0].length):-1}function h(t,e,n){var r=nt.exec(e.slice(n));return r?(t.W=+r[0],n+r[0].length):-1}function d(t,e,n){var r=nt.exec(e.slice(n,n+4));return r?(t.y=+r[0],n+r[0].length):-1}function v(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.y=+r[0]+(+r[0]>68?1900:2e3),n+r[0].length):-1}function g(t,e,n){var r=/^(Z)|([+-]\\d\\d)(?:\\:?(\\d\\d))?/.exec(e.slice(n,n+6));return r?(t.Z=r[1]?0:-(r[2]+(r[3]||\"00\")),n+r[0].length):-1}function m(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.m=r[0]-1,n+r[0].length):-1}function y(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.d=+r[0],n+r[0].length):-1}function _(t,e,n){var r=nt.exec(e.slice(n,n+3));return r?(t.m=0,t.d=+r[0],n+r[0].length):-1}function b(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.H=+r[0],n+r[0].length):-1}function x(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.M=+r[0],n+r[0].length):-1}function w(t,e,n){var r=nt.exec(e.slice(n,n+2));return r?(t.S=+r[0],n+r[0].length):-1}function C(t,e,n){var r=nt.exec(e.slice(n,n+3));return r?(t.L=+r[0],n+r[0].length):-1}function M(t,e,n){var r=rt.exec(e.slice(n,n+1));return r?n+r[0].length:-1}function k(t,e){return u(t.getDate(),e,2)}function E(t,e){return u(t.getHours(),e,2)}function T(t,e){return u(t.getHours()%12||12,e,2)}function S(t,e){return u(1+tt.a.count(n.i(tt.b)(t),t),e,3)}function P(t,e){return u(t.getMilliseconds(),e,3)}function N(t,e){return u(t.getMonth()+1,e,2)}function A(t,e){return u(t.getMinutes(),e,2)}function O(t,e){return u(t.getSeconds(),e,2)}function I(t,e){return u(tt.c.count(n.i(tt.b)(t),t),e,2)}function D(t){return t.getDay()}function R(t,e){return u(tt.d.count(n.i(tt.b)(t),t),e,2)}function L(t,e){return u(t.getFullYear()%100,e,2)}function U(t,e){return u(t.getFullYear()%1e4,e,4)}function F(t){var e=t.getTimezoneOffset();return(e>0?\"-\":(e*=-1,\"+\"))+u(e/60|0,\"0\",2)+u(e%60,\"0\",2)}function j(t,e){return u(t.getUTCDate(),e,2)}function B(t,e){return u(t.getUTCHours(),e,2)}function W(t,e){return u(t.getUTCHours()%12||12,e,2)}function V(t,e){return u(1+tt.e.count(n.i(tt.f)(t),t),e,3)}function z(t,e){return u(t.getUTCMilliseconds(),e,3)}function H(t,e){return u(t.getUTCMonth()+1,e,2)}function q(t,e){return u(t.getUTCMinutes(),e,2)}function Y(t,e){return u(t.getUTCSeconds(),e,2)}function K(t,e){return u(tt.g.count(n.i(tt.f)(t),t),e,2)}function G(t){return t.getUTCDay()}function $(t,e){return u(tt.h.count(n.i(tt.f)(t),t),e,2)}function X(t,e){return u(t.getUTCFullYear()%100,e,2)}function Z(t,e){return u(t.getUTCFullYear()%1e4,e,4)}function Q(){return\"+0000\"}function J(){return\"%\"}var tt=n(79);e.a=a;var et={\"-\":\"\",_:\" \",0:\"0\"},nt=/^\\s*\\d+/,rt=/^%/,it=/[\\\\\\^\\$\\*\\+\\?\\|\\[\\]\\(\\)\\.\\{\\}]/g},function(t,e,n){\"use strict\";var r=n(8),i={listen:function(t,e,n){return t.addEventListener?(t.addEventListener(e,n,!1),{remove:function(){t.removeEventListener(e,n,!1)}}):t.attachEvent?(t.attachEvent(\"on\"+e,n),{remove:function(){t.detachEvent(\"on\"+e,n)}}):void 0},capture:function(t,e,n){return t.addEventListener?(t.addEventListener(e,n,!0),{remove:function(){t.removeEventListener(e,n,!0)}}):{remove:r}},registerDefault:function(){}};t.exports=i},function(t,e,n){\"use strict\";function r(t){try{t.focus()}catch(t){}}t.exports=r},function(t,e,n){\"use strict\";function r(){if(\"undefined\"==typeof document)return null;try{return document.activeElement||document.body}catch(t){return document.body}}t.exports=r},function(t,e){function n(){throw new Error(\"setTimeout has not been defined\")}function r(){throw new Error(\"clearTimeout has not been defined\")}function i(t){if(l===setTimeout)return setTimeout(t,0);if((l===n||!l)&&setTimeout)return l=setTimeout,setTimeout(t,0);try{return l(t,0)}catch(e){try{return l.call(null,t,0)}catch(e){return l.call(this,t,0)}}}function o(t){if(f===clearTimeout)return clearTimeout(t);if((f===r||!f)&&clearTimeout)return f=clearTimeout,clearTimeout(t);try{return f(t)}catch(e){try{return f.call(null,t)}catch(e){return f.call(this,t)}}}function a(){v&&h&&(v=!1,h.length?d=h.concat(d):g=-1,d.length&&u())}function u(){if(!v){var t=i(a);v=!0;for(var e=d.length;e;){for(h=d,d=[];++g<e;)h&&h[g].run();g=-1,e=d.length}h=null,v=!1,o(t)}}function c(t,e){this.fun=t,this.array=e}function s(){}var l,f,p=t.exports={};!function(){try{l=\"function\"==typeof setTimeout?setTimeout:n}catch(t){l=n}try{f=\"function\"==typeof clearTimeout?clearTimeout:r}catch(t){f=r}}();var h,d=[],v=!1,g=-1;p.nextTick=function(t){var e=new Array(arguments.length-1);if(arguments.length>1)for(var n=1;n<arguments.length;n++)e[n-1]=arguments[n];d.push(new c(t,e)),1!==d.length||v||i(u)},c.prototype.run=function(){this.fun.apply(null,this.array)},p.title=\"browser\",p.browser=!0,p.env={},p.argv=[],p.version=\"\",p.versions={},p.on=s,p.addListener=s,p.once=s,p.off=s,p.removeListener=s,p.removeAllListeners=s,p.emit=s,p.binding=function(t){throw new Error(\"process.binding is not supported\")},p.cwd=function(){return\"/\"},p.chdir=function(t){throw new Error(\"process.chdir is not supported\")},p.umask=function(){\n",
              "return 0}},function(t,e,n){\"use strict\";function r(t,e){return t+e.charAt(0).toUpperCase()+e.substring(1)}var i={animationIterationCount:!0,borderImageOutset:!0,borderImageSlice:!0,borderImageWidth:!0,boxFlex:!0,boxFlexGroup:!0,boxOrdinalGroup:!0,columnCount:!0,flex:!0,flexGrow:!0,flexPositive:!0,flexShrink:!0,flexNegative:!0,flexOrder:!0,gridRow:!0,gridColumn:!0,fontWeight:!0,lineClamp:!0,lineHeight:!0,opacity:!0,order:!0,orphans:!0,tabSize:!0,widows:!0,zIndex:!0,zoom:!0,fillOpacity:!0,floodOpacity:!0,stopOpacity:!0,strokeDasharray:!0,strokeDashoffset:!0,strokeMiterlimit:!0,strokeOpacity:!0,strokeWidth:!0},o=[\"Webkit\",\"ms\",\"Moz\",\"O\"];Object.keys(i).forEach(function(t){o.forEach(function(e){i[r(e,t)]=i[t]})});var a={background:{backgroundAttachment:!0,backgroundColor:!0,backgroundImage:!0,backgroundPositionX:!0,backgroundPositionY:!0,backgroundRepeat:!0},backgroundPosition:{backgroundPositionX:!0,backgroundPositionY:!0},border:{borderWidth:!0,borderStyle:!0,borderColor:!0},borderBottom:{borderBottomWidth:!0,borderBottomStyle:!0,borderBottomColor:!0},borderLeft:{borderLeftWidth:!0,borderLeftStyle:!0,borderLeftColor:!0},borderRight:{borderRightWidth:!0,borderRightStyle:!0,borderRightColor:!0},borderTop:{borderTopWidth:!0,borderTopStyle:!0,borderTopColor:!0},font:{fontStyle:!0,fontVariant:!0,fontWeight:!0,fontSize:!0,lineHeight:!0,fontFamily:!0},outline:{outlineWidth:!0,outlineStyle:!0,outlineColor:!0}},u={isUnitlessNumber:i,shorthandPropertyExpansions:a};t.exports=u},function(t,e,n){\"use strict\";function r(t,e){if(!(t instanceof e))throw new TypeError(\"Cannot call a class as a function\")}var i=n(2),o=n(17),a=(n(0),function(){function t(e){r(this,t),this._callbacks=null,this._contexts=null,this._arg=e}return t.prototype.enqueue=function(t,e){this._callbacks=this._callbacks||[],this._callbacks.push(t),this._contexts=this._contexts||[],this._contexts.push(e)},t.prototype.notifyAll=function(){var t=this._callbacks,e=this._contexts,n=this._arg;if(t&&e){t.length!==e.length?i(\"24\"):void 0,this._callbacks=null,this._contexts=null;for(var r=0;r<t.length;r++)t[r].call(e[r],n);t.length=0,e.length=0}},t.prototype.checkpoint=function(){return this._callbacks?this._callbacks.length:0},t.prototype.rollback=function(t){this._callbacks&&this._contexts&&(this._callbacks.length=t,this._contexts.length=t)},t.prototype.reset=function(){this._callbacks=null,this._contexts=null},t.prototype.destructor=function(){this.reset()},t}());t.exports=o.addPoolingTo(a)},function(t,e,n){\"use strict\";function r(t){return!!s.hasOwnProperty(t)||!c.hasOwnProperty(t)&&(u.test(t)?(s[t]=!0,!0):(c[t]=!0,!1))}function i(t,e){return null==e||t.hasBooleanValue&&!e||t.hasNumericValue&&isNaN(e)||t.hasPositiveNumericValue&&e<1||t.hasOverloadedBooleanValue&&e===!1}var o=n(21),a=(n(4),n(9),n(394)),u=(n(1),new RegExp(\"^[\"+o.ATTRIBUTE_NAME_START_CHAR+\"][\"+o.ATTRIBUTE_NAME_CHAR+\"]*$\")),c={},s={},l={createMarkupForID:function(t){return o.ID_ATTRIBUTE_NAME+\"=\"+a(t)},setAttributeForID:function(t,e){t.setAttribute(o.ID_ATTRIBUTE_NAME,e)},createMarkupForRoot:function(){return o.ROOT_ATTRIBUTE_NAME+'=\"\"'},setAttributeForRoot:function(t){t.setAttribute(o.ROOT_ATTRIBUTE_NAME,\"\")},createMarkupForProperty:function(t,e){var n=o.properties.hasOwnProperty(t)?o.properties[t]:null;if(n){if(i(n,e))return\"\";var r=n.attributeName;return n.hasBooleanValue||n.hasOverloadedBooleanValue&&e===!0?r+'=\"\"':r+\"=\"+a(e)}return o.isCustomAttribute(t)?null==e?\"\":t+\"=\"+a(e):null},createMarkupForCustomAttribute:function(t,e){return r(t)&&null!=e?t+\"=\"+a(e):\"\"},setValueForProperty:function(t,e,n){var r=o.properties.hasOwnProperty(e)?o.properties[e]:null;if(r){var a=r.mutationMethod;if(a)a(t,n);else{if(i(r,n))return void this.deleteValueForProperty(t,e);if(r.mustUseProperty)t[r.propertyName]=n;else{var u=r.attributeName,c=r.attributeNamespace;c?t.setAttributeNS(c,u,\"\"+n):r.hasBooleanValue||r.hasOverloadedBooleanValue&&n===!0?t.setAttribute(u,\"\"):t.setAttribute(u,\"\"+n)}}}else if(o.isCustomAttribute(e))return void l.setValueForAttribute(t,e,n)},setValueForAttribute:function(t,e,n){if(r(e)){null==n?t.removeAttribute(e):t.setAttribute(e,\"\"+n)}},deleteValueForAttribute:function(t,e){t.removeAttribute(e)},deleteValueForProperty:function(t,e){var n=o.properties.hasOwnProperty(e)?o.properties[e]:null;if(n){var r=n.mutationMethod;if(r)r(t,void 0);else if(n.mustUseProperty){var i=n.propertyName;n.hasBooleanValue?t[i]=!1:t[i]=\"\"}else t.removeAttribute(n.attributeName)}else o.isCustomAttribute(e)&&t.removeAttribute(e)}};t.exports=l},function(t,e,n){\"use strict\";var r={hasCachedChildNodes:1};t.exports=r},function(t,e,n){\"use strict\";function r(){if(this._rootNodeID&&this._wrapperState.pendingUpdate){this._wrapperState.pendingUpdate=!1;var t=this._currentElement.props,e=u.getValue(t);null!=e&&i(this,Boolean(t.multiple),e)}}function i(t,e,n){var r,i,o=c.getNodeFromInstance(t).options;if(e){for(r={},i=0;i<n.length;i++)r[\"\"+n[i]]=!0;for(i=0;i<o.length;i++){var a=r.hasOwnProperty(o[i].value);o[i].selected!==a&&(o[i].selected=a)}}else{for(r=\"\"+n,i=0;i<o.length;i++)if(o[i].value===r)return void(o[i].selected=!0);o.length&&(o[0].selected=!0)}}function o(t){var e=this._currentElement.props,n=u.executeOnChange(e,t);return this._rootNodeID&&(this._wrapperState.pendingUpdate=!0),s.asap(r,this),n}var a=n(3),u=n(85),c=n(4),s=n(11),l=(n(1),!1),f={getHostProps:function(t,e){return a({},e,{onChange:t._wrapperState.onChange,value:void 0})},mountWrapper:function(t,e){var n=u.getValue(e);t._wrapperState={pendingUpdate:!1,initialValue:null!=n?n:e.defaultValue,listeners:null,onChange:o.bind(t),wasMultiple:Boolean(e.multiple)},void 0===e.value||void 0===e.defaultValue||l||(l=!0)},getSelectValueContext:function(t){return t._wrapperState.initialValue},postUpdateWrapper:function(t){var e=t._currentElement.props;t._wrapperState.initialValue=void 0;var n=t._wrapperState.wasMultiple;t._wrapperState.wasMultiple=Boolean(e.multiple);var r=u.getValue(e);null!=r?(t._wrapperState.pendingUpdate=!1,i(t,Boolean(e.multiple),r)):n!==Boolean(e.multiple)&&(null!=e.defaultValue?i(t,Boolean(e.multiple),e.defaultValue):i(t,Boolean(e.multiple),e.multiple?[]:\"\"))}};t.exports=f},function(t,e,n){\"use strict\";var r,i={injectEmptyComponentFactory:function(t){r=t}},o={create:function(t){return r(t)}};o.injection=i,t.exports=o},function(t,e,n){\"use strict\";var r={logTopLevelRenders:!1};t.exports=r},function(t,e,n){\"use strict\";function r(t){return u?void 0:a(\"111\",t.type),new u(t)}function i(t){return new c(t)}function o(t){return t instanceof c}var a=n(2),u=(n(0),null),c=null,s={injectGenericComponentClass:function(t){u=t},injectTextComponentClass:function(t){c=t}},l={createInternalComponent:r,createInstanceForText:i,isTextComponent:o,injection:s};t.exports=l},function(t,e,n){\"use strict\";function r(t){return o(document.documentElement,t)}var i=n(353),o=n(320),a=n(151),u=n(152),c={hasSelectionCapabilities:function(t){var e=t&&t.nodeName&&t.nodeName.toLowerCase();return e&&(\"input\"===e&&\"text\"===t.type||\"textarea\"===e||\"true\"===t.contentEditable)},getSelectionInformation:function(){var t=u();return{focusedElem:t,selectionRange:c.hasSelectionCapabilities(t)?c.getSelection(t):null}},restoreSelection:function(t){var e=u(),n=t.focusedElem,i=t.selectionRange;e!==n&&r(n)&&(c.hasSelectionCapabilities(n)&&c.setSelection(n,i),a(n))},getSelection:function(t){var e;if(\"selectionStart\"in t)e={start:t.selectionStart,end:t.selectionEnd};else if(document.selection&&t.nodeName&&\"input\"===t.nodeName.toLowerCase()){var n=document.selection.createRange();n.parentElement()===t&&(e={start:-n.moveStart(\"character\",-t.value.length),end:-n.moveEnd(\"character\",-t.value.length)})}else e=i.getOffsets(t);return e||{start:0,end:0}},setSelection:function(t,e){var n=e.start,r=e.end;if(void 0===r&&(r=n),\"selectionStart\"in t)t.selectionStart=n,t.selectionEnd=Math.min(r,t.value.length);else if(document.selection&&t.nodeName&&\"input\"===t.nodeName.toLowerCase()){var o=t.createTextRange();o.collapse(!0),o.moveStart(\"character\",n),o.moveEnd(\"character\",r-n),o.select()}else i.setOffsets(t,e)}};t.exports=c},function(t,e,n){\"use strict\";function r(t,e){for(var n=Math.min(t.length,e.length),r=0;r<n;r++)if(t.charAt(r)!==e.charAt(r))return r;return t.length===e.length?-1:n}function i(t){return t?t.nodeType===D?t.documentElement:t.firstChild:null}function o(t){return t.getAttribute&&t.getAttribute(A)||\"\"}function a(t,e,n,r,i){var o;if(x.logTopLevelRenders){var a=t._currentElement.props.child,u=a.type;o=\"React mount: \"+(\"string\"==typeof u?u:u.displayName||u.name),console.time(o)}var c=M.mountComponent(t,n,null,_(t,e),i,0);o&&console.timeEnd(o),t._renderedComponent._topLevelWrapper=t,j._mountImageIntoNode(c,e,t,r,n)}function u(t,e,n,r){var i=E.ReactReconcileTransaction.getPooled(!n&&b.useCreateElement);i.perform(a,null,t,e,i,n,r),E.ReactReconcileTransaction.release(i)}function c(t,e,n){for(M.unmountComponent(t,n),e.nodeType===D&&(e=e.documentElement);e.lastChild;)e.removeChild(e.lastChild)}function s(t){var e=i(t);if(e){var n=y.getInstanceFromNode(e);return!(!n||!n._hostParent)}}function l(t){return!(!t||t.nodeType!==I&&t.nodeType!==D&&t.nodeType!==R)}function f(t){var e=i(t),n=e&&y.getInstanceFromNode(e);return n&&!n._hostParent?n:null}function p(t){var e=f(t);return e?e._hostContainerInfo._topLevelWrapper:null}var h=n(2),d=n(20),v=n(21),g=n(26),m=n(51),y=(n(15),n(4)),_=n(347),b=n(349),x=n(160),w=n(40),C=(n(9),n(363)),M=n(24),k=n(88),E=n(11),T=n(38),S=n(169),P=(n(0),n(55)),N=n(95),A=(n(1),v.ID_ATTRIBUTE_NAME),O=v.ROOT_ATTRIBUTE_NAME,I=1,D=9,R=11,L={},U=1,F=function(){this.rootID=U++};F.prototype.isReactComponent={},F.prototype.render=function(){return this.props.child},F.isReactTopLevelWrapper=!0;var j={TopLevelWrapper:F,_instancesByReactRootID:L,scrollMonitor:function(t,e){e()},_updateRootComponent:function(t,e,n,r,i){return j.scrollMonitor(r,function(){k.enqueueElementInternal(t,e,n),i&&k.enqueueCallbackInternal(t,i)}),t},_renderNewRootComponent:function(t,e,n,r){l(e)?void 0:h(\"37\"),m.ensureScrollValueMonitoring();var i=S(t,!1);E.batchedUpdates(u,i,e,n,r);var o=i._instance.rootID;return L[o]=i,i},renderSubtreeIntoContainer:function(t,e,n,r){return null!=t&&w.has(t)?void 0:h(\"38\"),j._renderSubtreeIntoContainer(t,e,n,r)},_renderSubtreeIntoContainer:function(t,e,n,r){k.validateCallback(r,\"ReactDOM.render\"),g.isValidElement(e)?void 0:h(\"39\",\"string\"==typeof e?\" Instead of passing a string like 'div', pass React.createElement('div') or <div />.\":\"function\"==typeof e?\" Instead of passing a class like Foo, pass React.createElement(Foo) or <Foo />.\":null!=e&&void 0!==e.props?\" This may be caused by unintentionally loading two independent copies of React.\":\"\");var a,u=g.createElement(F,{child:e});if(t){var c=w.get(t);a=c._processChildContext(c._context)}else a=T;var l=p(n);if(l){var f=l._currentElement,d=f.props.child;if(N(d,e)){var v=l._renderedComponent.getPublicInstance(),m=r&&function(){r.call(v)};return j._updateRootComponent(l,u,a,n,m),v}j.unmountComponentAtNode(n)}var y=i(n),_=y&&!!o(y),b=s(n),x=_&&!l&&!b,C=j._renderNewRootComponent(u,n,x,a)._renderedComponent.getPublicInstance();return r&&r.call(C),C},render:function(t,e,n){return j._renderSubtreeIntoContainer(null,t,e,n)},unmountComponentAtNode:function(t){l(t)?void 0:h(\"40\");var e=p(t);if(!e){s(t),1===t.nodeType&&t.hasAttribute(O);return!1}return delete L[e._instance.rootID],E.batchedUpdates(c,e,t,!1),!0},_mountImageIntoNode:function(t,e,n,o,a){if(l(e)?void 0:h(\"41\"),o){var u=i(e);if(C.canReuseMarkup(t,u))return void y.precacheNode(n,u);var c=u.getAttribute(C.CHECKSUM_ATTR_NAME);u.removeAttribute(C.CHECKSUM_ATTR_NAME);var s=u.outerHTML;u.setAttribute(C.CHECKSUM_ATTR_NAME,c);var f=t,p=r(f,s),v=\" (client) \"+f.substring(p-20,p+20)+\"\\n (server) \"+s.substring(p-20,p+20);e.nodeType===D?h(\"42\",v):void 0}if(e.nodeType===D?h(\"43\"):void 0,a.useCreateElement){for(;e.lastChild;)e.removeChild(e.lastChild);d.insertTreeBefore(e,t,null)}else P(e,t),y.precacheNode(n,e.firstChild)}};t.exports=j},function(t,e,n){\"use strict\";var r=n(2),i=n(26),o=(n(0),{HOST:0,COMPOSITE:1,EMPTY:2,getType:function(t){return null===t||t===!1?o.EMPTY:i.isValidElement(t)?\"function\"==typeof t.type?o.COMPOSITE:o.HOST:void r(\"26\",t)}});t.exports=o},function(t,e,n){\"use strict\";function r(t,e){return null==e?i(\"30\"):void 0,null==t?e:Array.isArray(t)?Array.isArray(e)?(t.push.apply(t,e),t):(t.push(e),t):Array.isArray(e)?[t].concat(e):[t,e]}var i=n(2);n(0);t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n){Array.isArray(t)?t.forEach(e,n):t&&e.call(n,t)}t.exports=r},function(t,e,n){\"use strict\";function r(t){for(var e;(e=t._renderedNodeType)===i.COMPOSITE;)t=t._renderedComponent;return e===i.HOST?t._renderedComponent:e===i.EMPTY?null:void 0}var i=n(164);t.exports=r},function(t,e,n){\"use strict\";function r(){return!o&&i.canUseDOM&&(o=\"textContent\"in document.documentElement?\"textContent\":\"innerText\"),o}var i=n(6),o=null;t.exports=r},function(t,e,n){\"use strict\";function r(t){if(t){var e=t.getName();if(e)return\" Check the render method of `\"+e+\"`.\"}return\"\"}function i(t){return\"function\"==typeof t&&\"undefined\"!=typeof t.prototype&&\"function\"==typeof t.prototype.mountComponent&&\"function\"==typeof t.prototype.receiveComponent}function o(t,e){var n;if(null===t||t===!1)n=s.create(o);else if(\"object\"==typeof t){var u=t,c=u.type;if(\"function\"!=typeof c&&\"string\"!=typeof c){var p=\"\";p+=r(u._owner),a(\"130\",null==c?c:typeof c,p)}\"string\"==typeof u.type?n=l.createInternalComponent(u):i(u.type)?(n=new u.type(u),n.getHostNode||(n.getHostNode=n.getNativeNode)):n=new f(u)}else\"string\"==typeof t||\"number\"==typeof t?n=l.createInstanceForText(t):a(\"131\",typeof t);return n._mountIndex=0,n._mountImage=null,n}var a=n(2),u=n(3),c=n(344),s=n(159),l=n(161),f=(n(391),n(0),n(1),function(t){this.construct(t)});u(f.prototype,c,{_instantiateReactComponent:o}),t.exports=o},function(t,e,n){\"use strict\";function r(t){var e=t&&t.nodeName&&t.nodeName.toLowerCase();return\"input\"===e?!!i[t.type]:\"textarea\"===e}var i={color:!0,date:!0,datetime:!0,\"datetime-local\":!0,email:!0,month:!0,number:!0,password:!0,range:!0,search:!0,tel:!0,text:!0,time:!0,url:!0,week:!0};t.exports=r},function(t,e,n){\"use strict\";var r=n(6),i=n(54),o=n(55),a=function(t,e){if(e){var n=t.firstChild;if(n&&n===t.lastChild&&3===n.nodeType)return void(n.nodeValue=e)}t.textContent=e};r.canUseDOM&&(\"textContent\"in document.documentElement||(a=function(t,e){return 3===t.nodeType?void(t.nodeValue=e):void o(t,i(e))})),t.exports=a},function(t,e,n){\"use strict\";function r(t,e){return t&&\"object\"==typeof t&&null!=t.key?s.escape(t.key):e.toString(36)}function i(t,e,n,o){var p=typeof t;if(\"undefined\"!==p&&\"boolean\"!==p||(t=null),null===t||\"string\"===p||\"number\"===p||\"object\"===p&&t.$$typeof===u)return n(o,t,\"\"===e?l+r(t,0):e),1;var h,d,v=0,g=\"\"===e?l:e+f;if(Array.isArray(t))for(var m=0;m<t.length;m++)h=t[m],d=g+r(h,m),v+=i(h,d,n,o);else{var y=c(t);if(y){var _,b=y.call(t);if(y!==t.entries)for(var x=0;!(_=b.next()).done;)h=_.value,d=g+r(h,x++),v+=i(h,d,n,o);else for(;!(_=b.next()).done;){var w=_.value;w&&(h=w[1],d=g+s.escape(w[0])+f+r(h,0),v+=i(h,d,n,o))}}else if(\"object\"===p){var C=\"\",M=String(t);a(\"31\",\"[object Object]\"===M?\"object with keys {\"+Object.keys(t).join(\", \")+\"}\":M,C)}}return v}function o(t,e,n){return null==t?0:i(t,\"\",e,n)}var a=n(2),u=(n(15),n(359)),c=n(390),s=(n(0),n(84)),l=(n(1),\".\"),f=\":\";t.exports=o},function(t,e,n){\"use strict\";function r(t){var e=Function.prototype.toString,n=Object.prototype.hasOwnProperty,r=RegExp(\"^\"+e.call(n).replace(/[\\\\^$.*+?()[\\]{}|]/g,\"\\\\$&\").replace(/hasOwnProperty|(function).*?(?=\\\\\\()| for .+?(?=\\\\\\])/g,\"$1.*?\")+\"$\");try{var i=e.call(t);return r.test(i)}catch(t){return!1}}function i(t){var e=s(t);if(e){var n=e.childIDs;l(t),n.forEach(i)}}function o(t,e,n){return\"\\n    in \"+(t||\"Unknown\")+(e?\" (at \"+e.fileName.replace(/^.*[\\\\\\/]/,\"\")+\":\"+e.lineNumber+\")\":n?\" (created by \"+n+\")\":\"\")}function a(t){return null==t?\"#empty\":\"string\"==typeof t||\"number\"==typeof t?\"#text\":\"string\"==typeof t.type?t.type:t.type.displayName||t.type.name||\"Unknown\"}function u(t){var e,n=k.getDisplayName(t),r=k.getElement(t),i=k.getOwnerID(t);return i&&(e=k.getDisplayName(i)),o(n,r&&r._source,e)}var c,s,l,f,p,h,d,v=n(28),g=n(15),m=(n(0),n(1),\"function\"==typeof Array.from&&\"function\"==typeof Map&&r(Map)&&null!=Map.prototype&&\"function\"==typeof Map.prototype.keys&&r(Map.prototype.keys)&&\"function\"==typeof Set&&r(Set)&&null!=Set.prototype&&\"function\"==typeof Set.prototype.keys&&r(Set.prototype.keys));if(m){var y=new Map,_=new Set;c=function(t,e){y.set(t,e)},s=function(t){return y.get(t)},l=function(t){y.delete(t)},f=function(){return Array.from(y.keys())},p=function(t){_.add(t)},h=function(t){_.delete(t)},d=function(){return Array.from(_.keys())}}else{var b={},x={},w=function(t){return\".\"+t},C=function(t){return parseInt(t.substr(1),10)};c=function(t,e){var n=w(t);b[n]=e},s=function(t){var e=w(t);return b[e]},l=function(t){var e=w(t);delete b[e]},f=function(){return Object.keys(b).map(C)},p=function(t){var e=w(t);x[e]=!0},h=function(t){var e=w(t);delete x[e]},d=function(){return Object.keys(x).map(C)}}var M=[],k={onSetChildren:function(t,e){var n=s(t);n?void 0:v(\"144\"),n.childIDs=e;for(var r=0;r<e.length;r++){var i=e[r],o=s(i);o?void 0:v(\"140\"),null==o.childIDs&&\"object\"==typeof o.element&&null!=o.element?v(\"141\"):void 0,o.isMounted?void 0:v(\"71\"),null==o.parentID&&(o.parentID=t),o.parentID!==t?v(\"142\",i,o.parentID,t):void 0}},onBeforeMountComponent:function(t,e,n){var r={element:e,parentID:n,text:null,childIDs:[],isMounted:!1,updateCount:0};c(t,r)},onBeforeUpdateComponent:function(t,e){var n=s(t);n&&n.isMounted&&(n.element=e)},onMountComponent:function(t){var e=s(t);e?void 0:v(\"144\"),e.isMounted=!0;var n=0===e.parentID;n&&p(t)},onUpdateComponent:function(t){var e=s(t);e&&e.isMounted&&e.updateCount++},onUnmountComponent:function(t){var e=s(t);if(e){e.isMounted=!1;var n=0===e.parentID;n&&h(t)}M.push(t)},purgeUnmountedComponents:function(){if(!k._preventPurging){for(var t=0;t<M.length;t++){var e=M[t];i(e)}M.length=0}},isMounted:function(t){var e=s(t);return!!e&&e.isMounted},getCurrentStackAddendum:function(t){var e=\"\";if(t){var n=a(t),r=t._owner;e+=o(n,t._source,r&&r.getName())}var i=g.current,u=i&&i._debugID;return e+=k.getStackAddendumByID(u)},getStackAddendumByID:function(t){for(var e=\"\";t;)e+=u(t),t=k.getParentID(t);return e},getChildIDs:function(t){var e=s(t);return e?e.childIDs:[]},getDisplayName:function(t){var e=k.getElement(t);return e?a(e):null},getElement:function(t){var e=s(t);return e?e.element:null},getOwnerID:function(t){var e=k.getElement(t);return e&&e._owner?e._owner._debugID:null},getParentID:function(t){var e=s(t);return e?e.parentID:null},getSource:function(t){var e=s(t),n=e?e.element:null,r=null!=n?n._source:null;return r},getText:function(t){var e=k.getElement(t);return\"string\"==typeof e?e:\"number\"==typeof e?\"\"+e:null},getUpdateCount:function(t){var e=s(t);return e?e.updateCount:0},getRootIDs:d,getRegisteredIDs:f};t.exports=k},function(t,e,n){\"use strict\";var r=\"function\"==typeof Symbol&&Symbol.for&&Symbol.for(\"react.element\")||60103;t.exports=r},function(t,e,n){\"use strict\";var r={};t.exports=r},function(t,e,n){\"use strict\";var r=!1;t.exports=r},function(t,e,n){\"use strict\";function r(t){var e=t&&(i&&t[i]||t[o]);if(\"function\"==typeof e)return e}var i=\"function\"==typeof Symbol&&Symbol.iterator,o=\"@@iterator\";t.exports=r},,function(t,e,n){\"use strict\";function r(t){return t&&t.__esModule?t:{default:t}}function i(t,e){if(!(t instanceof e))throw new TypeError(\"Cannot call a class as a function\")}function o(t,e){if(!t)throw new ReferenceError(\"this hasn't been initialised - super() hasn't been called\");return!e||\"object\"!=typeof e&&\"function\"!=typeof e?t:e}function a(t,e){if(\"function\"!=typeof e&&null!==e)throw new TypeError(\"Super expression must either be null or a function, not \"+typeof e);t.prototype=Object.create(e&&e.prototype,{constructor:{value:t,enumerable:!1,writable:!0,configurable:!0}}),e&&(Object.setPrototypeOf?Object.setPrototypeOf(t,e):t.__proto__=e)}Object.defineProperty(e,\"__esModule\",{value:!0});var u=\"function\"==typeof Symbol&&\"symbol\"==typeof Symbol.iterator?function(t){return typeof t}:function(t){return t&&\"function\"==typeof Symbol&&t.constructor===Symbol&&t!==Symbol.prototype?\"symbol\":typeof t},c=function(){function t(t,e){for(var n=0;n<e.length;n++){var r=e[n];r.enumerable=r.enumerable||!1,r.configurable=!0,\"value\"in r&&(r.writable=!0),Object.defineProperty(t,r.key,r)}}return function(e,n,r){return n&&t(e.prototype,n),r&&t(e,r),e}}(),s=n(41),l=r(s),f=n(129),p=n(64),h=n(30),d=n(77),v=n(112),g=n(134),m=n(10),y=n(39),_=n(56),b=r(_),x=function(t){function e(){i(this,e);var t=o(this,(e.__proto__||Object.getPrototypeOf(e)).call(this));return window.lastAdditiveForceArrayVisualizer=t,t.topOffset=28,t.leftOffset=80,t.height=350,t.effectFormat=(0,h.format)(\".2\"),t.redraw=(0,y.debounce)(function(){return t.draw()},200),t}return a(e,t),c(e,[{key:\"componentDidMount\",value:function(){var t=this;this.mainGroup=this.svg.append(\"g\"),this.onTopGroup=this.svg.append(\"g\"),this.xaxisElement=this.onTopGroup.append(\"g\").attr(\"transform\",\"translate(0,35)\").attr(\"class\",\"force-bar-array-xaxis\"),this.yaxisElement=this.onTopGroup.append(\"g\").attr(\"transform\",\"translate(0,35)\").attr(\"class\",\"force-bar-array-yaxis\"),this.hoverGroup1=this.svg.append(\"g\"),this.hoverGroup2=this.svg.append(\"g\"),this.baseValueTitle=this.svg.append(\"text\"),this.hoverLine=this.svg.append(\"line\"),this.hoverxOutline=this.svg.append(\"text\").attr(\"text-anchor\",\"middle\").attr(\"font-weight\",\"bold\").attr(\"fill\",\"#fff\").attr(\"stroke\",\"#fff\").attr(\"stroke-width\",\"6\").attr(\"font-size\",\"12px\"),this.hoverx=this.svg.append(\"text\").attr(\"text-anchor\",\"middle\").attr(\"font-weight\",\"bold\").attr(\"fill\",\"#000\").attr(\"font-size\",\"12px\"),this.hoverxTitle=this.svg.append(\"text\").attr(\"text-anchor\",\"middle\").attr(\"opacity\",.6).attr(\"font-size\",\"12px\"),this.hoveryOutline=this.svg.append(\"text\").attr(\"text-anchor\",\"end\").attr(\"font-weight\",\"bold\").attr(\"fill\",\"#fff\").attr(\"stroke\",\"#fff\").attr(\"stroke-width\",\"6\").attr(\"font-size\",\"12px\"),this.hovery=this.svg.append(\"text\").attr(\"text-anchor\",\"end\").attr(\"font-weight\",\"bold\").attr(\"fill\",\"#000\").attr(\"font-size\",\"12px\"),this.xlabel=this.wrapper.select(\".additive-force-array-xlabel\"),this.ylabel=this.wrapper.select(\".additive-force-array-ylabel\");var e=void 0;\"string\"==typeof this.props.plot_cmap?this.props.plot_cmap in b.default.colors?e=b.default.colors[this.props.plot_cmap]:(console.log(\"Invalid color map name, reverting to default.\"),e=b.default.colors.RdBu):Array.isArray(this.props.plot_cmap)&&(e=this.props.plot_cmap),this.colors=e.map(function(t){return(0,m.hsl)(t)}),this.brighterColors=[1.45,1.6].map(function(e,n){return t.colors[n].brighter(e)});var n=(0,h.format)(\",.4\");if(null!=this.props.ordering_keys&&null!=this.props.ordering_keys_time_format){var r=function(t){return\"object\"==(\"undefined\"==typeof t?\"undefined\":u(t))?this.formatTime(t):n(t)};this.parseTime=(0,d.timeParse)(this.props.ordering_keys_time_format),this.formatTime=(0,d.timeFormat)(this.props.ordering_keys_time_format),this.xtickFormat=r}else this.parseTime=null,this.formatTime=null,this.xtickFormat=n;this.xscale=(0,p.scaleLinear)(),this.xaxis=(0,v.axisBottom)().scale(this.xscale).tickSizeInner(4).tickSizeOuter(0).tickFormat(function(e){return t.xtickFormat(e)}).tickPadding(-18),this.ytickFormat=n,this.yscale=(0,p.scaleLinear)(),this.yaxis=(0,v.axisLeft)().scale(this.yscale).tickSizeInner(4).tickSizeOuter(0).tickFormat(function(e){return t.ytickFormat(t.invLinkFunction(e))}).tickPadding(2),this.xlabel.node().onchange=function(){return t.internalDraw()},this.ylabel.node().onchange=function(){return t.internalDraw()},this.svg.on(\"mousemove\",function(e){return t.mouseMoved(e)}),this.svg.on(\"click\",function(){return alert(\"This original index of the sample you clicked is \"+t.nearestExpIndex)}),this.svg.on(\"mouseout\",function(e){return t.mouseOut(e)}),window.addEventListener(\"resize\",this.redraw),window.setTimeout(this.redraw,50)}},{key:\"componentDidUpdate\",value:function(){this.draw()}},{key:\"mouseOut\",value:function(){this.hoverLine.attr(\"display\",\"none\"),this.hoverx.attr(\"display\",\"none\"),this.hoverxOutline.attr(\"display\",\"none\"),this.hoverxTitle.attr(\"display\",\"none\"),this.hovery.attr(\"display\",\"none\"),this.hoveryOutline.attr(\"display\",\"none\"),this.hoverGroup1.attr(\"display\",\"none\"),this.hoverGroup2.attr(\"display\",\"none\")}},{key:\"mouseMoved\",value:function(){var t=this,e=void 0,n=void 0;this.hoverLine.attr(\"display\",\"\"),this.hoverx.attr(\"display\",\"\"),this.hoverxOutline.attr(\"display\",\"\"),this.hoverxTitle.attr(\"display\",\"\"),this.hovery.attr(\"display\",\"\"),this.hoveryOutline.attr(\"display\",\"\"),this.hoverGroup1.attr(\"display\",\"\"),this.hoverGroup2.attr(\"display\",\"\");var r=(0,f.mouse)(this.svg.node())[0];if(this.props.explanations){for(e=0;e<this.currExplanations.length;++e)(!n||Math.abs(n.xmapScaled-r)>Math.abs(this.currExplanations[e].xmapScaled-r))&&(n=this.currExplanations[e]);this.nearestExpIndex=n.origInd,this.hoverLine.attr(\"x1\",n.xmapScaled).attr(\"x2\",n.xmapScaled).attr(\"y1\",0+this.topOffset).attr(\"y2\",this.height),this.hoverx.attr(\"x\",n.xmapScaled).attr(\"y\",this.topOffset-5).text(this.xtickFormat(n.xmap)),this.hoverxOutline.attr(\"x\",n.xmapScaled).attr(\"y\",this.topOffset-5).text(this.xtickFormat(n.xmap)),this.hoverxTitle.attr(\"x\",n.xmapScaled).attr(\"y\",this.topOffset-18).text(n.count>1?n.count+\" averaged samples\":\"\"),this.hovery.attr(\"x\",this.leftOffset-6).attr(\"y\",n.joinPointy).text(this.ytickFormat(this.invLinkFunction(n.joinPoint))),this.hoveryOutline.attr(\"x\",this.leftOffset-6).attr(\"y\",n.joinPointy).text(this.ytickFormat(this.invLinkFunction(n.joinPoint)));for(var i=[],o=void 0,a=void 0,u=this.currPosOrderedFeatures.length-1;u>=0;--u){var c=this.currPosOrderedFeatures[u],s=n.features[c];a=5+(s.posyTop+s.posyBottom)/2,(!o||a-o>=15)&&s.posyTop-s.posyBottom>=6&&(i.push(s),o=a)}var l=[];o=void 0;var p=!0,h=!1,d=void 0;try{for(var v,g=this.currNegOrderedFeatures[Symbol.iterator]();!(p=(v=g.next()).done);p=!0){var m=v.value,y=n.features[m];a=5+(y.negyTop+y.negyBottom)/2,(!o||o-a>=15)&&y.negyTop-y.negyBottom>=6&&(l.push(y),o=a)}}catch(t){h=!0,d=t}finally{try{!p&&g.return&&g.return()}finally{if(h)throw d}}var _=function(e){var r=\"\";return null!==e.value&&void 0!==e.value&&(r=\" = \"+(isNaN(e.value)?e.value:t.ytickFormat(e.value))),n.count>1?\"mean(\"+t.props.featureNames[e.ind]+\")\"+r:t.props.featureNames[e.ind]+r},b=this.hoverGroup1.selectAll(\".pos-values\").data(i);b.enter().append(\"text\").attr(\"class\",\"pos-values\").merge(b).attr(\"x\",n.xmapScaled+5).attr(\"y\",function(t){return 4+(t.posyTop+t.posyBottom)/2}).attr(\"text-anchor\",\"start\").attr(\"font-size\",12).attr(\"stroke\",\"#fff\").attr(\"fill\",\"#fff\").attr(\"stroke-width\",\"4\").attr(\"stroke-linejoin\",\"round\").attr(\"opacity\",1).text(_),b.exit().remove();var x=this.hoverGroup2.selectAll(\".pos-values\").data(i);x.enter().append(\"text\").attr(\"class\",\"pos-values\").merge(x).attr(\"x\",n.xmapScaled+5).attr(\"y\",function(t){return 4+(t.posyTop+t.posyBottom)/2}).attr(\"text-anchor\",\"start\").attr(\"font-size\",12).attr(\"fill\",this.colors[0]).text(_),x.exit().remove();var w=this.hoverGroup1.selectAll(\".neg-values\").data(l);w.enter().append(\"text\").attr(\"class\",\"neg-values\").merge(w).attr(\"x\",n.xmapScaled+5).attr(\"y\",function(t){return 4+(t.negyTop+t.negyBottom)/2}).attr(\"text-anchor\",\"start\").attr(\"font-size\",12).attr(\"stroke\",\"#fff\").attr(\"fill\",\"#fff\").attr(\"stroke-width\",\"4\").attr(\"stroke-linejoin\",\"round\").attr(\"opacity\",1).text(_),w.exit().remove();var C=this.hoverGroup2.selectAll(\".neg-values\").data(l);C.enter().append(\"text\").attr(\"class\",\"neg-values\").merge(C).attr(\"x\",n.xmapScaled+5).attr(\"y\",function(t){return 4+(t.negyTop+t.negyBottom)/2}).attr(\"text-anchor\",\"start\").attr(\"font-size\",12).attr(\"fill\",this.colors[1]).text(_),C.exit().remove()}}},{key:\"draw\",value:function(){var t=this;if(this.props.explanations&&0!==this.props.explanations.length){(0,y.each)(this.props.explanations,function(t,e){return t.origInd=e});var e={},n={},r={},i=!0,o=!1,a=void 0;try{for(var u,c=this.props.explanations[Symbol.iterator]();!(i=(u=c.next()).done);i=!0){var s=u.value;for(var l in s.features)void 0===e[l]&&(e[l]=0,n[l]=0,r[l]=0),s.features[l].effect>0?e[l]+=s.features[l].effect:n[l]-=s.features[l].effect,null!==s.features[l].value&&void 0!==s.features[l].value&&(r[l]+=1)}}catch(t){o=!0,a=t}finally{try{!i&&c.return&&c.return()}finally{if(o)throw a}}this.usedFeatures=(0,y.sortBy)((0,y.keys)(e),function(t){return-(e[t]+n[t])}),console.log(\"found \",this.usedFeatures.length,\" used features\"),this.posOrderedFeatures=(0,y.sortBy)(this.usedFeatures,function(t){return e[t]}),this.negOrderedFeatures=(0,y.sortBy)(this.usedFeatures,function(t){return-n[t]}),this.singleValueFeatures=(0,y.filter)(this.usedFeatures,function(t){return r[t]>0});var f=[\"sample order by similarity\",\"sample order by output value\",\"original sample ordering\"].concat(this.singleValueFeatures.map(function(e){return t.props.featureNames[e]}));null!=this.props.ordering_keys&&f.unshift(\"sample order by key\");var p=this.xlabel.selectAll(\"option\").data(f);p.enter().append(\"option\").merge(p).attr(\"value\",function(t){return t}).text(function(t){return t}),p.exit().remove();var h=this.props.outNames[0]?this.props.outNames[0]:\"model output value\";f=(0,y.map)(this.usedFeatures,function(e){return[t.props.featureNames[e],t.props.featureNames[e]+\" effects\"]}),f.unshift([\"model output value\",h]);var d=this.ylabel.selectAll(\"option\").data(f);d.enter().append(\"option\").merge(d).attr(\"value\",function(t){return t[0]}).text(function(t){return t[1]}),d.exit().remove(),this.ylabel.style(\"top\",(this.height-10-this.topOffset)/2+this.topOffset+\"px\").style(\"left\",10-this.ylabel.node().offsetWidth/2+\"px\"),this.internalDraw()}}},{key:\"internalDraw\",value:function(){var t=this,e=!0,n=!1,r=void 0;try{for(var i,o=this.props.explanations[Symbol.iterator]();!(e=(i=o.next()).done);e=!0){var a=i.value,c=!0,s=!1,l=void 0;try{for(var f,h=this.usedFeatures[Symbol.iterator]();!(c=(f=h.next()).done);c=!0){var d=f.value;a.features.hasOwnProperty(d)||(a.features[d]={effect:0,value:0}),a.features[d].ind=d}}catch(t){s=!0,l=t}finally{try{!c&&h.return&&h.return()}finally{if(s)throw l}}}}catch(t){n=!0,r=t}finally{try{!e&&o.return&&o.return()}finally{if(n)throw r}}var v=void 0,m=this.xlabel.node().value,_=\"sample order by key\"===m&&null!=this.props.ordering_keys_time_format;if(_?this.xscale=(0,p.scaleTime)():this.xscale=(0,p.scaleLinear)(),this.xaxis.scale(this.xscale),\"sample order by similarity\"===m)v=(0,y.sortBy)(this.props.explanations,function(t){return t.simIndex}),(0,y.each)(v,function(t,e){return t.xmap=e});else if(\"sample order by output value\"===m)v=(0,y.sortBy)(this.props.explanations,function(t){return-t.outValue}),(0,y.each)(v,function(t,e){return t.xmap=e});else if(\"original sample ordering\"===m)v=(0,y.sortBy)(this.props.explanations,function(t){return t.origInd}),(0,y.each)(v,function(t,e){return t.xmap=e});else if(\"sample order by key\"===m)v=this.props.explanations,_?(0,y.each)(v,function(e,n){return e.xmap=t.parseTime(t.props.ordering_keys[n])}):(0,y.each)(v,function(e,n){return e.xmap=t.props.ordering_keys[n]}),v=(0,y.sortBy)(v,function(t){return t.xmap});else{var b=function(){var e=(0,y.findKey)(t.props.featureNames,function(t){return t===m});(0,y.each)(t.props.explanations,function(t,n){return t.xmap=t.features[e].value});var n=(0,y.sortBy)(t.props.explanations,function(t){return t.xmap}),r=(0,y.map)(n,function(t){return t.xmap});if(\"string\"==typeof r[0])return alert(\"Ordering by category names is not yet supported.\"),{v:void 0};var i=(0,y.min)(r),o=(0,y.max)(r),a=(o-i)/100;v=[];for(var u=void 0,c=void 0,s=0;s<n.length;++s){var l=n[s];if(u&&!c&&l.xmap-u.xmap<=a||c&&l.xmap-c.xmap<=a){c||(c=(0,y.cloneDeep)(u),c.count=1);var f=!0,p=!1,h=void 0;try{for(var d,g=t.usedFeatures[Symbol.iterator]();!(f=(d=g.next()).done);f=!0){var _=d.value;c.features[_].effect+=l.features[_].effect,c.features[_].value+=l.features[_].value;\n",
              "}}catch(t){p=!0,h=t}finally{try{!f&&g.return&&g.return()}finally{if(p)throw h}}c.count+=1}else if(u)if(c){var b=!0,x=!1,w=void 0;try{for(var C,M=t.usedFeatures[Symbol.iterator]();!(b=(C=M.next()).done);b=!0){var k=C.value;c.features[k].effect/=c.count,c.features[k].value/=c.count}}catch(t){x=!0,w=t}finally{try{!b&&M.return&&M.return()}finally{if(x)throw w}}v.push(c),c=void 0}else v.push(u);u=l}u.xmap-v[v.length-1].xmap>a&&v.push(u)}();if(\"object\"===(\"undefined\"==typeof b?\"undefined\":u(b)))return b.v}this.currUsedFeatures=this.usedFeatures,this.currPosOrderedFeatures=this.posOrderedFeatures,this.currNegOrderedFeatures=this.negOrderedFeatures;var x=this.ylabel.node().value;if(\"model output value\"!==x){var w=v;v=(0,y.cloneDeep)(v);for(var C=(0,y.findKey)(this.props.featureNames,function(t){return t===x}),M=0;M<v.length;++M){var k=v[M].features[C];v[M].features={},v[M].features[C]=k,w[M].remapped_version=v[M]}this.currUsedFeatures=[C],this.currPosOrderedFeatures=[C],this.currNegOrderedFeatures=[C]}this.currExplanations=v,\"identity\"===this.props.link?this.invLinkFunction=function(e){return t.props.baseValue+e}:\"logit\"===this.props.link?this.invLinkFunction=function(e){return 1/(1+Math.exp(-(t.props.baseValue+e)))}:console.log(\"ERROR: Unrecognized link function: \",this.props.link),this.predValues=(0,y.map)(v,function(t){return(0,y.sum)((0,y.map)(t.features,function(t){return t.effect}))});var E=this.wrapper.node().offsetWidth;if(0==E)return setTimeout(function(){return t.draw(v)},500);this.svg.style(\"height\",this.height+\"px\"),this.svg.style(\"width\",E+\"px\");var T=(0,y.map)(v,function(t){return t.xmap});this.xscale.domain([(0,y.min)(T),(0,y.max)(T)]).range([this.leftOffset,E]).clamp(!0),this.xaxisElement.attr(\"transform\",\"translate(0,\"+this.topOffset+\")\").call(this.xaxis);for(var S=0;S<this.currExplanations.length;++S)this.currExplanations[S].xmapScaled=this.xscale(this.currExplanations[S].xmap);for(var P=v.length,N=0,A=0;A<P;++A){var O=v[A].features,I=(0,y.sum)((0,y.map)((0,y.filter)(O,function(t){return t.effect>0}),function(t){return t.effect}))||0,D=(0,y.sum)((0,y.map)((0,y.filter)(O,function(t){return t.effect<0}),function(t){return-t.effect}))||0;N=Math.max(N,2.2*Math.max(I,D))}this.yscale.domain([-N/2,N/2]).range([this.height-10,this.topOffset]),this.yaxisElement.attr(\"transform\",\"translate(\"+this.leftOffset+\",0)\").call(this.yaxis);for(var R=0;R<P;++R){var L=v[R].features,U=(0,y.sum)((0,y.map)((0,y.filter)(L,function(t){return t.effect<0}),function(t){return-t.effect}))||0,F=-U,j=void 0,B=!0,W=!1,V=void 0;try{for(var z,H=this.currPosOrderedFeatures[Symbol.iterator]();!(B=(z=H.next()).done);B=!0)j=z.value,L[j].posyTop=this.yscale(F),L[j].effect>0&&(F+=L[j].effect),L[j].posyBottom=this.yscale(F),L[j].ind=j}catch(t){W=!0,V=t}finally{try{!B&&H.return&&H.return()}finally{if(W)throw V}}var q=F,Y=!0,K=!1,G=void 0;try{for(var $,X=this.currNegOrderedFeatures[Symbol.iterator]();!(Y=($=X.next()).done);Y=!0)j=$.value,L[j].negyTop=this.yscale(F),L[j].effect<0&&(F-=L[j].effect),L[j].negyBottom=this.yscale(F)}catch(t){K=!0,G=t}finally{try{!Y&&X.return&&X.return()}finally{if(K)throw G}}v[R].joinPoint=q,v[R].joinPointy=this.yscale(q)}var Z=(0,g.line)().x(function(t){return t[0]}).y(function(t){return t[1]}),Q=this.mainGroup.selectAll(\".force-bar-array-area-pos\").data(this.currUsedFeatures);Q.enter().append(\"path\").attr(\"class\",\"force-bar-array-area-pos\").merge(Q).attr(\"d\",function(t){var e=(0,y.map)((0,y.range)(P),function(e){return[v[e].xmapScaled,v[e].features[t].posyTop]}),n=(0,y.map)((0,y.rangeRight)(P),function(e){return[v[e].xmapScaled,v[e].features[t].posyBottom]});return Z(e.concat(n))}).attr(\"fill\",this.colors[0]),Q.exit().remove();var J=this.mainGroup.selectAll(\".force-bar-array-area-neg\").data(this.currUsedFeatures);J.enter().append(\"path\").attr(\"class\",\"force-bar-array-area-neg\").merge(J).attr(\"d\",function(t){var e=(0,y.map)((0,y.range)(P),function(e){return[v[e].xmapScaled,v[e].features[t].negyTop]}),n=(0,y.map)((0,y.rangeRight)(P),function(e){return[v[e].xmapScaled,v[e].features[t].negyBottom]});return Z(e.concat(n))}).attr(\"fill\",this.colors[1]),J.exit().remove();var tt=this.mainGroup.selectAll(\".force-bar-array-divider-pos\").data(this.currUsedFeatures);tt.enter().append(\"path\").attr(\"class\",\"force-bar-array-divider-pos\").merge(tt).attr(\"d\",function(t){var e=(0,y.map)((0,y.range)(P),function(e){return[v[e].xmapScaled,v[e].features[t].posyBottom]});return Z(e)}).attr(\"fill\",\"none\").attr(\"stroke-width\",1).attr(\"stroke\",function(){return t.colors[0].brighter(1.2)}),tt.exit().remove();var et=this.mainGroup.selectAll(\".force-bar-array-divider-neg\").data(this.currUsedFeatures);et.enter().append(\"path\").attr(\"class\",\"force-bar-array-divider-neg\").merge(et).attr(\"d\",function(t){var e=(0,y.map)((0,y.range)(P),function(e){return[v[e].xmapScaled,v[e].features[t].negyTop]});return Z(e)}).attr(\"fill\",\"none\").attr(\"stroke-width\",1).attr(\"stroke\",function(){return t.colors[1].brighter(1.5)}),et.exit().remove();for(var nt=function(t,e,n,r,i){var o=void 0,a=void 0;\"pos\"===i?(o=t[n].features[e].posyBottom,a=t[n].features[e].posyTop):(o=t[n].features[e].negyBottom,a=t[n].features[e].negyTop);for(var u=void 0,c=void 0,s=n+1;s<=r;++s)\"pos\"===i?(u=t[s].features[e].posyBottom,c=t[s].features[e].posyTop):(u=t[s].features[e].negyBottom,c=t[s].features[e].negyTop),u>o&&(o=u),c<a&&(a=c);return{top:o,bottom:a}},rt=100,it=20,ot=100,at=[],ut=[\"pos\",\"neg\"],ct=0;ct<ut.length;ct++){var st=ut[ct],lt=!0,ft=!1,pt=void 0;try{for(var ht,dt=this.currUsedFeatures[Symbol.iterator]();!(lt=(ht=dt.next()).done);lt=!0)for(var vt=ht.value,gt=0,mt=0,yt=0,_t={top:0,bottom:0},bt=void 0;mt<P-1;){for(;yt<rt&&mt<P-1;)++mt,yt=v[mt].xmapScaled-v[gt].xmapScaled;for(_t=nt(v,vt,gt,mt,st);_t.bottom-_t.top<it&&gt<mt;)++gt,_t=nt(v,vt,gt,mt,st);if(yt=v[mt].xmapScaled-v[gt].xmapScaled,_t.bottom-_t.top>=it&&yt>=rt){for(;mt<P-1;){if(++mt,bt=nt(v,vt,gt,mt,st),!(bt.bottom-bt.top>it)){--mt;break}_t=bt}yt=v[mt].xmapScaled-v[gt].xmapScaled,at.push([(v[mt].xmapScaled+v[gt].xmapScaled)/2,(_t.top+_t.bottom)/2,this.props.featureNames[vt]]);var xt=v[mt].xmapScaled;for(gt=mt;xt+ot>v[gt].xmapScaled&&gt<P-1;)++gt;mt=gt}}}catch(t){ft=!0,pt=t}finally{try{!lt&&dt.return&&dt.return()}finally{if(ft)throw pt}}}var wt=this.onTopGroup.selectAll(\".force-bar-array-flabels\").data(at);wt.enter().append(\"text\").attr(\"class\",\"force-bar-array-flabels\").merge(wt).attr(\"x\",function(t){return t[0]}).attr(\"y\",function(t){return t[1]+4}).text(function(t){return t[2]}),wt.exit().remove()}},{key:\"componentWillUnmount\",value:function(){window.removeEventListener(\"resize\",this.redraw)}},{key:\"render\",value:function(){var t=this;return l.default.createElement(\"div\",{ref:function(e){return t.wrapper=(0,f.select)(e)},style:{textAlign:\"center\"}},l.default.createElement(\"style\",{dangerouslySetInnerHTML:{__html:\"\\n          .force-bar-array-wrapper {\\n            text-align: center;\\n          }\\n          .force-bar-array-xaxis path {\\n            fill: none;\\n            opacity: 0.4;\\n          }\\n          .force-bar-array-xaxis .domain {\\n            opacity: 0;\\n          }\\n          .force-bar-array-xaxis paths {\\n            display: none;\\n          }\\n          .force-bar-array-yaxis path {\\n            fill: none;\\n            opacity: 0.4;\\n          }\\n          .force-bar-array-yaxis paths {\\n            display: none;\\n          }\\n          .tick line {\\n            stroke: #000;\\n            stroke-width: 1px;\\n            opacity: 0.4;\\n          }\\n          .tick text {\\n            fill: #000;\\n            opacity: 0.5;\\n            font-size: 12px;\\n            padding: 0px;\\n          }\\n          .force-bar-array-flabels {\\n            font-size: 12px;\\n            fill: #fff;\\n            text-anchor: middle;\\n          }\\n          .additive-force-array-xlabel {\\n            background: none;\\n            border: 1px solid #ccc;\\n            opacity: 0.5;\\n            margin-bottom: 0px;\\n            font-size: 12px;\\n            font-family: arial;\\n            margin-left: 80px;\\n            max-width: 300px;\\n          }\\n          .additive-force-array-xlabel:focus {\\n            outline: none;\\n          }\\n          .additive-force-array-ylabel {\\n            position: relative;\\n            top: 0px;\\n            left: 0px;\\n            transform: rotate(-90deg);\\n            background: none;\\n            border: 1px solid #ccc;\\n            opacity: 0.5;\\n            margin-bottom: 0px;\\n            font-size: 12px;\\n            font-family: arial;\\n            max-width: 150px;\\n          }\\n          .additive-force-array-ylabel:focus {\\n            outline: none;\\n          }\\n          .additive-force-array-hoverLine {\\n            stroke-width: 1px;\\n            stroke: #fff;\\n            opacity: 1;\\n          }\"}}),l.default.createElement(\"select\",{className:\"additive-force-array-xlabel\"}),l.default.createElement(\"div\",{style:{height:\"0px\",textAlign:\"left\"}},l.default.createElement(\"select\",{className:\"additive-force-array-ylabel\"})),l.default.createElement(\"svg\",{ref:function(e){return t.svg=(0,f.select)(e)},style:{userSelect:\"none\",display:\"block\",fontFamily:\"arial\",sansSerif:!0}}))}}]),e}(l.default.Component);x.defaultProps={plot_cmap:\"RdBu\",ordering_keys:null,ordering_keys_time_format:null},e.default=x},function(t,e,n){\"use strict\";function r(t){return t&&t.__esModule?t:{default:t}}function i(t,e){if(!(t instanceof e))throw new TypeError(\"Cannot call a class as a function\")}function o(t,e){if(!t)throw new ReferenceError(\"this hasn't been initialised - super() hasn't been called\");return!e||\"object\"!=typeof e&&\"function\"!=typeof e?t:e}function a(t,e){if(\"function\"!=typeof e&&null!==e)throw new TypeError(\"Super expression must either be null or a function, not \"+typeof e);t.prototype=Object.create(e&&e.prototype,{constructor:{value:t,enumerable:!1,writable:!0,configurable:!0}}),e&&(Object.setPrototypeOf?Object.setPrototypeOf(t,e):t.__proto__=e)}Object.defineProperty(e,\"__esModule\",{value:!0});var u=function(){function t(t,e){for(var n=0;n<e.length;n++){var r=e[n];r.enumerable=r.enumerable||!1,r.configurable=!0,\"value\"in r&&(r.writable=!0),Object.defineProperty(t,r.key,r)}}return function(e,n,r){return n&&t(e.prototype,n),r&&t(e,r),e}}(),c=n(41),s=r(c),l=n(129),f=n(64),p=n(30),h=n(112),d=n(134),v=n(10),g=n(39),m=n(56),y=r(m),b=function(t){function e(){i(this,e);var t=o(this,(e.__proto__||Object.getPrototypeOf(e)).call(this));return window.lastAdditiveForceVisualizer=t,t.effectFormat=(0,p.format)(\".2\"),t.redraw=(0,g.debounce)(function(){return t.draw()},200),t}return a(e,t),u(e,[{key:\"componentDidMount\",value:function(){var t=this;this.mainGroup=this.svg.append(\"g\"),this.axisElement=this.mainGroup.append(\"g\").attr(\"transform\",\"translate(0,35)\").attr(\"class\",\"force-bar-axis\"),this.onTopGroup=this.svg.append(\"g\"),this.baseValueTitle=this.svg.append(\"text\"),this.joinPointLine=this.svg.append(\"line\"),this.joinPointLabelOutline=this.svg.append(\"text\"),this.joinPointLabel=this.svg.append(\"text\"),this.joinPointTitleLeft=this.svg.append(\"text\"),this.joinPointTitleLeftArrow=this.svg.append(\"text\"),this.joinPointTitle=this.svg.append(\"text\"),this.joinPointTitleRightArrow=this.svg.append(\"text\"),this.joinPointTitleRight=this.svg.append(\"text\"),this.hoverLabelBacking=this.svg.append(\"text\").attr(\"x\",10).attr(\"y\",20).attr(\"text-anchor\",\"middle\").attr(\"font-size\",12).attr(\"stroke\",\"#fff\").attr(\"fill\",\"#fff\").attr(\"stroke-width\",\"4\").attr(\"stroke-linejoin\",\"round\").text(\"\").on(\"mouseover\",function(){t.hoverLabel.attr(\"opacity\",1),t.hoverLabelBacking.attr(\"opacity\",1)}).on(\"mouseout\",function(){t.hoverLabel.attr(\"opacity\",0),t.hoverLabelBacking.attr(\"opacity\",0)}),this.hoverLabel=this.svg.append(\"text\").attr(\"x\",10).attr(\"y\",20).attr(\"text-anchor\",\"middle\").attr(\"font-size\",12).attr(\"fill\",\"#0f0\").text(\"\").on(\"mouseover\",function(){t.hoverLabel.attr(\"opacity\",1),t.hoverLabelBacking.attr(\"opacity\",1)}).on(\"mouseout\",function(){t.hoverLabel.attr(\"opacity\",0),t.hoverLabelBacking.attr(\"opacity\",0)});var e=void 0;\"string\"==typeof this.props.plot_cmap?this.props.plot_cmap in y.default.colors?e=y.default.colors[this.props.plot_cmap]:(console.log(\"Invalid color map name, reverting to default.\"),e=y.default.colors.RdBu):Array.isArray(this.props.plot_cmap)&&(e=this.props.plot_cmap),this.colors=e.map(function(t){return(0,v.hsl)(t)}),this.brighterColors=[1.45,1.6].map(function(e,n){return t.colors[n].brighter(e)}),this.colors.map(function(e,n){var r=t.svg.append(\"linearGradient\").attr(\"id\",\"linear-grad-\"+n).attr(\"x1\",\"0%\").attr(\"y1\",\"0%\").attr(\"x2\",\"0%\").attr(\"y2\",\"100%\");r.append(\"stop\").attr(\"offset\",\"0%\").attr(\"stop-color\",e).attr(\"stop-opacity\",.6),r.append(\"stop\").attr(\"offset\",\"100%\").attr(\"stop-color\",e).attr(\"stop-opacity\",0);var i=t.svg.append(\"linearGradient\").attr(\"id\",\"linear-backgrad-\"+n).attr(\"x1\",\"0%\").attr(\"y1\",\"0%\").attr(\"x2\",\"0%\").attr(\"y2\",\"100%\");i.append(\"stop\").attr(\"offset\",\"0%\").attr(\"stop-color\",e).attr(\"stop-opacity\",.5),i.append(\"stop\").attr(\"offset\",\"100%\").attr(\"stop-color\",e).attr(\"stop-opacity\",0)}),this.tickFormat=(0,p.format)(\",.4\"),this.scaleCentered=(0,f.scaleLinear)(),this.axis=(0,h.axisBottom)().scale(this.scaleCentered).tickSizeInner(4).tickSizeOuter(0).tickFormat(function(e){return t.tickFormat(t.invLinkFunction(e))}).tickPadding(-18),window.addEventListener(\"resize\",this.redraw),window.setTimeout(this.redraw,50)}},{key:\"componentDidUpdate\",value:function(){this.draw()}},{key:\"draw\",value:function(){var t=this;(0,g.each)(this.props.featureNames,function(e,n){t.props.features[n]&&(t.props.features[n].name=e)}),\"identity\"===this.props.link?this.invLinkFunction=function(e){return t.props.baseValue+e}:\"logit\"===this.props.link?this.invLinkFunction=function(e){return 1/(1+Math.exp(-(t.props.baseValue+e)))}:console.log(\"ERROR: Unrecognized link function: \",this.props.link);var e=this.svg.node().parentNode.offsetWidth;if(0==e)return setTimeout(function(){return t.draw(t.props)},500);this.svg.style(\"height\",\"150px\"),this.svg.style(\"width\",e+\"px\");var n=50,r=(0,g.sortBy)(this.props.features,function(t){return-1/(t.effect+1e-10)}),i=(0,g.sum)((0,g.map)(r,function(t){return Math.abs(t.effect)})),o=(0,g.sum)((0,g.map)((0,g.filter)(r,function(t){return t.effect>0}),function(t){return t.effect}))||0,a=(0,g.sum)((0,g.map)((0,g.filter)(r,function(t){return t.effect<0}),function(t){return-t.effect}))||0;this.domainSize=3*Math.max(o,a);var u=(0,f.scaleLinear)().domain([0,this.domainSize]).range([0,e]),c=e/2-u(a);this.scaleCentered.domain([-this.domainSize/2,this.domainSize/2]).range([0,e]).clamp(!0),this.axisElement.attr(\"transform\",\"translate(0,\"+n+\")\").call(this.axis);var s=0,l=void 0,h=void 0,v=void 0;for(l=0;l<r.length;++l)r[l].x=s,r[l].effect<0&&void 0===h&&(h=s,v=l),s+=Math.abs(r[l].effect);void 0===h&&(h=s,v=l);var m=(0,d.line)().x(function(t){return t[0]}).y(function(t){return t[1]}),y=function(e){return void 0!==e.value&&null!==e.value&&\"\"!==e.value?e.name+\" = \"+(isNaN(e.value)?e.value:t.tickFormat(e.value)):e.name};r=this.props.hideBars?[]:r;var b=this.mainGroup.selectAll(\".force-bar-blocks\").data(r);b.enter().append(\"path\").attr(\"class\",\"force-bar-blocks\").merge(b).attr(\"d\",function(t,e){var r=u(t.x)+c,i=u(Math.abs(t.effect)),o=t.effect<0?-4:4,a=o;return e===v&&(o=0),e===v-1&&(a=0),m([[r,6+n],[r+i,6+n],[r+i+a,14.5+n],[r+i,23+n],[r,23+n],[r+o,14.5+n]])}).attr(\"fill\",function(e){return e.effect>0?t.colors[0]:t.colors[1]}).on(\"mouseover\",function(e){if(u(Math.abs(e.effect))<u(i)/50||u(Math.abs(e.effect))<10){var r=u(e.x)+c,o=u(Math.abs(e.effect));t.hoverLabel.attr(\"opacity\",1).attr(\"x\",r+o/2).attr(\"y\",n+.5).attr(\"fill\",e.effect>0?t.colors[0]:t.colors[1]).text(y(e)),t.hoverLabelBacking.attr(\"opacity\",1).attr(\"x\",r+o/2).attr(\"y\",n+.5).text(y(e))}}).on(\"mouseout\",function(){t.hoverLabel.attr(\"opacity\",0),t.hoverLabelBacking.attr(\"opacity\",0)}),b.exit().remove();var x=_.filter(r,function(t){return u(Math.abs(t.effect))>u(i)/50&&u(Math.abs(t.effect))>10}),w=this.onTopGroup.selectAll(\".force-bar-labels\").data(x);if(w.exit().remove(),w=w.enter().append(\"text\").attr(\"class\",\"force-bar-labels\").attr(\"font-size\",\"12px\").attr(\"y\",48+n).merge(w).text(function(e){return void 0!==e.value&&null!==e.value&&\"\"!==e.value?e.name+\" = \"+(isNaN(e.value)?e.value:t.tickFormat(e.value)):e.name}).attr(\"fill\",function(e){return e.effect>0?t.colors[0]:t.colors[1]}).attr(\"stroke\",function(t){return t.textWidth=Math.max(this.getComputedTextLength(),u(Math.abs(t.effect))-10),t.innerTextWidth=this.getComputedTextLength(),\"none\"}),this.filteredData=x,r.length>0){s=h+u.invert(5);for(var C=v;C<r.length;++C)r[C].textx=s,s+=u.invert(r[C].textWidth+10);s=h-u.invert(5);for(var M=v-1;M>=0;--M)r[M].textx=s,s-=u.invert(r[M].textWidth+10)}w.attr(\"x\",function(t){return u(t.textx)+c+(t.effect>0?-t.textWidth/2:t.textWidth/2)}).attr(\"text-anchor\",\"middle\"),x=(0,g.filter)(x,function(n){return u(n.textx)+c>t.props.labelMargin&&u(n.textx)+c<e-t.props.labelMargin}),this.filteredData2=x;var k=x.slice(),E=(0,g.findIndex)(r,x[0])-1;E>=0&&k.unshift(r[E]);var T=this.mainGroup.selectAll(\".force-bar-labelBacking\").data(x);T.enter().append(\"path\").attr(\"class\",\"force-bar-labelBacking\").attr(\"stroke\",\"none\").attr(\"opacity\",.2).merge(T).attr(\"d\",function(t){return m([[u(t.x)+u(Math.abs(t.effect))+c,23+n],[(t.effect>0?u(t.textx):u(t.textx)+t.textWidth)+c+5,33+n],[(t.effect>0?u(t.textx):u(t.textx)+t.textWidth)+c+5,54+n],[(t.effect>0?u(t.textx)-t.textWidth:u(t.textx))+c-5,54+n],[(t.effect>0?u(t.textx)-t.textWidth:u(t.textx))+c-5,33+n],[u(t.x)+c,23+n]])}).attr(\"fill\",function(t){return\"url(#linear-backgrad-\"+(t.effect>0?0:1)+\")\"}),T.exit().remove();var S=this.mainGroup.selectAll(\".force-bar-labelDividers\").data(x.slice(0,-1));S.enter().append(\"rect\").attr(\"class\",\"force-bar-labelDividers\").attr(\"height\",\"21px\").attr(\"width\",\"1px\").attr(\"y\",33+n).merge(S).attr(\"x\",function(t){return(t.effect>0?u(t.textx):u(t.textx)+t.textWidth)+c+4.5}).attr(\"fill\",function(t){return\"url(#linear-grad-\"+(t.effect>0?0:1)+\")\"}),S.exit().remove();var P=this.mainGroup.selectAll(\".force-bar-labelLinks\").data(x.slice(0,-1));P.enter().append(\"line\").attr(\"class\",\"force-bar-labelLinks\").attr(\"y1\",23+n).attr(\"y2\",33+n).attr(\"stroke-opacity\",.5).attr(\"stroke-width\",1).merge(P).attr(\"x1\",function(t){return u(t.x)+u(Math.abs(t.effect))+c}).attr(\"x2\",function(t){return(t.effect>0?u(t.textx):u(t.textx)+t.textWidth)+c+5}).attr(\"stroke\",function(e){return e.effect>0?t.colors[0]:t.colors[1]}),P.exit().remove();var N=this.mainGroup.selectAll(\".force-bar-blockDividers\").data(r.slice(0,-1));N.enter().append(\"path\").attr(\"class\",\"force-bar-blockDividers\").attr(\"stroke-width\",2).attr(\"fill\",\"none\").merge(N).attr(\"d\",function(t){var e=u(t.x)+u(Math.abs(t.effect))+c;return m([[e,6+n],[e+(t.effect<0?-4:4),14.5+n],[e,23+n]])}).attr(\"stroke\",function(e,n){return v===n+1||Math.abs(e.effect)<1e-8?\"#rgba(0,0,0,0)\":e.effect>0?t.brighterColors[0]:t.brighterColors[1]}),N.exit().remove(),this.joinPointLine.attr(\"x1\",u(h)+c).attr(\"x2\",u(h)+c).attr(\"y1\",0+n).attr(\"y2\",6+n).attr(\"stroke\",\"#F2F2F2\").attr(\"stroke-width\",1).attr(\"opacity\",1),this.joinPointLabelOutline.attr(\"x\",u(h)+c).attr(\"y\",-5+n).attr(\"color\",\"#fff\").attr(\"text-anchor\",\"middle\").attr(\"font-weight\",\"bold\").attr(\"stroke\",\"#fff\").attr(\"stroke-width\",6).text((0,p.format)(\",.2f\")(this.invLinkFunction(h-a))).attr(\"opacity\",1),console.log(\"joinPoint\",h,c,n,a),this.joinPointLabel.attr(\"x\",u(h)+c).attr(\"y\",-5+n).attr(\"text-anchor\",\"middle\").attr(\"font-weight\",\"bold\").attr(\"fill\",\"#000\").text((0,p.format)(\",.2f\")(this.invLinkFunction(h-a))).attr(\"opacity\",1),this.joinPointTitle.attr(\"x\",u(h)+c).attr(\"y\",-22+n).attr(\"text-anchor\",\"middle\").attr(\"font-size\",\"12\").attr(\"fill\",\"#000\").text(this.props.outNames[0]).attr(\"opacity\",.5),this.props.hideBars||(this.joinPointTitleLeft.attr(\"x\",u(h)+c-16).attr(\"y\",-38+n).attr(\"text-anchor\",\"end\").attr(\"font-size\",\"13\").attr(\"fill\",this.colors[0]).text(\"higher\").attr(\"opacity\",1),this.joinPointTitleRight.attr(\"x\",u(h)+c+16).attr(\"y\",-38+n).attr(\"text-anchor\",\"start\").attr(\"font-size\",\"13\").attr(\"fill\",this.colors[1]).text(\"lower\").attr(\"opacity\",1),this.joinPointTitleLeftArrow.attr(\"x\",u(h)+c+7).attr(\"y\",-42+n).attr(\"text-anchor\",\"end\").attr(\"font-size\",\"13\").attr(\"fill\",this.colors[0]).text(\"→\").attr(\"opacity\",1),this.joinPointTitleRightArrow.attr(\"x\",u(h)+c-7).attr(\"y\",-36+n).attr(\"text-anchor\",\"start\").attr(\"font-size\",\"13\").attr(\"fill\",this.colors[1]).text(\"←\").attr(\"opacity\",1)),this.props.hideBaseValueLabel||this.baseValueTitle.attr(\"x\",this.scaleCentered(0)).attr(\"y\",-22+n).attr(\"text-anchor\",\"middle\").attr(\"font-size\",\"12\").attr(\"fill\",\"#000\").text(\"base value\").attr(\"opacity\",.5)}},{key:\"componentWillUnmount\",value:function(){window.removeEventListener(\"resize\",this.redraw)}},{key:\"render\",value:function(){var t=this;return s.default.createElement(\"svg\",{ref:function(e){return t.svg=(0,l.select)(e)},style:{userSelect:\"none\",display:\"block\",fontFamily:\"arial\",sansSerif:!0}},s.default.createElement(\"style\",{dangerouslySetInnerHTML:{__html:\"\\n          .force-bar-axis path {\\n            fill: none;\\n            opacity: 0.4;\\n          }\\n          .force-bar-axis paths {\\n            display: none;\\n          }\\n          .tick line {\\n            stroke: #000;\\n            stroke-width: 1px;\\n            opacity: 0.4;\\n          }\\n          .tick text {\\n            fill: #000;\\n            opacity: 0.5;\\n            font-size: 12px;\\n            padding: 0px;\\n          }\"}}))}}]),e}(s.default.Component);b.defaultProps={plot_cmap:\"RdBu\"},e.default=b},function(t,e,n){\"use strict\";function r(t){return t&&t.__esModule?t:{default:t}}function i(t,e){if(!(t instanceof e))throw new TypeError(\"Cannot call a class as a function\")}function o(t,e){if(!t)throw new ReferenceError(\"this hasn't been initialised - super() hasn't been called\");return!e||\"object\"!=typeof e&&\"function\"!=typeof e?t:e}function a(t,e){if(\"function\"!=typeof e&&null!==e)throw new TypeError(\"Super expression must either be null or a function, not \"+typeof e);t.prototype=Object.create(e&&e.prototype,{constructor:{value:t,enumerable:!1,writable:!0,configurable:!0}}),e&&(Object.setPrototypeOf?Object.setPrototypeOf(t,e):t.__proto__=e)}Object.defineProperty(e,\"__esModule\",{value:!0});var u=function(){function t(t,e){for(var n=0;n<e.length;n++){var r=e[n];r.enumerable=r.enumerable||!1,r.configurable=!0,\"value\"in r&&(r.writable=!0),Object.defineProperty(t,r.key,r)}}return function(e,n,r){return n&&t(e.prototype,n),r&&t(e,r),e}}(),c=n(41),s=r(c),l=n(64),f=n(30),p=n(39),h=n(56),d=r(h),v=function(t){function e(){i(this,e);var t=o(this,(e.__proto__||Object.getPrototypeOf(e)).call(this));return t.width=100,window.lastSimpleListInstance=t,t.effectFormat=(0,f.format)(\".2\"),t}return a(e,t),u(e,[{key:\"render\",value:function(){var t=this,e=void 0;\"string\"==typeof this.props.plot_cmap?this.props.plot_cmap in d.default.colors?e=d.default.colors[this.props.plot_cmap]:(console.log(\"Invalid color map name, reverting to default.\"),e=d.default.colors.RdBu):Array.isArray(this.props.plot_cmap)&&(e=this.props.plot_cmap),console.log(this.props.features,this.props.features),this.scale=(0,l.scaleLinear)().domain([0,(0,p.max)((0,p.map)(this.props.features,function(t){return Math.abs(t.effect)}))]).range([0,this.width]);var n=(0,p.reverse)((0,p.sortBy)(Object.keys(this.props.features),function(e){return Math.abs(t.props.features[e].effect)})),r=n.map(function(n){var r=t.props.features[n],i=t.props.featureNames[n],o={width:t.scale(Math.abs(r.effect)),height:\"20px\",background:r.effect<0?e[0]:e[1],display:\"inline-block\"},a=void 0,u=void 0,c={lineHeight:\"20px\",display:\"inline-block\",width:t.width+40,verticalAlign:\"top\",marginRight:\"5px\",textAlign:\"right\"},l={lineHeight:\"20px\",display:\"inline-block\",width:t.width+40,verticalAlign:\"top\",marginLeft:\"5px\"};return r.effect<0?(u=s.default.createElement(\"span\",{style:l},i),c.width=40+t.width-t.scale(Math.abs(r.effect)),c.textAlign=\"right\",c.color=\"#999\",c.fontSize=\"13px\",a=s.default.createElement(\"span\",{style:c},t.effectFormat(r.effect))):(c.textAlign=\"right\",a=s.default.createElement(\"span\",{style:c},i),l.width=40,l.textAlign=\"left\",l.color=\"#999\",l.fontSize=\"13px\",u=s.default.createElement(\"span\",{style:l},t.effectFormat(r.effect))),s.default.createElement(\"div\",{key:n,style:{marginTop:\"2px\"}},a,s.default.createElement(\"div\",{style:o}),u)});return s.default.createElement(\"span\",null,r)}}]),e}(s.default.Component);v.defaultProps={plot_cmap:\"RdBu\"},e.default=v},function(t,e,n){\"use strict\";t.exports=n(345)},function(t,e,n){var r=(n(0),n(398)),i=!1;t.exports=function(t){t=t||{};var e=t.shouldRejectClick||r;i=!0,n(22).injection.injectEventPluginsByName({TapEventPlugin:n(396)(e)})}},function(t,e,n){\"use strict\";e.a=function(t){return function(){return t}}},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";n(101),n(102),n(184),n(105),n(187),n(109),n(108)},function(t,e,n){\"use strict\";e.a=function(t){return t}},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";n(29)},function(t,e,n){\"use strict\";n(18),n(29),n(57)},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";n(18)},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";n(101),n(18),n(29),n(57)},function(t,e,n){\"use strict\";n(104)},function(t,e,n){\"use strict\";n(110)},function(t,e,n){\"use strict\";n.d(e,\"a\",function(){return r});var r=Array.prototype.slice},function(t,e,n){\"use strict\";function r(t,e,n){var r=t(n);return\"translate(\"+(isFinite(r)?r:e(n))+\",0)\"}function i(t,e,n){var r=t(n);return\"translate(0,\"+(isFinite(r)?r:e(n))+\")\"}function o(t){var e=t.bandwidth()/2;return t.round()&&(e=Math.round(e)),function(n){return t(n)+e}}function a(){return!this.__axis}function u(t,e){function n(n){var p,b=null==c?e.ticks?e.ticks.apply(e,u):e.domain():c,x=null==s?e.tickFormat?e.tickFormat.apply(e,u):h.a:s,w=Math.max(l,0)+_,C=t===d||t===g?r:i,M=e.range(),k=M[0]+.5,E=M[M.length-1]+.5,T=(e.bandwidth?o:h.a)(e.copy()),S=n.selection?n.selection():n,P=S.selectAll(\".domain\").data([null]),N=S.selectAll(\".tick\").data(b,e).order(),A=N.exit(),O=N.enter().append(\"g\").attr(\"class\",\"tick\"),I=N.select(\"line\"),D=N.select(\"text\"),R=t===d||t===m?-1:1,L=t===m||t===v?(p=\"x\",\"y\"):(p=\"y\",\"x\");P=P.merge(P.enter().insert(\"path\",\".tick\").attr(\"class\",\"domain\").attr(\"stroke\",\"#000\")),N=N.merge(O),I=I.merge(O.append(\"line\").attr(\"stroke\",\"#000\").attr(p+\"2\",R*l).attr(L+\"1\",.5).attr(L+\"2\",.5)),D=D.merge(O.append(\"text\").attr(\"fill\",\"#000\").attr(p,R*w).attr(L,.5).attr(\"dy\",t===d?\"0em\":t===g?\"0.71em\":\"0.32em\")),n!==S&&(P=P.transition(n),N=N.transition(n),I=I.transition(n),D=D.transition(n),A=A.transition(n).attr(\"opacity\",y).attr(\"transform\",function(t){return C(T,this.parentNode.__axis||T,t)}),O.attr(\"opacity\",y).attr(\"transform\",function(t){return C(this.parentNode.__axis||T,T,t)})),A.remove(),P.attr(\"d\",t===m||t==v?\"M\"+R*f+\",\"+k+\"H0.5V\"+E+\"H\"+R*f:\"M\"+k+\",\"+R*f+\"V0.5H\"+E+\"V\"+R*f),N.attr(\"opacity\",1).attr(\"transform\",function(t){return C(T,T,t)}),I.attr(p+\"2\",R*l),D.attr(p,R*w).text(x),S.filter(a).attr(\"fill\",\"none\").attr(\"font-size\",10).attr(\"font-family\",\"sans-serif\").attr(\"text-anchor\",t===v?\"start\":t===m?\"end\":\"middle\"),S.each(function(){this.__axis=T})}var u=[],c=null,s=null,l=6,f=6,_=3;return n.scale=function(t){return arguments.length?(e=t,n):e},n.ticks=function(){return u=p.a.call(arguments),n},n.tickArguments=function(t){return arguments.length?(u=null==t?[]:p.a.call(t),n):u.slice()},n.tickValues=function(t){return arguments.length?(c=null==t?null:p.a.call(t),n):c&&c.slice()},n.tickFormat=function(t){return arguments.length?(s=t,n):s},n.tickSize=function(t){return arguments.length?(l=f=+t,n):l},n.tickSizeInner=function(t){return arguments.length?(l=+t,n):l},n.tickSizeOuter=function(t){return arguments.length?(f=+t,n):f},n.tickPadding=function(t){return arguments.length?(_=+t,n):_},n}function c(t){return u(d,t)}function s(t){return u(v,t)}function l(t){return u(g,t)}function f(t){return u(m,t)}var p=n(200),h=n(202);e.a=c,e.b=s,e.c=l,e.d=f;var d=1,v=2,g=3,m=4,y=1e-6},function(t,e,n){\"use strict\";e.a=function(t){return t}},function(t,e,n){\"use strict\";var r=(n(206),n(207),n(58));n.d(e,\"a\",function(){return r.a});n(205),n(208),n(204)},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";n(58)},function(t,e,n){\"use strict\";function r(){}function i(t,e){var n=new r;if(t instanceof r)t.each(function(t){n.add(t)});else if(t){var i=-1,o=t.length;if(null==e)for(;++i<o;)n.add(t[i]);else for(;++i<o;)n.add(e(t[i],i,t))}return n}var o=n(58),a=o.a.prototype;r.prototype=i.prototype={constructor:r,has:a.has,add:function(t){return t+=\"\",this[o.b+t]=t,this},remove:a.remove,clear:a.clear,values:a.keys,size:a.size,empty:a.empty,each:a.each}},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";function r(t){if(t instanceof o)return new o(t.h,t.s,t.l,t.opacity);t instanceof u.d||(t=n.i(u.e)(t));var e=t.r/255,r=t.g/255,i=t.b/255,a=(g*i+d*e-v*r)/(g+d-v),s=i-a,l=(h*(r-a)-f*s)/p,m=Math.sqrt(l*l+s*s)/(h*a*(1-a)),y=m?Math.atan2(l,s)*c.a-120:NaN;return new o(y<0?y+360:y,m,a,t.opacity)}function i(t,e,n,i){return 1===arguments.length?r(t):new o(t,e,n,null==i?1:i)}function o(t,e,n,r){this.h=+t,this.s=+e,this.l=+n,this.opacity=+r}var a=n(60),u=n(59),c=n(113);e.a=i;var s=-.14861,l=1.78277,f=-.29227,p=-.90649,h=1.97294,d=h*p,v=h*l,g=l*f-p*s;n.i(a.a)(o,i,n.i(a.b)(u.f,{brighter:function(t){return t=null==t?u.g:Math.pow(u.g,t),new o(this.h,this.s,this.l*t,this.opacity)},darker:function(t){return t=null==t?u.h:Math.pow(u.h,t),new o(this.h,this.s,this.l*t,this.opacity)},rgb:function(){var t=isNaN(this.h)?0:(this.h+120)*c.b,e=+this.l,n=isNaN(this.s)?0:this.s*e*(1-e),r=Math.cos(t),i=Math.sin(t);return new u.d(255*(e+n*(s*r+l*i)),255*(e+n*(f*r+p*i)),255*(e+n*(h*r)),this.opacity)}}))},function(t,e,n){\"use strict\";function r(t){if(t instanceof o)return new o(t.l,t.a,t.b,t.opacity);if(t instanceof p){var e=t.h*v.b;return new o(t.l,Math.cos(e)*t.c,Math.sin(e)*t.c,t.opacity)}t instanceof d.d||(t=n.i(d.e)(t));var r=s(t.r),i=s(t.g),u=s(t.b),c=a((.4124564*r+.3575761*i+.1804375*u)/m),l=a((.2126729*r+.7151522*i+.072175*u)/y),f=a((.0193339*r+.119192*i+.9503041*u)/_);return new o(116*l-16,500*(c-l),200*(l-f),t.opacity)}function i(t,e,n,i){return 1===arguments.length?r(t):new o(t,e,n,null==i?1:i)}function o(t,e,n,r){this.l=+t,this.a=+e,this.b=+n,this.opacity=+r}function a(t){return t>C?Math.pow(t,1/3):t/w+b}function u(t){return t>x?t*t*t:w*(t-b)}function c(t){return 255*(t<=.0031308?12.92*t:1.055*Math.pow(t,1/2.4)-.055)}function s(t){return(t/=255)<=.04045?t/12.92:Math.pow((t+.055)/1.055,2.4)}function l(t){if(t instanceof p)return new p(t.h,t.c,t.l,t.opacity);t instanceof o||(t=r(t));var e=Math.atan2(t.b,t.a)*v.a;return new p(e<0?e+360:e,Math.sqrt(t.a*t.a+t.b*t.b),t.l,t.opacity)}function f(t,e,n,r){return 1===arguments.length?l(t):new p(t,e,n,null==r?1:r)}function p(t,e,n,r){this.h=+t,this.c=+e,this.l=+n,this.opacity=+r}var h=n(60),d=n(59),v=n(113);e.a=i,e.b=f;var g=18,m=.95047,y=1,_=1.08883,b=4/29,x=6/29,w=3*x*x,C=x*x*x;n.i(h.a)(o,i,n.i(h.b)(d.f,{brighter:function(t){return new o(this.l+g*(null==t?1:t),this.a,this.b,this.opacity)},darker:function(t){return new o(this.l-g*(null==t?1:t),this.a,this.b,this.opacity)},rgb:function(){var t=(this.l+16)/116,e=isNaN(this.a)?t:t+this.a/500,n=isNaN(this.b)?t:t-this.b/200;return t=y*u(t),e=m*u(e),n=_*u(n),new d.d(c(3.2404542*e-1.5371385*t-.4985314*n),c(-.969266*e+1.8760108*t+.041556*n),c(.0556434*e-.2040259*t+1.0572252*n),this.opacity)}})),n.i(h.a)(p,f,n.i(h.b)(d.f,{brighter:function(t){return new p(this.h,this.c,this.l+g*(null==t?1:t),this.opacity)},darker:function(t){return new p(this.h,this.c,this.l-g*(null==t?1:t),this.opacity)},rgb:function(){return r(this).rgb()}}))},function(t,e,n){\"use strict\";function r(t){return o=n.i(i.a)(t),a=o.format,u=o.formatPrefix,o}var i=n(117);n.d(e,\"b\",function(){return a}),n.d(e,\"c\",function(){\n",
              "return u}),e.a=r;var o,a,u;r({decimal:\".\",thousands:\",\",grouping:[3],currency:[\"$\",\"\"]})},function(t,e,n){\"use strict\";e.a=function(t,e){t=t.toPrecision(e);t:for(var n,r=t.length,i=1,o=-1;i<r;++i)switch(t[i]){case\".\":o=n=i;break;case\"0\":0===o&&(o=i),n=i;break;case\"e\":break t;default:o>0&&(o=0)}return o>0?t.slice(0,o)+t.slice(n+1):t}},function(t,e,n){\"use strict\";e.a=function(t,e){return function(n,r){for(var i=n.length,o=[],a=0,u=t[0],c=0;i>0&&u>0&&(c+u+1>r&&(u=Math.max(1,r-c)),o.push(n.substring(i-=u,i+u)),!((c+=u+1)>r));)u=t[a=(a+1)%t.length];return o.reverse().join(e)}}},function(t,e,n){\"use strict\";var r=n(61);e.a=function(t,e){var i=n.i(r.a)(t,e);if(!i)return t+\"\";var o=i[0],a=i[1];return a<0?\"0.\"+new Array(-a).join(\"0\")+o:o.length>a+1?o.slice(0,a+1)+\".\"+o.slice(a+1):o+new Array(a-o.length+2).join(\"0\")}},function(t,e,n){\"use strict\";var r=n(42);e.a=function(t){return Math.max(0,-n.i(r.a)(Math.abs(t)))}},function(t,e,n){\"use strict\";var r=n(42);e.a=function(t,e){return Math.max(0,3*Math.max(-8,Math.min(8,Math.floor(n.i(r.a)(e)/3)))-n.i(r.a)(Math.abs(t)))}},function(t,e,n){\"use strict\";var r=n(42);e.a=function(t,e){return t=Math.abs(t),e=Math.abs(e)-t,Math.max(0,n.i(r.a)(e)-n.i(r.a)(t))+1}},function(t,e,n){\"use strict\";function r(t){return function e(r){function a(e,a){var u=t((e=n.i(i.cubehelix)(e)).h,(a=n.i(i.cubehelix)(a)).h),c=n.i(o.a)(e.s,a.s),s=n.i(o.a)(e.l,a.l),l=n.i(o.a)(e.opacity,a.opacity);return function(t){return e.h=u(t),e.s=c(t),e.l=s(Math.pow(t,r)),e.opacity=l(t),e+\"\"}}return r=+r,a.gamma=e,a}(1)}var i=n(10),o=n(32);n.d(e,\"a\",function(){return a});var a=(r(o.b),r(o.a))},function(t,e,n){\"use strict\";function r(t){return function(e,r){var a=t((e=n.i(i.hcl)(e)).h,(r=n.i(i.hcl)(r)).h),u=n.i(o.a)(e.c,r.c),c=n.i(o.a)(e.l,r.l),s=n.i(o.a)(e.opacity,r.opacity);return function(t){return e.h=a(t),e.c=u(t),e.l=c(t),e.opacity=s(t),e+\"\"}}}var i=n(10),o=n(32);r(o.b),r(o.a)},function(t,e,n){\"use strict\";function r(t){return function(e,r){var a=t((e=n.i(i.hsl)(e)).h,(r=n.i(i.hsl)(r)).h),u=n.i(o.a)(e.s,r.s),c=n.i(o.a)(e.l,r.l),s=n.i(o.a)(e.opacity,r.opacity);return function(t){return e.h=a(t),e.s=u(t),e.l=c(t),e.opacity=s(t),e+\"\"}}}var i=n(10),o=n(32);r(o.b),r(o.a)},function(t,e,n){\"use strict\";n(10),n(32)},function(t,e,n){\"use strict\"},function(t,e,n){\"use strict\";e.a=function(t,e){return t=+t,e-=t,function(n){return Math.round(t+e*n)}}},function(t,e,n){\"use strict\";n.d(e,\"a\",function(){return i});var r=180/Math.PI,i={translateX:0,translateY:0,rotate:0,skewX:0,scaleX:1,scaleY:1};e.b=function(t,e,n,i,o,a){var u,c,s;return(u=Math.sqrt(t*t+e*e))&&(t/=u,e/=u),(s=t*n+e*i)&&(n-=t*s,i-=e*s),(c=Math.sqrt(n*n+i*i))&&(n/=c,i/=c,s/=c),t*i<e*n&&(t=-t,e=-e,s=-s,u=-u),{translateX:o,translateY:a,rotate:Math.atan2(e,t)*r,skewX:Math.atan(s)*r,scaleX:u,scaleY:c}}},function(t,e,n){\"use strict\";function r(t,e,r,o){function a(t){return t.length?t.pop()+\" \":\"\"}function u(t,o,a,u,c,s){if(t!==a||o!==u){var l=c.push(\"translate(\",null,e,null,r);s.push({i:l-4,x:n.i(i.a)(t,a)},{i:l-2,x:n.i(i.a)(o,u)})}else(a||u)&&c.push(\"translate(\"+a+e+u+r)}function c(t,e,r,u){t!==e?(t-e>180?e+=360:e-t>180&&(t+=360),u.push({i:r.push(a(r)+\"rotate(\",null,o)-2,x:n.i(i.a)(t,e)})):e&&r.push(a(r)+\"rotate(\"+e+o)}function s(t,e,r,u){t!==e?u.push({i:r.push(a(r)+\"skewX(\",null,o)-2,x:n.i(i.a)(t,e)}):e&&r.push(a(r)+\"skewX(\"+e+o)}function l(t,e,r,o,u,c){if(t!==r||e!==o){var s=u.push(a(u)+\"scale(\",null,\",\",null,\")\");c.push({i:s-4,x:n.i(i.a)(t,r)},{i:s-2,x:n.i(i.a)(e,o)})}else 1===r&&1===o||u.push(a(u)+\"scale(\"+r+\",\"+o+\")\")}return function(e,n){var r=[],i=[];return e=t(e),n=t(n),u(e.translateX,e.translateY,n.translateX,n.translateY,r,i),c(e.rotate,n.rotate,r,i),s(e.skewX,n.skewX,r,i),l(e.scaleX,e.scaleY,n.scaleX,n.scaleY,r,i),e=n=null,function(t){for(var e,n=-1,o=i.length;++n<o;)r[(e=i[n]).i]=e.x(t);return r.join(\"\")}}}var i=n(43),o=n(226);r(o.a,\"px, \",\"px)\",\"deg)\"),r(o.b,\", \",\")\",\")\")},function(t,e,n){\"use strict\";function r(t){return\"none\"===t?o.a:(a||(a=document.createElement(\"DIV\"),u=document.documentElement,c=document.defaultView),a.style.transform=t,t=c.getComputedStyle(u.appendChild(a),null).getPropertyValue(\"transform\"),u.removeChild(a),t=t.slice(7,-1).split(\",\"),n.i(o.b)(+t[0],+t[1],+t[2],+t[3],+t[4],+t[5]))}function i(t){return null==t?o.a:(s||(s=document.createElementNS(\"http://www.w3.org/2000/svg\",\"g\")),s.setAttribute(\"transform\",t),(t=s.transform.baseVal.consolidate())?(t=t.matrix,n.i(o.b)(t.a,t.b,t.c,t.d,t.e,t.f)):o.a)}var o=n(224);e.a=r,e.b=i;var a,u,c,s},function(t,e,n){\"use strict\";Math.SQRT2},function(t,e,n){\"use strict\";function r(){this._x0=this._y0=this._x1=this._y1=null,this._=\"\"}function i(){return new r}var o=Math.PI,a=2*o,u=1e-6,c=a-u;r.prototype=i.prototype={constructor:r,moveTo:function(t,e){this._+=\"M\"+(this._x0=this._x1=+t)+\",\"+(this._y0=this._y1=+e)},closePath:function(){null!==this._x1&&(this._x1=this._x0,this._y1=this._y0,this._+=\"Z\")},lineTo:function(t,e){this._+=\"L\"+(this._x1=+t)+\",\"+(this._y1=+e)},quadraticCurveTo:function(t,e,n,r){this._+=\"Q\"+ +t+\",\"+ +e+\",\"+(this._x1=+n)+\",\"+(this._y1=+r)},bezierCurveTo:function(t,e,n,r,i,o){this._+=\"C\"+ +t+\",\"+ +e+\",\"+ +n+\",\"+ +r+\",\"+(this._x1=+i)+\",\"+(this._y1=+o)},arcTo:function(t,e,n,r,i){t=+t,e=+e,n=+n,r=+r,i=+i;var a=this._x1,c=this._y1,s=n-t,l=r-e,f=a-t,p=c-e,h=f*f+p*p;if(i<0)throw new Error(\"negative radius: \"+i);if(null===this._x1)this._+=\"M\"+(this._x1=t)+\",\"+(this._y1=e);else if(h>u)if(Math.abs(p*s-l*f)>u&&i){var d=n-a,v=r-c,g=s*s+l*l,m=d*d+v*v,y=Math.sqrt(g),_=Math.sqrt(h),b=i*Math.tan((o-Math.acos((g+h-m)/(2*y*_)))/2),x=b/_,w=b/y;Math.abs(x-1)>u&&(this._+=\"L\"+(t+x*f)+\",\"+(e+x*p)),this._+=\"A\"+i+\",\"+i+\",0,0,\"+ +(p*d>f*v)+\",\"+(this._x1=t+w*s)+\",\"+(this._y1=e+w*l)}else this._+=\"L\"+(this._x1=t)+\",\"+(this._y1=e);else;},arc:function(t,e,n,r,i,s){t=+t,e=+e,n=+n;var l=n*Math.cos(r),f=n*Math.sin(r),p=t+l,h=e+f,d=1^s,v=s?r-i:i-r;if(n<0)throw new Error(\"negative radius: \"+n);null===this._x1?this._+=\"M\"+p+\",\"+h:(Math.abs(this._x1-p)>u||Math.abs(this._y1-h)>u)&&(this._+=\"L\"+p+\",\"+h),n&&(v>c?this._+=\"A\"+n+\",\"+n+\",0,1,\"+d+\",\"+(t-l)+\",\"+(e-f)+\"A\"+n+\",\"+n+\",0,1,\"+d+\",\"+(this._x1=p)+\",\"+(this._y1=h):(v<0&&(v=v%a+a),this._+=\"A\"+n+\",\"+n+\",0,\"+ +(v>=o)+\",\"+d+\",\"+(this._x1=t+n*Math.cos(i))+\",\"+(this._y1=e+n*Math.sin(i))))},rect:function(t,e,n,r){this._+=\"M\"+(this._x0=this._x1=+t)+\",\"+(this._y0=this._y1=+e)+\"h\"+ +n+\"v\"+ +r+\"h\"+-n+\"Z\"},toString:function(){return this._}},e.a=i},function(t,e,n){\"use strict\";function r(){function t(){var t=c().length,r=l[1]<l[0],o=l[r-0],u=l[1-r];e=(u-o)/Math.max(1,t-p+2*h),f&&(e=Math.floor(e)),o+=(u-o-e*(t-p))*d,i=e*(1-p),f&&(o=Math.round(o),i=Math.round(i));var v=n.i(a.g)(t).map(function(t){return o+e*t});return s(r?v.reverse():v)}var e,i,o=n.i(u.a)().unknown(void 0),c=o.domain,s=o.range,l=[0,1],f=!1,p=0,h=0,d=.5;return delete o.unknown,o.domain=function(e){return arguments.length?(c(e),t()):c()},o.range=function(e){return arguments.length?(l=[+e[0],+e[1]],t()):l.slice()},o.rangeRound=function(e){return l=[+e[0],+e[1]],f=!0,t()},o.bandwidth=function(){return i},o.step=function(){return e},o.round=function(e){return arguments.length?(f=!!e,t()):f},o.padding=function(e){return arguments.length?(p=h=Math.max(0,Math.min(1,e)),t()):p},o.paddingInner=function(e){return arguments.length?(p=Math.max(0,Math.min(1,e)),t()):p},o.paddingOuter=function(e){return arguments.length?(h=Math.max(0,Math.min(1,e)),t()):h},o.align=function(e){return arguments.length?(d=Math.max(0,Math.min(1,e)),t()):d},o.copy=function(){return r().domain(c()).range(l).round(f).paddingInner(p).paddingOuter(h).align(d)},t()}function i(t){var e=t.copy;return t.padding=t.paddingOuter,delete t.paddingInner,delete t.paddingOuter,t.copy=function(){return i(e())},t}function o(){return i(r().paddingInner(1))}var a=n(12),u=n(127);e.a=r,e.b=o},function(t,e,n){\"use strict\";var r=n(33);e.a=n.i(r.a)(\"1f77b4ff7f0e2ca02cd627289467bd8c564be377c27f7f7fbcbd2217becf\")},function(t,e,n){\"use strict\";var r=n(33);e.a=n.i(r.a)(\"1f77b4aec7e8ff7f0effbb782ca02c98df8ad62728ff98969467bdc5b0d58c564bc49c94e377c2f7b6d27f7f7fc7c7c7bcbd22dbdb8d17becf9edae5\")},function(t,e,n){\"use strict\";var r=n(33);e.a=n.i(r.a)(\"393b795254a36b6ecf9c9ede6379398ca252b5cf6bcedb9c8c6d31bd9e39e7ba52e7cb94843c39ad494ad6616be7969c7b4173a55194ce6dbdde9ed6\")},function(t,e,n){\"use strict\";var r=n(33);e.a=n.i(r.a)(\"3182bd6baed69ecae1c6dbefe6550dfd8d3cfdae6bfdd0a231a35474c476a1d99bc7e9c0756bb19e9ac8bcbddcdadaeb636363969696bdbdbdd9d9d9\")},function(t,e,n){\"use strict\";var r=n(10),i=n(31);e.a=n.i(i.d)(n.i(r.cubehelix)(300,.5,0),n.i(r.cubehelix)(-240,.5,1))},function(t,e,n){\"use strict\";function r(){function t(t){return+t}var e=[0,1];return t.invert=t,t.domain=t.range=function(n){return arguments.length?(e=i.a.call(n,a.a),t):e.slice()},t.copy=function(){return r().domain(e)},n.i(o.b)(t)}var i=n(16),o=n(34),a=n(126);e.a=r},function(t,e,n){\"use strict\";function r(t,e){return(e=Math.log(e/t))?function(n){return Math.log(n/t)/e}:n.i(p.a)(e)}function i(t,e){return t<0?function(n){return-Math.pow(-e,n)*Math.pow(-t,1-n)}:function(n){return Math.pow(e,n)*Math.pow(t,1-n)}}function o(t){return isFinite(t)?+(\"1e\"+t):t<0?0:t}function a(t){return 10===t?o:t===Math.E?Math.exp:function(e){return Math.pow(t,e)}}function u(t){return t===Math.E?Math.log:10===t&&Math.log10||2===t&&Math.log2||(t=Math.log(t),function(e){return Math.log(e)/t})}function c(t){return function(e){return-t(-e)}}function s(){function t(){return v=u(p),g=a(p),o()[0]<0&&(v=c(v),g=c(g)),e}var e=n.i(d.a)(r,i).domain([1,10]),o=e.domain,p=10,v=u(10),g=a(10);return e.base=function(e){return arguments.length?(p=+e,t()):p},e.domain=function(e){return arguments.length?(o(e),t()):o()},e.ticks=function(t){var e,r=o(),i=r[0],a=r[r.length-1];(e=a<i)&&(f=i,i=a,a=f);var u,c,s,f=v(i),h=v(a),d=null==t?10:+t,m=[];if(!(p%1)&&h-f<d){if(f=Math.round(f)-1,h=Math.round(h)+1,i>0){for(;f<h;++f)for(c=1,u=g(f);c<p;++c)if(s=u*c,!(s<i)){if(s>a)break;m.push(s)}}else for(;f<h;++f)for(c=p-1,u=g(f);c>=1;--c)if(s=u*c,!(s<i)){if(s>a)break;m.push(s)}}else m=n.i(l.a)(f,h,Math.min(h-f,d)).map(g);return e?m.reverse():m},e.tickFormat=function(t,r){if(null==r&&(r=10===p?\".0e\":\",\"),\"function\"!=typeof r&&(r=n.i(f.format)(r)),t===1/0)return r;null==t&&(t=10);var i=Math.max(1,p*t/e.ticks().length);return function(t){var e=t/g(Math.round(v(t)));return e*p<p-.5&&(e*=p),e<=i?r(t):\"\"}},e.nice=function(){return o(n.i(h.a)(o(),{floor:function(t){return g(Math.floor(v(t)))},ceil:function(t){return g(Math.ceil(v(t)))}}))},e.copy=function(){return n.i(d.c)(e,s().base(p))},e}var l=n(12),f=n(30),p=n(65),h=n(125),d=n(45);e.a=s},function(t,e,n){\"use strict\";function r(t,e){return t<0?-Math.pow(-t,e):Math.pow(t,e)}function i(){function t(t,e){return(e=r(e,o)-(t=r(t,o)))?function(n){return(r(n,o)-t)/e}:n.i(a.a)(e)}function e(t,e){return e=r(e,o)-(t=r(t,o)),function(n){return r(t+e*n,1/o)}}var o=1,s=n.i(c.a)(t,e),l=s.domain;return s.exponent=function(t){return arguments.length?(o=+t,l(l())):o},s.copy=function(){return n.i(c.c)(s,i().exponent(o))},n.i(u.b)(s)}function o(){return i().exponent(.5)}var a=n(65),u=n(34),c=n(45);e.a=i,e.b=o},function(t,e,n){\"use strict\";function r(){function t(){var t=0,r=Math.max(1,u.length);for(c=new Array(r-1);++t<r;)c[t-1]=n.i(i.e)(a,t/r);return e}function e(t){if(!isNaN(t=+t))return u[n.i(i.c)(c,t)]}var a=[],u=[],c=[];return e.invertExtent=function(t){var e=u.indexOf(t);return e<0?[NaN,NaN]:[e>0?c[e-1]:a[0],e<c.length?c[e]:a[a.length-1]]},e.domain=function(e){if(!arguments.length)return a.slice();a=[];for(var n,r=0,o=e.length;r<o;++r)n=e[r],null==n||isNaN(n=+n)||a.push(n);return a.sort(i.f),t()},e.range=function(e){return arguments.length?(u=o.b.call(e),t()):u.slice()},e.quantiles=function(){return c.slice()},e.copy=function(){return r().domain(a).range(u)},e}var i=n(12),o=n(16);e.a=r},function(t,e,n){\"use strict\";function r(){function t(t){if(t<=t)return f[n.i(i.c)(l,t,0,s)]}function e(){var e=-1;for(l=new Array(s);++e<s;)l[e]=((e+1)*c-(e-s)*u)/(s+1);return t}var u=0,c=1,s=1,l=[.5],f=[0,1];return t.domain=function(t){return arguments.length?(u=+t[0],c=+t[1],e()):[u,c]},t.range=function(t){return arguments.length?(s=(f=o.b.call(t)).length-1,e()):f.slice()},t.invertExtent=function(t){var e=f.indexOf(t);return e<0?[NaN,NaN]:e<1?[u,l[0]]:e>=s?[l[s-1],c]:[l[e-1],l[e]]},t.copy=function(){return r().domain([u,c]).range(f)},n.i(a.b)(t)}var i=n(12),o=n(16),a=n(34);e.a=r},function(t,e,n){\"use strict\";var r=n(10),i=n(31);n.d(e,\"b\",function(){return o}),n.d(e,\"c\",function(){return a});var o=n.i(i.d)(n.i(r.cubehelix)(-100,.75,.35),n.i(r.cubehelix)(80,1.5,.8)),a=n.i(i.d)(n.i(r.cubehelix)(260,.75,.35),n.i(r.cubehelix)(80,1.5,.8)),u=n.i(r.cubehelix)();e.a=function(t){(t<0||t>1)&&(t-=Math.floor(t));var e=Math.abs(t-.5);return u.h=360*t-100,u.s=1.5-1.5*e,u.l=.8-.9*e,u+\"\"}},function(t,e,n){\"use strict\";function r(t){function e(e){var n=(e-o)/(a-o);return t(u?Math.max(0,Math.min(1,n)):n)}var o=0,a=1,u=!1;return e.domain=function(t){return arguments.length?(o=+t[0],a=+t[1],e):[o,a]},e.clamp=function(t){return arguments.length?(u=!!t,e):u},e.interpolator=function(n){return arguments.length?(t=n,e):t},e.copy=function(){return r(t).domain([o,a]).clamp(u)},n.i(i.b)(e)}var i=n(34);e.a=r},function(t,e,n){\"use strict\";function r(){function t(t){if(t<=t)return a[n.i(i.c)(e,t,0,u)]}var e=[.5],a=[0,1],u=1;return t.domain=function(n){return arguments.length?(e=o.b.call(n),u=Math.min(e.length,a.length-1),t):e.slice()},t.range=function(n){return arguments.length?(a=o.b.call(n),u=Math.min(e.length,a.length-1),t):a.slice()},t.invertExtent=function(t){var n=a.indexOf(t);return[e[n-1],e[n]]},t.copy=function(){return r().domain(e).range(a)},t}var i=n(12),o=n(16);e.a=r},function(t,e,n){\"use strict\";var r=n(12),i=n(30);e.a=function(t,e,o){var a,u=t[0],c=t[t.length-1],s=n.i(r.b)(u,c,null==e?10:e);switch(o=n.i(i.formatSpecifier)(null==o?\",f\":o),o.type){case\"s\":var l=Math.max(Math.abs(u),Math.abs(c));return null!=o.precision||isNaN(a=n.i(i.precisionPrefix)(s,l))||(o.precision=a),n.i(i.formatPrefix)(o,l);case\"\":case\"e\":case\"g\":case\"p\":case\"r\":null!=o.precision||isNaN(a=n.i(i.precisionRound)(s,Math.max(Math.abs(u),Math.abs(c))))||(o.precision=a-(\"e\"===o.type));break;case\"f\":case\"%\":null!=o.precision||isNaN(a=n.i(i.precisionFixed)(s))||(o.precision=a-2*(\"%\"===o.type))}return n.i(i.format)(o)}},function(t,e,n){\"use strict\";var r=n(128),i=n(77),o=n(79);e.a=function(){return n.i(r.b)(o.f,o.i,o.j,o.e,o.k,o.l,o.m,o.n,i.utcFormat).domain([Date.UTC(2e3,0,1),Date.UTC(2e3,0,2)])}},function(t,e,n){\"use strict\";function r(t){var e=t.length;return function(n){return t[Math.max(0,Math.min(e-1,Math.floor(n*e)))]}}var i=n(33);n.d(e,\"b\",function(){return o}),n.d(e,\"c\",function(){return a}),n.d(e,\"d\",function(){return u}),e.a=r(n.i(i.a)(\"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\"));var o=r(n.i(i.a)(\"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\")),a=r(n.i(i.a)(\"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\")),u=r(n.i(i.a)(\"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\"))},function(t,e,n){\"use strict\";e.a=function(t){return function(){return t}}},function(t,e,n){\"use strict\";function r(){return new i}function i(){this._=\"@\"+(++o).toString(36)}e.a=r;var o=0;i.prototype=r.prototype={constructor:i,get:function(t){for(var e=this._;!(e in t);)if(!(t=t.parentNode))return;return t[e]},set:function(t,e){return t[this._]=e},remove:function(t){return this._ in t&&delete t[this._]},toString:function(){return this._}}},function(t,e,n){\"use strict\";var r=n(72),i=n(69);e.a=function(t){var e=n.i(r.a)();return e.changedTouches&&(e=e.changedTouches[0]),n.i(i.a)(t,e)}},function(t,e,n){\"use strict\";var r=n(7);e.a=function(t){return\"string\"==typeof t?new r.b([[document.querySelector(t)]],[document.documentElement]):new r.b([[t]],r.c)}},function(t,e,n){\"use strict\";var r=n(7);e.a=function(t){return\"string\"==typeof t?new r.b([document.querySelectorAll(t)],[document.documentElement]):new r.b([null==t?[]:t],r.c)}},function(t,e,n){\"use strict\";var r=n(66);e.a=function(t){var e=\"function\"==typeof t?t:n.i(r.a)(t);return this.select(function(){return this.appendChild(e.apply(this,arguments))})}},function(t,e,n){\"use strict\";function r(t){return function(){this.removeAttribute(t)}}function i(t){return function(){this.removeAttributeNS(t.space,t.local)}}function o(t,e){return function(){this.setAttribute(t,e)}}function a(t,e){return function(){this.setAttributeNS(t.space,t.local,e)}}function u(t,e){return function(){var n=e.apply(this,arguments);null==n?this.removeAttribute(t):this.setAttribute(t,n)}}function c(t,e){return function(){var n=e.apply(this,arguments);null==n?this.removeAttributeNS(t.space,t.local):this.setAttributeNS(t.space,t.local,n)}}var s=n(67);e.a=function(t,e){var l=n.i(s.a)(t);if(arguments.length<2){var f=this.node();return l.local?f.getAttributeNS(l.space,l.local):f.getAttribute(l)}return this.each((null==e?l.local?i:r:\"function\"==typeof e?l.local?c:u:l.local?a:o)(l,e))}},function(t,e,n){\"use strict\";e.a=function(){var t=arguments[0];return arguments[0]=this,t.apply(null,arguments),this}},function(t,e,n){\"use strict\";function r(t){return t.trim().split(/^|\\s+/)}function i(t){return t.classList||new o(t)}function o(t){this._node=t,this._names=r(t.getAttribute(\"class\")||\"\")}function a(t,e){for(var n=i(t),r=-1,o=e.length;++r<o;)n.add(e[r])}function u(t,e){for(var n=i(t),r=-1,o=e.length;++r<o;)n.remove(e[r])}function c(t){return function(){a(this,t)}}function s(t){return function(){u(this,t)}}function l(t,e){return function(){(e.apply(this,arguments)?a:u)(this,t)}}o.prototype={add:function(t){var e=this._names.indexOf(t);e<0&&(this._names.push(t),this._node.setAttribute(\"class\",this._names.join(\" \")))},remove:function(t){var e=this._names.indexOf(t);e>=0&&(this._names.splice(e,1),this._node.setAttribute(\"class\",this._names.join(\" \")))},contains:function(t){return this._names.indexOf(t)>=0}},e.a=function(t,e){var n=r(t+\"\");if(arguments.length<2){for(var o=i(this.node()),a=-1,u=n.length;++a<u;)if(!o.contains(n[a]))return!1;return!0}return this.each((\"function\"==typeof e?l:e?c:s)(n,e))}},function(t,e,n){\"use strict\";function r(t,e,n,r,i,o){for(var u,c=0,s=e.length,l=o.length;c<l;++c)(u=e[c])?(u.__data__=o[c],r[c]=u):n[c]=new a.b(t,o[c]);for(;c<s;++c)(u=e[c])&&(i[c]=u)}function i(t,e,n,r,i,o,u){var s,l,f,p={},h=e.length,d=o.length,v=new Array(h);for(s=0;s<h;++s)(l=e[s])&&(v[s]=f=c+u.call(l,l.__data__,s,e),f in p?i[s]=l:p[f]=l);for(s=0;s<d;++s)f=c+u.call(t,o[s],s,o),(l=p[f])?(r[s]=l,l.__data__=o[s],p[f]=null):n[s]=new a.b(t,o[s]);for(s=0;s<h;++s)(l=e[s])&&p[v[s]]===l&&(i[s]=l)}var o=n(7),a=n(131),u=n(246),c=\"$\";e.a=function(t,e){if(!t)return y=new Array(this.size()),d=-1,this.each(function(t){y[++d]=t}),y;var a=e?i:r,c=this._parents,s=this._groups;\"function\"!=typeof t&&(t=n.i(u.a)(t));for(var l=s.length,f=new Array(l),p=new Array(l),h=new Array(l),d=0;d<l;++d){var v=c[d],g=s[d],m=g.length,y=t.call(v,v&&v.__data__,d,c),_=y.length,b=p[d]=new Array(_),x=f[d]=new Array(_),w=h[d]=new Array(m);a(v,g,b,x,w,y,e);for(var C,M,k=0,E=0;k<_;++k)if(C=b[k]){for(k>=E&&(E=k+1);!(M=x[E])&&++E<_;);C._next=M||null}}return f=new o.b(f,c),f._enter=p,f._exit=h,f}},function(t,e,n){\"use strict\";e.a=function(t){return arguments.length?this.property(\"__data__\",t):this.node().__data__}},function(t,e,n){\"use strict\";function r(t,e,r){var i=n.i(a.a)(t),o=i.CustomEvent;o?o=new o(e,r):(o=i.document.createEvent(\"Event\"),r?(o.initEvent(e,r.bubbles,r.cancelable),o.detail=r.detail):o.initEvent(e,!1,!1)),t.dispatchEvent(o)}function i(t,e){return function(){return r(this,t,e)}}function o(t,e){return function(){return r(this,t,e.apply(this,arguments))}}var a=n(73);e.a=function(t,e){return this.each((\"function\"==typeof e?o:i)(t,e))}},function(t,e,n){\"use strict\";e.a=function(t){for(var e=this._groups,n=0,r=e.length;n<r;++n)for(var i,o=e[n],a=0,u=o.length;a<u;++a)(i=o[a])&&t.call(i,i.__data__,a,o);return this}},function(t,e,n){\"use strict\";e.a=function(){return!this.node()}},function(t,e,n){\"use strict\";var r=n(132),i=n(7);e.a=function(){return new i.b(this._exit||this._groups.map(r.a),this._parents)}},function(t,e,n){\"use strict\";var r=n(7),i=n(130);e.a=function(t){\"function\"!=typeof t&&(t=n.i(i.a)(t));for(var e=this._groups,o=e.length,a=new Array(o),u=0;u<o;++u)for(var c,s=e[u],l=s.length,f=a[u]=[],p=0;p<l;++p)(c=s[p])&&t.call(c,c.__data__,p,s)&&f.push(c);return new r.b(a,this._parents)}},function(t,e,n){\"use strict\";function r(){this.innerHTML=\"\"}function i(t){return function(){this.innerHTML=t}}function o(t){return function(){var e=t.apply(this,arguments);this.innerHTML=null==e?\"\":e}}e.a=function(t){return arguments.length?this.each(null==t?r:(\"function\"==typeof t?o:i)(t)):this.node().innerHTML}},function(t,e,n){\"use strict\";function r(){return null}var i=n(66),o=n(71);e.a=function(t,e){var a=\"function\"==typeof t?t:n.i(i.a)(t),u=null==e?r:\"function\"==typeof e?e:n.i(o.a)(e);return this.select(function(){return this.insertBefore(a.apply(this,arguments),u.apply(this,arguments)||null)})}},function(t,e,n){\"use strict\";function r(){this.previousSibling&&this.parentNode.insertBefore(this,this.parentNode.firstChild)}e.a=function(){return this.each(r)}},function(t,e,n){\"use strict\";var r=n(7);e.a=function(t){for(var e=this._groups,n=t._groups,i=e.length,o=n.length,a=Math.min(i,o),u=new Array(i),c=0;c<a;++c)for(var s,l=e[c],f=n[c],p=l.length,h=u[c]=new Array(p),d=0;d<p;++d)(s=l[d]||f[d])&&(h[d]=s);for(;c<i;++c)u[c]=e[c];return new r.b(u,this._parents)}},function(t,e,n){\"use strict\";e.a=function(){for(var t=this._groups,e=0,n=t.length;e<n;++e)for(var r=t[e],i=0,o=r.length;i<o;++i){var a=r[i];if(a)return a}return null}},function(t,e,n){\"use strict\";e.a=function(){var t=new Array(this.size()),e=-1;return this.each(function(){t[++e]=this}),t}},function(t,e,n){\"use strict\";e.a=function(){for(var t=this._groups,e=-1,n=t.length;++e<n;)for(var r,i=t[e],o=i.length-1,a=i[o];--o>=0;)(r=i[o])&&(a&&a!==r.nextSibling&&a.parentNode.insertBefore(r,a),a=r);return this}},function(t,e,n){\"use strict\";function r(t){return function(){delete this[t]}}function i(t,e){return function(){this[t]=e}}function o(t,e){return function(){var n=e.apply(this,arguments);null==n?delete this[t]:this[t]=n}}e.a=function(t,e){return arguments.length>1?this.each((null==e?r:\"function\"==typeof e?o:i)(t,e)):this.node()[t]}},function(t,e,n){\"use strict\";function r(){this.nextSibling&&this.parentNode.appendChild(this)}e.a=function(){return this.each(r)}},function(t,e,n){\"use strict\";function r(){var t=this.parentNode;t&&t.removeChild(this)}e.a=function(){return this.each(r)}},function(t,e,n){\"use strict\";var r=n(7),i=n(71);e.a=function(t){\"function\"!=typeof t&&(t=n.i(i.a)(t));for(var e=this._groups,o=e.length,a=new Array(o),u=0;u<o;++u)for(var c,s,l=e[u],f=l.length,p=a[u]=new Array(f),h=0;h<f;++h)(c=l[h])&&(s=t.call(c,c.__data__,h,l))&&(\"__data__\"in c&&(s.__data__=c.__data__),p[h]=s);return new r.b(a,this._parents)}},function(t,e,n){\"use strict\";var r=n(7),i=n(133);e.a=function(t){\"function\"!=typeof t&&(t=n.i(i.a)(t));for(var e=this._groups,o=e.length,a=[],u=[],c=0;c<o;++c)for(var s,l=e[c],f=l.length,p=0;p<f;++p)(s=l[p])&&(a.push(t.call(s,s.__data__,p,l)),u.push(s));return new r.b(a,u)}},function(t,e,n){\"use strict\";e.a=function(){var t=0;return this.each(function(){++t}),t}},function(t,e,n){\"use strict\";function r(t,e){return t<e?-1:t>e?1:t>=e?0:NaN}var i=n(7);e.a=function(t){function e(e,n){return e&&n?t(e.__data__,n.__data__):!e-!n}t||(t=r);for(var n=this._groups,o=n.length,a=new Array(o),u=0;u<o;++u){for(var c,s=n[u],l=s.length,f=a[u]=new Array(l),p=0;p<l;++p)(c=s[p])&&(f[p]=c);f.sort(e)}return new i.b(a,this._parents).order()}},function(t,e,n){\"use strict\";function r(t){return function(){this.style.removeProperty(t)}}function i(t,e,n){return function(){this.style.setProperty(t,e,n)}}function o(t,e,n){return function(){var r=e.apply(this,arguments);null==r?this.style.removeProperty(t):this.style.setProperty(t,r,n)}}var a=n(73);e.a=function(t,e,u){var c;return arguments.length>1?this.each((null==e?r:\"function\"==typeof e?o:i)(t,e,null==u?\"\":u)):n.i(a.a)(c=this.node()).getComputedStyle(c,null).getPropertyValue(t)}},function(t,e,n){\"use strict\";function r(){this.textContent=\"\"}function i(t){return function(){this.textContent=t}}function o(t){return function(){var e=t.apply(this,arguments);this.textContent=null==e?\"\":e}}e.a=function(t){return arguments.length?this.each(null==t?r:(\"function\"==typeof t?o:i)(t)):this.node().textContent}},function(t,e,n){\"use strict\";var r=n(72),i=n(69);e.a=function(t,e,o){arguments.length<3&&(o=e,e=n.i(r.a)().changedTouches);for(var a,u=0,c=e?e.length:0;u<c;++u)if((a=e[u]).identifier===o)return n.i(i.a)(t,a);return null}},function(t,e,n){\"use strict\";var r=n(72),i=n(69);e.a=function(t,e){null==e&&(e=n.i(r.a)().touches);for(var o=0,a=e?e.length:0,u=new Array(a);o<a;++o)u[o]=n.i(i.a)(t,e[o]);return u}},function(t,e,n){\"use strict\";function r(t){return t.innerRadius}function i(t){return t.outerRadius}function o(t){return t.startAngle}function a(t){return t.endAngle}function u(t){return t&&t.padAngle}function c(t){return t>=1?h.d:t<=-1?-h.d:Math.asin(t)}function s(t,e,n,r,i,o,a,u){var c=n-t,s=r-e,l=a-i,f=u-o,p=(l*(e-o)-f*(t-i))/(f*c-l*s);return[t+p*c,e+p*s]}function l(t,e,n,r,i,o,a){var u=t-n,c=e-r,s=(a?o:-o)/Math.sqrt(u*u+c*c),l=s*c,f=-s*u,p=t+l,h=e+f,d=n+l,v=r+f,g=(p+d)/2,m=(h+v)/2,y=d-p,_=v-h,b=y*y+_*_,x=i-o,w=p*v-d*h,C=(_<0?-1:1)*Math.sqrt(Math.max(0,x*x*b-w*w)),M=(w*_-y*C)/b,k=(-w*y-_*C)/b,E=(w*_+y*C)/b,T=(-w*y+_*C)/b,S=M-g,P=k-m,N=E-g,A=T-m;return S*S+P*P>N*N+A*A&&(M=E,k=T),{cx:M,cy:k,x01:-l,y01:-f,x11:M*(i/x-1),y11:k*(i/x-1)}}var f=n(44),p=n(19),h=n(35);e.a=function(){function t(){var t,r,i=+e.apply(this,arguments),o=+d.apply(this,arguments),a=m.apply(this,arguments)-h.d,u=y.apply(this,arguments)-h.d,p=Math.abs(u-a),x=u>a;if(b||(b=t=n.i(f.a)()),o<i&&(r=o,o=i,i=r),o>h.a)if(p>h.c-h.a)b.moveTo(o*Math.cos(a),o*Math.sin(a)),b.arc(0,0,o,a,u,!x),i>h.a&&(b.moveTo(i*Math.cos(u),i*Math.sin(u)),b.arc(0,0,i,u,a,x));else{var w,C,M=a,k=u,E=a,T=u,S=p,P=p,N=_.apply(this,arguments)/2,A=N>h.a&&(g?+g.apply(this,arguments):Math.sqrt(i*i+o*o)),O=Math.min(Math.abs(o-i)/2,+v.apply(this,arguments)),I=O,D=O;\n",
              "if(A>h.a){var R=c(A/i*Math.sin(N)),L=c(A/o*Math.sin(N));(S-=2*R)>h.a?(R*=x?1:-1,E+=R,T-=R):(S=0,E=T=(a+u)/2),(P-=2*L)>h.a?(L*=x?1:-1,M+=L,k-=L):(P=0,M=k=(a+u)/2)}var U=o*Math.cos(M),F=o*Math.sin(M),j=i*Math.cos(T),B=i*Math.sin(T);if(O>h.a){var W=o*Math.cos(k),V=o*Math.sin(k),z=i*Math.cos(E),H=i*Math.sin(E);if(p<h.b){var q=S>h.a?s(U,F,z,H,W,V,j,B):[j,B],Y=U-q[0],K=F-q[1],G=W-q[0],$=V-q[1],X=1/Math.sin(Math.acos((Y*G+K*$)/(Math.sqrt(Y*Y+K*K)*Math.sqrt(G*G+$*$)))/2),Z=Math.sqrt(q[0]*q[0]+q[1]*q[1]);I=Math.min(O,(i-Z)/(X-1)),D=Math.min(O,(o-Z)/(X+1))}}P>h.a?D>h.a?(w=l(z,H,U,F,o,D,x),C=l(W,V,j,B,o,D,x),b.moveTo(w.cx+w.x01,w.cy+w.y01),D<O?b.arc(w.cx,w.cy,D,Math.atan2(w.y01,w.x01),Math.atan2(C.y01,C.x01),!x):(b.arc(w.cx,w.cy,D,Math.atan2(w.y01,w.x01),Math.atan2(w.y11,w.x11),!x),b.arc(0,0,o,Math.atan2(w.cy+w.y11,w.cx+w.x11),Math.atan2(C.cy+C.y11,C.cx+C.x11),!x),b.arc(C.cx,C.cy,D,Math.atan2(C.y11,C.x11),Math.atan2(C.y01,C.x01),!x))):(b.moveTo(U,F),b.arc(0,0,o,M,k,!x)):b.moveTo(U,F),i>h.a&&S>h.a?I>h.a?(w=l(j,B,W,V,i,-I,x),C=l(U,F,z,H,i,-I,x),b.lineTo(w.cx+w.x01,w.cy+w.y01),I<O?b.arc(w.cx,w.cy,I,Math.atan2(w.y01,w.x01),Math.atan2(C.y01,C.x01),!x):(b.arc(w.cx,w.cy,I,Math.atan2(w.y01,w.x01),Math.atan2(w.y11,w.x11),!x),b.arc(0,0,i,Math.atan2(w.cy+w.y11,w.cx+w.x11),Math.atan2(C.cy+C.y11,C.cx+C.x11),x),b.arc(C.cx,C.cy,I,Math.atan2(C.y11,C.x11),Math.atan2(C.y01,C.x01),!x))):b.arc(0,0,i,T,E,x):b.lineTo(j,B)}else b.moveTo(0,0);if(b.closePath(),t)return b=null,t+\"\"||null}var e=r,d=i,v=n.i(p.a)(0),g=null,m=o,y=a,_=u,b=null;return t.centroid=function(){var t=(+e.apply(this,arguments)+ +d.apply(this,arguments))/2,n=(+m.apply(this,arguments)+ +y.apply(this,arguments))/2-h.b/2;return[Math.cos(n)*t,Math.sin(n)*t]},t.innerRadius=function(r){return arguments.length?(e=\"function\"==typeof r?r:n.i(p.a)(+r),t):e},t.outerRadius=function(e){return arguments.length?(d=\"function\"==typeof e?e:n.i(p.a)(+e),t):d},t.cornerRadius=function(e){return arguments.length?(v=\"function\"==typeof e?e:n.i(p.a)(+e),t):v},t.padRadius=function(e){return arguments.length?(g=null==e?null:\"function\"==typeof e?e:n.i(p.a)(+e),t):g},t.startAngle=function(e){return arguments.length?(m=\"function\"==typeof e?e:n.i(p.a)(+e),t):m},t.endAngle=function(e){return arguments.length?(y=\"function\"==typeof e?e:n.i(p.a)(+e),t):y},t.padAngle=function(e){return arguments.length?(_=\"function\"==typeof e?e:n.i(p.a)(+e),t):_},t.context=function(e){return arguments.length?(b=null==e?null:e,t):b},t}},function(t,e,n){\"use strict\";n.d(e,\"a\",function(){return r});var r=Array.prototype.slice},function(t,e,n){\"use strict\";function r(t){this._context=t}var i=n(49),o=n(46);r.prototype={areaStart:i.a,areaEnd:i.a,lineStart:function(){this._x0=this._x1=this._x2=this._x3=this._x4=this._y0=this._y1=this._y2=this._y3=this._y4=NaN,this._point=0},lineEnd:function(){switch(this._point){case 1:this._context.moveTo(this._x2,this._y2),this._context.closePath();break;case 2:this._context.moveTo((this._x2+2*this._x3)/3,(this._y2+2*this._y3)/3),this._context.lineTo((this._x3+2*this._x2)/3,(this._y3+2*this._y2)/3),this._context.closePath();break;case 3:this.point(this._x2,this._y2),this.point(this._x3,this._y3),this.point(this._x4,this._y4)}},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._x2=t,this._y2=e;break;case 1:this._point=2,this._x3=t,this._y3=e;break;case 2:this._point=3,this._x4=t,this._y4=e,this._context.moveTo((this._x0+4*this._x1+t)/6,(this._y0+4*this._y1+e)/6);break;default:n.i(o.c)(this,t,e)}this._x0=this._x1,this._x1=t,this._y0=this._y1,this._y1=e}},e.a=function(t){return new r(t)}},function(t,e,n){\"use strict\";function r(t){this._context=t}var i=n(46);r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._y0=this._y1=NaN,this._point=0},lineEnd:function(){(this._line||0!==this._line&&3===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1;break;case 1:this._point=2;break;case 2:this._point=3;var r=(this._x0+4*this._x1+t)/6,o=(this._y0+4*this._y1+e)/6;this._line?this._context.lineTo(r,o):this._context.moveTo(r,o);break;case 3:this._point=4;default:n.i(i.c)(this,t,e)}this._x0=this._x1,this._x1=t,this._y0=this._y1,this._y1=e}},e.a=function(t){return new r(t)}},function(t,e,n){\"use strict\";function r(t,e){this._basis=new i.b(t),this._beta=e}var i=n(46);r.prototype={lineStart:function(){this._x=[],this._y=[],this._basis.lineStart()},lineEnd:function(){var t=this._x,e=this._y,n=t.length-1;if(n>0)for(var r,i=t[0],o=e[0],a=t[n]-i,u=e[n]-o,c=-1;++c<=n;)r=c/n,this._basis.point(this._beta*t[c]+(1-this._beta)*(i+r*a),this._beta*e[c]+(1-this._beta)*(o+r*u));this._x=this._y=null,this._basis.lineEnd()},point:function(t,e){this._x.push(+t),this._y.push(+e)}},e.a=function t(e){function n(t){return 1===e?new i.b(t):new r(t,e)}return n.beta=function(e){return t(+e)},n}(.85)},function(t,e,n){\"use strict\";function r(t,e){this._context=t,this._alpha=e}var i=n(136),o=n(49),a=n(74);r.prototype={areaStart:o.a,areaEnd:o.a,lineStart:function(){this._x0=this._x1=this._x2=this._x3=this._x4=this._x5=this._y0=this._y1=this._y2=this._y3=this._y4=this._y5=NaN,this._l01_a=this._l12_a=this._l23_a=this._l01_2a=this._l12_2a=this._l23_2a=this._point=0},lineEnd:function(){switch(this._point){case 1:this._context.moveTo(this._x3,this._y3),this._context.closePath();break;case 2:this._context.lineTo(this._x3,this._y3),this._context.closePath();break;case 3:this.point(this._x3,this._y3),this.point(this._x4,this._y4),this.point(this._x5,this._y5)}},point:function(t,e){if(t=+t,e=+e,this._point){var r=this._x2-t,i=this._y2-e;this._l23_a=Math.sqrt(this._l23_2a=Math.pow(r*r+i*i,this._alpha))}switch(this._point){case 0:this._point=1,this._x3=t,this._y3=e;break;case 1:this._point=2,this._context.moveTo(this._x4=t,this._y4=e);break;case 2:this._point=3,this._x5=t,this._y5=e;break;default:n.i(a.b)(this,t,e)}this._l01_a=this._l12_a,this._l12_a=this._l23_a,this._l01_2a=this._l12_2a,this._l12_2a=this._l23_2a,this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return e?new r(t,e):new i.b(t,0)}return n.alpha=function(e){return t(+e)},n}(.5)},function(t,e,n){\"use strict\";function r(t,e){this._context=t,this._alpha=e}var i=n(137),o=n(74);r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._x2=this._y0=this._y1=this._y2=NaN,this._l01_a=this._l12_a=this._l23_a=this._l01_2a=this._l12_2a=this._l23_2a=this._point=0},lineEnd:function(){(this._line||0!==this._line&&3===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){if(t=+t,e=+e,this._point){var r=this._x2-t,i=this._y2-e;this._l23_a=Math.sqrt(this._l23_2a=Math.pow(r*r+i*i,this._alpha))}switch(this._point){case 0:this._point=1;break;case 1:this._point=2;break;case 2:this._point=3,this._line?this._context.lineTo(this._x2,this._y2):this._context.moveTo(this._x2,this._y2);break;case 3:this._point=4;default:n.i(o.b)(this,t,e)}this._l01_a=this._l12_a,this._l12_a=this._l23_a,this._l01_2a=this._l12_2a,this._l12_2a=this._l23_2a,this._x0=this._x1,this._x1=this._x2,this._x2=t,this._y0=this._y1,this._y1=this._y2,this._y2=e}},e.a=function t(e){function n(t){return e?new r(t,e):new i.b(t,0)}return n.alpha=function(e){return t(+e)},n}(.5)},function(t,e,n){\"use strict\";function r(t){this._context=t}var i=n(49);r.prototype={areaStart:i.a,areaEnd:i.a,lineStart:function(){this._point=0},lineEnd:function(){this._point&&this._context.closePath()},point:function(t,e){t=+t,e=+e,this._point?this._context.lineTo(t,e):(this._point=1,this._context.moveTo(t,e))}},e.a=function(t){return new r(t)}},function(t,e,n){\"use strict\";function r(t){return t<0?-1:1}function i(t,e,n){var i=t._x1-t._x0,o=e-t._x1,a=(t._y1-t._y0)/(i||o<0&&-0),u=(n-t._y1)/(o||i<0&&-0),c=(a*o+u*i)/(i+o);return(r(a)+r(u))*Math.min(Math.abs(a),Math.abs(u),.5*Math.abs(c))||0}function o(t,e){var n=t._x1-t._x0;return n?(3*(t._y1-t._y0)/n-e)/2:e}function a(t,e,n){var r=t._x0,i=t._y0,o=t._x1,a=t._y1,u=(o-r)/3;t._context.bezierCurveTo(r+u,i+u*e,o-u,a-u*n,o,a)}function u(t){this._context=t}function c(t){this._context=new s(t)}function s(t){this._context=t}function l(t){return new u(t)}function f(t){return new c(t)}e.a=l,e.b=f,u.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x0=this._x1=this._y0=this._y1=this._t0=NaN,this._point=0},lineEnd:function(){switch(this._point){case 2:this._context.lineTo(this._x1,this._y1);break;case 3:a(this,this._t0,o(this,this._t0))}(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line=1-this._line},point:function(t,e){var n=NaN;if(t=+t,e=+e,t!==this._x1||e!==this._y1){switch(this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2;break;case 2:this._point=3,a(this,o(this,n=i(this,t,e)),n);break;default:a(this,this._t0,n=i(this,t,e))}this._x0=this._x1,this._x1=t,this._y0=this._y1,this._y1=e,this._t0=n}}},(c.prototype=Object.create(u.prototype)).point=function(t,e){u.prototype.point.call(this,e,t)},s.prototype={moveTo:function(t,e){this._context.moveTo(e,t)},closePath:function(){this._context.closePath()},lineTo:function(t,e){this._context.lineTo(e,t)},bezierCurveTo:function(t,e,n,r,i,o){this._context.bezierCurveTo(e,t,r,n,o,i)}}},function(t,e,n){\"use strict\";function r(t){this._context=t}function i(t){var e,n,r=t.length-1,i=new Array(r),o=new Array(r),a=new Array(r);for(i[0]=0,o[0]=2,a[0]=t[0]+2*t[1],e=1;e<r-1;++e)i[e]=1,o[e]=4,a[e]=4*t[e]+2*t[e+1];for(i[r-1]=2,o[r-1]=7,a[r-1]=8*t[r-1]+t[r],e=1;e<r;++e)n=i[e]/o[e-1],o[e]-=n,a[e]-=n*a[e-1];for(i[r-1]=a[r-1]/o[r-1],e=r-2;e>=0;--e)i[e]=(a[e]-i[e+1])/o[e];for(o[r-1]=(t[r]+i[r-1])/2,e=0;e<r-1;++e)o[e]=2*t[e+1]-i[e+1];return[i,o]}r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x=[],this._y=[]},lineEnd:function(){var t=this._x,e=this._y,n=t.length;if(n)if(this._line?this._context.lineTo(t[0],e[0]):this._context.moveTo(t[0],e[0]),2===n)this._context.lineTo(t[1],e[1]);else for(var r=i(t),o=i(e),a=0,u=1;u<n;++a,++u)this._context.bezierCurveTo(r[0][a],o[0][a],r[1][a],o[1][a],t[u],e[u]);(this._line||0!==this._line&&1===n)&&this._context.closePath(),this._line=1-this._line,this._x=this._y=null},point:function(t,e){this._x.push(+t),this._y.push(+e)}},e.a=function(t){return new r(t)}},function(t,e,n){\"use strict\";function r(t,e){this._context=t,this._t=e}function i(t){return new r(t,0)}function o(t){return new r(t,1)}e.c=i,e.b=o,r.prototype={areaStart:function(){this._line=0},areaEnd:function(){this._line=NaN},lineStart:function(){this._x=this._y=NaN,this._point=0},lineEnd:function(){0<this._t&&this._t<1&&2===this._point&&this._context.lineTo(this._x,this._y),(this._line||0!==this._line&&1===this._point)&&this._context.closePath(),this._line>=0&&(this._t=1-this._t,this._line=1-this._line)},point:function(t,e){switch(t=+t,e=+e,this._point){case 0:this._point=1,this._line?this._context.lineTo(t,e):this._context.moveTo(t,e);break;case 1:this._point=2;default:if(this._t<=0)this._context.lineTo(this._x,e),this._context.lineTo(t,e);else{var n=this._x*(1-this._t)+t*this._t;this._context.lineTo(n,this._y),this._context.lineTo(n,e)}}this._x=t,this._y=e}},e.a=function(t){return new r(t,.5)}},function(t,e,n){\"use strict\";e.a=function(t,e){return e<t?-1:e>t?1:e>=t?0:NaN}},function(t,e,n){\"use strict\";e.a=function(t){return t}},function(t,e,n){\"use strict\";var r=n(36);e.a=function(t,e){if((o=t.length)>0){for(var i,o,a,u=0,c=t[0].length;u<c;++u){for(a=i=0;i<o;++i)a+=t[i][u][1]||0;if(a)for(i=0;i<o;++i)t[i][u][1]/=a}n.i(r.a)(t,e)}}},function(t,e,n){\"use strict\";var r=n(36);e.a=function(t,e){if((i=t.length)>0){for(var i,o=0,a=t[e[0]],u=a.length;o<u;++o){for(var c=0,s=0;c<i;++c)s+=t[c][o][1]||0;a[o][1]+=a[o][0]=-s/2}n.i(r.a)(t,e)}}},function(t,e,n){\"use strict\";var r=n(36);e.a=function(t,e){if((a=t.length)>0&&(o=(i=t[e[0]]).length)>0){for(var i,o,a,u=0,c=1;c<o;++c){for(var s=0,l=0,f=0;s<a;++s){for(var p=t[e[s]],h=p[c][1]||0,d=p[c-1][1]||0,v=(h-d)/2,g=0;g<s;++g){var m=t[e[g]],y=m[c][1]||0,_=m[c-1][1]||0;v+=y-_}l+=h,f+=v*h}i[c-1][1]+=i[c-1][0]=u,l&&(u-=f/l)}i[c-1][1]+=i[c-1][0]=u,n.i(r.a)(t,e)}}},function(t,e,n){\"use strict\";var r=n(76);e.a=function(t){return n.i(r.a)(t).reverse()}},function(t,e,n){\"use strict\";var r=n(37),i=n(76);e.a=function(t){var e,o,a=t.length,u=t.map(i.b),c=n.i(r.a)(t).sort(function(t,e){return u[e]-u[t]}),s=0,l=0,f=[],p=[];for(e=0;e<a;++e)o=c[e],s<l?(s+=u[o],f.push(o)):(l+=u[o],p.push(o));return p.reverse().concat(f)}},function(t,e,n){\"use strict\";var r=n(37);e.a=function(t){return n.i(r.a)(t).reverse()}},function(t,e,n){\"use strict\";var r=n(19),i=n(291),o=n(292),a=n(35);e.a=function(){function t(t){var n,r,i,o,p,h=t.length,d=0,v=new Array(h),g=new Array(h),m=+s.apply(this,arguments),y=Math.min(a.c,Math.max(-a.c,l.apply(this,arguments)-m)),_=Math.min(Math.abs(y)/h,f.apply(this,arguments)),b=_*(y<0?-1:1);for(n=0;n<h;++n)(p=g[v[n]=n]=+e(t[n],n,t))>0&&(d+=p);for(null!=u?v.sort(function(t,e){return u(g[t],g[e])}):null!=c&&v.sort(function(e,n){return c(t[e],t[n])}),n=0,i=d?(y-h*b)/d:0;n<h;++n,m=o)r=v[n],p=g[r],o=m+(p>0?p*i:0)+b,g[r]={data:t[r],index:n,value:p,startAngle:m,endAngle:o,padAngle:_};return g}var e=o.a,u=i.a,c=null,s=n.i(r.a)(0),l=n.i(r.a)(a.c),f=n.i(r.a)(0);return t.value=function(i){return arguments.length?(e=\"function\"==typeof i?i:n.i(r.a)(+i),t):e},t.sortValues=function(e){return arguments.length?(u=e,c=null,t):u},t.sort=function(e){return arguments.length?(c=e,u=null,t):c},t.startAngle=function(e){return arguments.length?(s=\"function\"==typeof e?e:n.i(r.a)(+e),t):s},t.endAngle=function(e){return arguments.length?(l=\"function\"==typeof e?e:n.i(r.a)(+e),t):l},t.padAngle=function(e){return arguments.length?(f=\"function\"==typeof e?e:n.i(r.a)(+e),t):f},t}},function(t,e,n){\"use strict\";var r=n(138),i=n(135),o=n(140);e.a=function(){var t=n.i(i.a)().curve(r.b),e=t.curve,a=t.lineX0,u=t.lineX1,c=t.lineY0,s=t.lineY1;return t.angle=t.x,delete t.x,t.startAngle=t.x0,delete t.x0,t.endAngle=t.x1,delete t.x1,t.radius=t.y,delete t.y,t.innerRadius=t.y0,delete t.y0,t.outerRadius=t.y1,delete t.y1,t.lineStartAngle=function(){return n.i(o.b)(a())},delete t.lineX0,t.lineEndAngle=function(){return n.i(o.b)(u())},delete t.lineX1,t.lineInnerRadius=function(){return n.i(o.b)(c())},delete t.lineY0,t.lineOuterRadius=function(){return n.i(o.b)(s())},delete t.lineY1,t.curve=function(t){return arguments.length?e(n.i(r.a)(t)):e()._curve},t}},function(t,e,n){\"use strict\";function r(t,e){return t[e]}var i=n(281),o=n(19),a=n(36),u=n(37);e.a=function(){function t(t){var n,r,i=e.apply(this,arguments),o=t.length,a=i.length,u=new Array(a);for(n=0;n<a;++n){for(var f,p=i[n],h=u[n]=new Array(o),d=0;d<o;++d)h[d]=f=[0,+l(t[d],p,d,t)],f.data=t[d];h.key=p}for(n=0,r=c(u);n<a;++n)u[r[n]].index=n;return s(u,r),u}var e=n.i(o.a)([]),c=u.a,s=a.a,l=r;return t.keys=function(r){return arguments.length?(e=\"function\"==typeof r?r:n.i(o.a)(i.a.call(r)),t):e},t.value=function(e){return arguments.length?(l=\"function\"==typeof e?e:n.i(o.a)(+e),t):l},t.order=function(e){return arguments.length?(c=null==e?u.a:\"function\"==typeof e?e:n.i(o.a)(i.a.call(e)),t):c},t.offset=function(e){return arguments.length?(s=null==e?a.a:e,t):s},t}},function(t,e,n){\"use strict\";var r=n(44),i=n(141),o=n(142),a=n(143),u=n(145),c=n(144),s=n(146),l=n(147),f=n(19);n.d(e,\"b\",function(){return p});var p=[i.a,o.a,a.a,c.a,u.a,s.a,l.a];e.a=function(){function t(){var t;if(a||(a=t=n.i(r.a)()),e.apply(this,arguments).draw(a,+o.apply(this,arguments)),t)return a=null,t+\"\"||null}var e=n.i(f.a)(i.a),o=n.i(f.a)(64),a=null;return t.type=function(r){return arguments.length?(e=\"function\"==typeof r?r:n.i(f.a)(r),t):e},t.size=function(e){return arguments.length?(o=\"function\"==typeof e?e:n.i(f.a)(+e),t):o},t.context=function(e){return arguments.length?(a=null==e?null:e,t):a},t}},function(t,e,n){\"use strict\";function r(t){var e=new Date(t);return isNaN(e)?null:e}var i=n(148),o=n(78),a=+new Date(\"2000-01-01T00:00:00.000Z\")?r:n.i(o.e)(i.b);e.a=a},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setHours(0,0,0,0)},function(t,e){t.setDate(t.getDate()+e)},function(t,e){return(e-t-(e.getTimezoneOffset()-t.getTimezoneOffset())*i.d)/i.b},function(t){return t.getDate()-1});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){var e=t.getTimezoneOffset()*i.d%i.c;e<0&&(e+=i.c),t.setTime(Math.floor((+t-e)/i.c)*i.c+e)},function(t,e){t.setTime(+t+e*i.c)},function(t,e){return(e-t)/i.c},function(t){return t.getHours()});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n.i(r.a)(function(){},function(t,e){t.setTime(+t+e)},function(t,e){return e-t});i.every=function(t){return t=Math.floor(t),isFinite(t)&&t>0?t>1?n.i(r.a)(function(e){e.setTime(Math.floor(e/t)*t)},function(e,n){e.setTime(+e+n*t)},function(e,n){return(n-e)/t}):i:null},e.a=i;i.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setTime(Math.floor(t/i.d)*i.d)},function(t,e){t.setTime(+t+e*i.d)},function(t,e){return(e-t)/i.d},function(t){return t.getMinutes()});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n.i(r.a)(function(t){t.setDate(1),t.setHours(0,0,0,0)},function(t,e){t.setMonth(t.getMonth()+e)},function(t,e){return e.getMonth()-t.getMonth()+12*(e.getFullYear()-t.getFullYear())},function(t){return t.getMonth()});e.a=i;i.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setTime(Math.floor(t/i.e)*i.e)},function(t,e){t.setTime(+t+e*i.e)},function(t,e){return(e-t)/i.e},function(t){return t.getUTCSeconds()});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setUTCHours(0,0,0,0)},function(t,e){t.setUTCDate(t.getUTCDate()+e)},function(t,e){return(e-t)/i.b},function(t){return t.getUTCDate()-1});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setUTCMinutes(0,0,0)},function(t,e){t.setTime(+t+e*i.c)},function(t,e){return(e-t)/i.c},function(t){return t.getUTCHours()});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n(13),o=n.i(r.a)(function(t){t.setUTCSeconds(0,0)},function(t,e){t.setTime(+t+e*i.d)},function(t,e){return(e-t)/i.d},function(t){return t.getUTCMinutes()});e.a=o;o.range},function(t,e,n){\"use strict\";var r=n(5),i=n.i(r.a)(function(t){t.setUTCDate(1),t.setUTCHours(0,0,0,0)},function(t,e){t.setUTCMonth(t.getUTCMonth()+e)},function(t,e){return e.getUTCMonth()-t.getUTCMonth()+12*(e.getUTCFullYear()-t.getUTCFullYear())},function(t){return t.getUTCMonth()});e.a=i;i.range},function(t,e,n){\"use strict\";function r(t){return n.i(i.a)(function(e){e.setUTCDate(e.getUTCDate()-(e.getUTCDay()+7-t)%7),e.setUTCHours(0,0,0,0)},function(t,e){t.setUTCDate(t.getUTCDate()+7*e)},function(t,e){return(e-t)/o.a})}var i=n(5),o=n(13);n.d(e,\"a\",function(){return a}),n.d(e,\"b\",function(){return u});var a=r(0),u=r(1),c=r(2),s=r(3),l=r(4),f=r(5),p=r(6);a.range,u.range,c.range,s.range,l.range,f.range,p.range},function(t,e,n){\"use strict\";var r=n(5),i=n.i(r.a)(function(t){t.setUTCMonth(0,1),t.setUTCHours(0,0,0,0)},function(t,e){t.setUTCFullYear(t.getUTCFullYear()+e)},function(t,e){return e.getUTCFullYear()-t.getUTCFullYear()},function(t){return t.getUTCFullYear()});i.every=function(t){return isFinite(t=Math.floor(t))&&t>0?n.i(r.a)(function(e){e.setUTCFullYear(Math.floor(e.getUTCFullYear()/t)*t),e.setUTCMonth(0,1),e.setUTCHours(0,0,0,0)},function(e,n){e.setUTCFullYear(e.getUTCFullYear()+n*t)}):null},e.a=i;i.range},function(t,e,n){\"use strict\";function r(t){return n.i(i.a)(function(e){e.setDate(e.getDate()-(e.getDay()+7-t)%7),e.setHours(0,0,0,0)},function(t,e){t.setDate(t.getDate()+7*e)},function(t,e){return(e-t-(e.getTimezoneOffset()-t.getTimezoneOffset())*o.d)/o.a})}var i=n(5),o=n(13);n.d(e,\"a\",function(){return a}),n.d(e,\"b\",function(){return u});var a=r(0),u=r(1),c=r(2),s=r(3),l=r(4),f=r(5),p=r(6);a.range,u.range,c.range,s.range,l.range,f.range,p.range},function(t,e,n){\"use strict\";var r=n(5),i=n.i(r.a)(function(t){t.setMonth(0,1),t.setHours(0,0,0,0)},function(t,e){t.setFullYear(t.getFullYear()+e)},function(t,e){return e.getFullYear()-t.getFullYear()},function(t){return t.getFullYear()});i.every=function(t){return isFinite(t=Math.floor(t))&&t>0?n.i(r.a)(function(e){e.setFullYear(Math.floor(e.getFullYear()/t)*t),e.setMonth(0,1),e.setHours(0,0,0,0)},function(e,n){e.setFullYear(e.getFullYear()+n*t)}):null},e.a=i;i.range},function(t,e,n){\"use strict\";function r(t){return t.replace(i,function(t,e){return e.toUpperCase()})}var i=/-(.)/g;t.exports=r},function(t,e,n){\"use strict\";function r(t){return i(t.replace(o,\"ms-\"))}var i=n(318),o=/^-ms-/;t.exports=r},function(t,e,n){\"use strict\";function r(t,e){return!(!t||!e)&&(t===e||!i(t)&&(i(e)?r(t,e.parentNode):\"contains\"in t?t.contains(e):!!t.compareDocumentPosition&&!!(16&t.compareDocumentPosition(e))))}var i=n(328);t.exports=r},function(t,e,n){\"use strict\";function r(t){var e=t.length;if(Array.isArray(t)||\"object\"!=typeof t&&\"function\"!=typeof t?a(!1):void 0,\"number\"!=typeof e?a(!1):void 0,0===e||e-1 in t?void 0:a(!1),\"function\"==typeof t.callee?a(!1):void 0,t.hasOwnProperty)try{return Array.prototype.slice.call(t)}catch(t){}for(var n=Array(e),r=0;r<e;r++)n[r]=t[r];return n}function i(t){return!!t&&(\"object\"==typeof t||\"function\"==typeof t)&&\"length\"in t&&!(\"setInterval\"in t)&&\"number\"!=typeof t.nodeType&&(Array.isArray(t)||\"callee\"in t||\"item\"in t)}function o(t){return i(t)?Array.isArray(t)?t.slice():r(t):[t]}var a=n(0);t.exports=o},function(t,e,n){\"use strict\";function r(t){var e=t.match(l);return e&&e[1].toLowerCase()}function i(t,e){var n=s;s?void 0:c(!1);var i=r(t),o=i&&u(i);if(o){n.innerHTML=o[1]+t+o[2];for(var l=o[0];l--;)n=n.lastChild}else n.innerHTML=t;var f=n.getElementsByTagName(\"script\");f.length&&(e?void 0:c(!1),a(f).forEach(e));for(var p=Array.from(n.childNodes);n.lastChild;)n.removeChild(n.lastChild);return p}var o=n(6),a=n(321),u=n(323),c=n(0),s=o.canUseDOM?document.createElement(\"div\"):null,l=/^\\s*<(\\w+)/;t.exports=i},function(t,e,n){\"use strict\";function r(t){return a?void 0:o(!1),p.hasOwnProperty(t)||(t=\"*\"),u.hasOwnProperty(t)||(\"*\"===t?a.innerHTML=\"<link />\":a.innerHTML=\"<\"+t+\"></\"+t+\">\",u[t]=!a.firstChild),u[t]?p[t]:null}var i=n(6),o=n(0),a=i.canUseDOM?document.createElement(\"div\"):null,u={},c=[1,'<select multiple=\"true\">',\"</select>\"],s=[1,\"<table>\",\"</table>\"],l=[3,\"<table><tbody><tr>\",\"</tr></tbody></table>\"],f=[1,'<svg xmlns=\"http://www.w3.org/2000/svg\">',\"</svg>\"],p={\"*\":[1,\"?<div>\",\"</div>\"],area:[1,\"<map>\",\"</map>\"],col:[2,\"<table><tbody></tbody><colgroup>\",\"</colgroup></table>\"],legend:[1,\"<fieldset>\",\"</fieldset>\"],param:[1,\"<object>\",\"</object>\"],tr:[2,\"<table><tbody>\",\"</tbody></table>\"],optgroup:c,option:c,caption:s,colgroup:s,tbody:s,tfoot:s,thead:s,td:l,th:l},h=[\"circle\",\"clipPath\",\"defs\",\"ellipse\",\"g\",\"image\",\"line\",\"linearGradient\",\"mask\",\"path\",\"pattern\",\"polygon\",\"polyline\",\"radialGradient\",\"rect\",\"stop\",\"text\",\"tspan\"];h.forEach(function(t){p[t]=f,u[t]=!0}),t.exports=r},function(t,e,n){\"use strict\";function r(t){return t===window?{x:window.pageXOffset||document.documentElement.scrollLeft,y:window.pageYOffset||document.documentElement.scrollTop}:{x:t.scrollLeft,y:t.scrollTop}}t.exports=r},function(t,e,n){\"use strict\";function r(t){return t.replace(i,\"-$1\").toLowerCase()}var i=/([A-Z])/g;t.exports=r},function(t,e,n){\"use strict\";function r(t){return i(t).replace(o,\"-ms-\")}var i=n(325),o=/^ms-/;t.exports=r},function(t,e,n){\"use strict\";function r(t){return!(!t||!(\"function\"==typeof Node?t instanceof Node:\"object\"==typeof t&&\"number\"==typeof t.nodeType&&\"string\"==typeof t.nodeName))}t.exports=r},function(t,e,n){\"use strict\";function r(t){return i(t)&&3==t.nodeType}var i=n(327);t.exports=r},function(t,e,n){\"use strict\";var r=function(t){var e;for(e in t)if(t.hasOwnProperty(e))return e;return null};t.exports=r},function(t,e,n){\"use strict\";function r(t){var e={};return function(n){return e.hasOwnProperty(n)||(e[n]=t.call(this,n)),e[n]}}t.exports=r},function(t,e,n){\"use strict\";var r={Properties:{\"aria-current\":0,\"aria-details\":0,\"aria-disabled\":0,\"aria-hidden\":0,\"aria-invalid\":0,\"aria-keyshortcuts\":0,\"aria-label\":0,\"aria-roledescription\":0,\"aria-autocomplete\":0,\"aria-checked\":0,\"aria-expanded\":0,\"aria-haspopup\":0,\"aria-level\":0,\"aria-modal\":0,\"aria-multiline\":0,\"aria-multiselectable\":0,\"aria-orientation\":0,\"aria-placeholder\":0,\"aria-pressed\":0,\"aria-readonly\":0,\"aria-required\":0,\"aria-selected\":0,\"aria-sort\":0,\"aria-valuemax\":0,\"aria-valuemin\":0,\"aria-valuenow\":0,\"aria-valuetext\":0,\"aria-atomic\":0,\"aria-busy\":0,\"aria-live\":0,\"aria-relevant\":0,\"aria-dropeffect\":0,\"aria-grabbed\":0,\"aria-activedescendant\":0,\"aria-colcount\":0,\"aria-colindex\":0,\"aria-colspan\":0,\"aria-controls\":0,\"aria-describedby\":0,\"aria-errormessage\":0,\"aria-flowto\":0,\"aria-labelledby\":0,\"aria-owns\":0,\"aria-posinset\":0,\"aria-rowcount\":0,\"aria-rowindex\":0,\"aria-rowspan\":0,\"aria-setsize\":0},DOMAttributeNames:{},DOMPropertyNames:{}};t.exports=r},function(t,e,n){\"use strict\";var r=n(4),i=n(151),o={focusDOMComponent:function(){i(r.getNodeFromInstance(this))}};t.exports=o},function(t,e,n){\"use strict\";function r(){var t=window.opera;return\"object\"==typeof t&&\"function\"==typeof t.version&&parseInt(t.version(),10)<=12}function i(t){return(t.ctrlKey||t.altKey||t.metaKey)&&!(t.ctrlKey&&t.altKey)}function o(t){switch(t){case\"topCompositionStart\":return E.compositionStart;case\"topCompositionEnd\":return E.compositionEnd;case\"topCompositionUpdate\":return E.compositionUpdate}}function a(t,e){return\"topKeyDown\"===t&&e.keyCode===_}function u(t,e){switch(t){case\"topKeyUp\":return y.indexOf(e.keyCode)!==-1;case\"topKeyDown\":return e.keyCode!==_;case\"topKeyPress\":case\"topMouseDown\":case\"topBlur\":return!0;default:return!1}}function c(t){var e=t.detail;return\"object\"==typeof e&&\"data\"in e?e.data:null}function s(t,e,n,r){var i,s;if(b?i=o(t):S?u(t,n)&&(i=E.compositionEnd):a(t,n)&&(i=E.compositionStart),!i)return null;C&&(S||i!==E.compositionStart?i===E.compositionEnd&&S&&(s=S.getData()):S=v.getPooled(r));var l=g.getPooled(i,e,n,r);if(s)l.data=s;else{var f=c(n);null!==f&&(l.data=f)}return h.accumulateTwoPhaseDispatches(l),l}function l(t,e){switch(t){case\"topCompositionEnd\":return c(e);case\"topKeyPress\":var n=e.which;return n!==M?null:(T=!0,k);case\"topTextInput\":var r=e.data;return r===k&&T?null:r;default:return null}}function f(t,e){if(S){if(\"topCompositionEnd\"===t||!b&&u(t,e)){var n=S.getData();return v.release(S),S=null,n}return null}switch(t){case\"topPaste\":return null;case\"topKeyPress\":return e.which&&!i(e)?String.fromCharCode(e.which):null;case\"topCompositionEnd\":return C?null:e.data;default:return null}}function p(t,e,n,r){var i;if(i=w?l(t,n):f(t,n),!i)return null;var o=m.getPooled(E.beforeInput,e,n,r);return o.data=i,h.accumulateTwoPhaseDispatches(o),o}var h=n(23),d=n(6),v=n(340),g=n(377),m=n(380),y=[9,13,27,32],_=229,b=d.canUseDOM&&\"CompositionEvent\"in window,x=null;d.canUseDOM&&\"documentMode\"in document&&(x=document.documentMode);var w=d.canUseDOM&&\"TextEvent\"in window&&!x&&!r(),C=d.canUseDOM&&(!b||x&&x>8&&x<=11),M=32,k=String.fromCharCode(M),E={beforeInput:{phasedRegistrationNames:{bubbled:\"onBeforeInput\",captured:\"onBeforeInputCapture\"},dependencies:[\"topCompositionEnd\",\"topKeyPress\",\"topTextInput\",\"topPaste\"]},compositionEnd:{phasedRegistrationNames:{bubbled:\"onCompositionEnd\",captured:\"onCompositionEndCapture\"},dependencies:[\"topBlur\",\"topCompositionEnd\",\"topKeyDown\",\"topKeyPress\",\"topKeyUp\",\"topMouseDown\"]},compositionStart:{phasedRegistrationNames:{bubbled:\"onCompositionStart\",captured:\"onCompositionStartCapture\"},dependencies:[\"topBlur\",\"topCompositionStart\",\"topKeyDown\",\"topKeyPress\",\"topKeyUp\",\"topMouseDown\"]},compositionUpdate:{phasedRegistrationNames:{bubbled:\"onCompositionUpdate\",captured:\"onCompositionUpdateCapture\"},dependencies:[\"topBlur\",\"topCompositionUpdate\",\"topKeyDown\",\"topKeyPress\",\"topKeyUp\",\"topMouseDown\"]}},T=!1,S=null,P={eventTypes:E,extractEvents:function(t,e,n,r){return[s(t,e,n,r),p(t,e,n,r)]}};t.exports=P},function(t,e,n){\"use strict\";var r=n(154),i=n(6),o=(n(9),n(319),n(386)),a=n(326),u=n(330),c=(n(1),u(function(t){return a(t)})),s=!1,l=\"cssFloat\";if(i.canUseDOM){var f=document.createElement(\"div\").style;try{f.font=\"\"}catch(t){s=!0}void 0===document.documentElement.style.cssFloat&&(l=\"styleFloat\")}var p={createMarkupForStyles:function(t,e){var n=\"\";for(var r in t)if(t.hasOwnProperty(r)){var i=t[r];null!=i&&(n+=c(r)+\":\",n+=o(r,i,e)+\";\")}return n||null},setValueForStyles:function(t,e,n){var i=t.style;for(var a in e)if(e.hasOwnProperty(a)){var u=o(a,e[a],n);if(\"float\"!==a&&\"cssFloat\"!==a||(a=l),u)i[a]=u;else{var c=s&&r.shorthandPropertyExpansions[a];if(c)for(var f in c)i[f]=\"\";else i[a]=\"\"}}}};t.exports=p},function(t,e,n){\"use strict\";function r(t){var e=t.nodeName&&t.nodeName.toLowerCase();return\"select\"===e||\"input\"===e&&\"file\"===t.type}function i(t){var e=C.getPooled(T.change,P,t,M(t));_.accumulateTwoPhaseDispatches(e),w.batchedUpdates(o,e)}function o(t){y.enqueueEvents(t),y.processEventQueue(!1)}function a(t,e){S=t,P=e,S.attachEvent(\"onchange\",i)}function u(){S&&(S.detachEvent(\"onchange\",i),S=null,P=null)}function c(t,e){if(\"topChange\"===t)return e}function s(t,e,n){\"topFocus\"===t?(u(),a(e,n)):\"topBlur\"===t&&u()}function l(t,e){S=t,P=e,N=t.value,A=Object.getOwnPropertyDescriptor(t.constructor.prototype,\"value\"),Object.defineProperty(S,\"value\",D),S.attachEvent?S.attachEvent(\"onpropertychange\",p):S.addEventListener(\"propertychange\",p,!1)}function f(){S&&(delete S.value,S.detachEvent?S.detachEvent(\"onpropertychange\",p):S.removeEventListener(\"propertychange\",p,!1),S=null,P=null,N=null,A=null)}function p(t){if(\"value\"===t.propertyName){var e=t.srcElement.value;e!==N&&(N=e,i(t))}}function h(t,e){if(\"topInput\"===t)return e}function d(t,e,n){\"topFocus\"===t?(f(),l(e,n)):\"topBlur\"===t&&f()}function v(t,e){if((\"topSelectionChange\"===t||\"topKeyUp\"===t||\"topKeyDown\"===t)&&S&&S.value!==N)return N=S.value,P}function g(t){return t.nodeName&&\"input\"===t.nodeName.toLowerCase()&&(\"checkbox\"===t.type||\"radio\"===t.type)}function m(t,e){if(\"topClick\"===t)return e}var y=n(22),_=n(23),b=n(6),x=n(4),w=n(11),C=n(14),M=n(93),k=n(94),E=n(170),T={change:{phasedRegistrationNames:{bubbled:\"onChange\",captured:\"onChangeCapture\"},dependencies:[\"topBlur\",\"topChange\",\"topClick\",\"topFocus\",\"topInput\",\"topKeyDown\",\"topKeyUp\",\"topSelectionChange\"]}},S=null,P=null,N=null,A=null,O=!1;b.canUseDOM&&(O=k(\"change\")&&(!document.documentMode||document.documentMode>8));var I=!1;b.canUseDOM&&(I=k(\"input\")&&(!document.documentMode||document.documentMode>11));var D={get:function(){return A.get.call(this)},set:function(t){N=\"\"+t,A.set.call(this,t)}},R={eventTypes:T,extractEvents:function(t,e,n,i){var o,a,u=e?x.getNodeFromInstance(e):window;if(r(u)?O?o=c:a=s:E(u)?I?o=h:(o=v,a=d):g(u)&&(o=m),o){var l=o(t,e);if(l){var f=C.getPooled(T.change,l,n,i);return f.type=\"change\",_.accumulateTwoPhaseDispatches(f),f}}a&&a(t,u,e)}};t.exports=R},function(t,e,n){\"use strict\";var r=n(2),i=n(20),o=n(6),a=n(322),u=n(8),c=(n(0),{dangerouslyReplaceNodeWithMarkup:function(t,e){if(o.canUseDOM?void 0:r(\"56\"),e?void 0:r(\"57\"),\"HTML\"===t.nodeName?r(\"58\"):void 0,\"string\"==typeof e){var n=a(e,u)[0];t.parentNode.replaceChild(n,t)}else i.replaceChildWithTree(t,e)}});t.exports=c},function(t,e,n){\"use strict\";var r=[\"ResponderEventPlugin\",\"SimpleEventPlugin\",\"TapEventPlugin\",\"EnterLeaveEventPlugin\",\"ChangeEventPlugin\",\"SelectEventPlugin\",\"BeforeInputEventPlugin\"];t.exports=r},function(t,e,n){\"use strict\";var r=n(23),i=n(4),o=n(52),a={mouseEnter:{registrationName:\"onMouseEnter\",dependencies:[\"topMouseOut\",\"topMouseOver\"]},mouseLeave:{registrationName:\"onMouseLeave\",dependencies:[\"topMouseOut\",\"topMouseOver\"]}},u={eventTypes:a,extractEvents:function(t,e,n,u){if(\"topMouseOver\"===t&&(n.relatedTarget||n.fromElement))return null;\n",
              "if(\"topMouseOut\"!==t&&\"topMouseOver\"!==t)return null;var c;if(u.window===u)c=u;else{var s=u.ownerDocument;c=s?s.defaultView||s.parentWindow:window}var l,f;if(\"topMouseOut\"===t){l=e;var p=n.relatedTarget||n.toElement;f=p?i.getClosestInstanceFromNode(p):null}else l=null,f=e;if(l===f)return null;var h=null==l?c:i.getNodeFromInstance(l),d=null==f?c:i.getNodeFromInstance(f),v=o.getPooled(a.mouseLeave,l,n,u);v.type=\"mouseleave\",v.target=h,v.relatedTarget=d;var g=o.getPooled(a.mouseEnter,f,n,u);return g.type=\"mouseenter\",g.target=d,g.relatedTarget=h,r.accumulateEnterLeaveDispatches(v,g,l,f),[v,g]}};t.exports=u},function(t,e,n){\"use strict\";var r={topAbort:null,topAnimationEnd:null,topAnimationIteration:null,topAnimationStart:null,topBlur:null,topCanPlay:null,topCanPlayThrough:null,topChange:null,topClick:null,topCompositionEnd:null,topCompositionStart:null,topCompositionUpdate:null,topContextMenu:null,topCopy:null,topCut:null,topDoubleClick:null,topDrag:null,topDragEnd:null,topDragEnter:null,topDragExit:null,topDragLeave:null,topDragOver:null,topDragStart:null,topDrop:null,topDurationChange:null,topEmptied:null,topEncrypted:null,topEnded:null,topError:null,topFocus:null,topInput:null,topInvalid:null,topKeyDown:null,topKeyPress:null,topKeyUp:null,topLoad:null,topLoadedData:null,topLoadedMetadata:null,topLoadStart:null,topMouseDown:null,topMouseMove:null,topMouseOut:null,topMouseOver:null,topMouseUp:null,topPaste:null,topPause:null,topPlay:null,topPlaying:null,topProgress:null,topRateChange:null,topReset:null,topScroll:null,topSeeked:null,topSeeking:null,topSelectionChange:null,topStalled:null,topSubmit:null,topSuspend:null,topTextInput:null,topTimeUpdate:null,topTouchCancel:null,topTouchEnd:null,topTouchMove:null,topTouchStart:null,topTransitionEnd:null,topVolumeChange:null,topWaiting:null,topWheel:null},i={topLevelTypes:r};t.exports=i},function(t,e,n){\"use strict\";function r(t){this._root=t,this._startText=this.getText(),this._fallbackText=null}var i=n(3),o=n(17),a=n(168);i(r.prototype,{destructor:function(){this._root=null,this._startText=null,this._fallbackText=null},getText:function(){return\"value\"in this._root?this._root.value:this._root[a()]},getData:function(){if(this._fallbackText)return this._fallbackText;var t,e,n=this._startText,r=n.length,i=this.getText(),o=i.length;for(t=0;t<r&&n[t]===i[t];t++);var a=r-t;for(e=1;e<=a&&n[r-e]===i[o-e];e++);var u=e>1?1-e:void 0;return this._fallbackText=i.slice(t,u),this._fallbackText}}),o.addPoolingTo(r),t.exports=r},function(t,e,n){\"use strict\";var r=n(21),i=r.injection.MUST_USE_PROPERTY,o=r.injection.HAS_BOOLEAN_VALUE,a=r.injection.HAS_NUMERIC_VALUE,u=r.injection.HAS_POSITIVE_NUMERIC_VALUE,c=r.injection.HAS_OVERLOADED_BOOLEAN_VALUE,s={isCustomAttribute:RegExp.prototype.test.bind(new RegExp(\"^(data|aria)-[\"+r.ATTRIBUTE_NAME_CHAR+\"]*$\")),Properties:{accept:0,acceptCharset:0,accessKey:0,action:0,allowFullScreen:o,allowTransparency:0,alt:0,as:0,async:o,autoComplete:0,autoPlay:o,capture:o,cellPadding:0,cellSpacing:0,charSet:0,challenge:0,checked:i|o,cite:0,classID:0,className:0,cols:u,colSpan:0,content:0,contentEditable:0,contextMenu:0,controls:o,coords:0,crossOrigin:0,data:0,dateTime:0,default:o,defer:o,dir:0,disabled:o,download:c,draggable:0,encType:0,form:0,formAction:0,formEncType:0,formMethod:0,formNoValidate:o,formTarget:0,frameBorder:0,headers:0,height:0,hidden:o,high:0,href:0,hrefLang:0,htmlFor:0,httpEquiv:0,icon:0,id:0,inputMode:0,integrity:0,is:0,keyParams:0,keyType:0,kind:0,label:0,lang:0,list:0,loop:o,low:0,manifest:0,marginHeight:0,marginWidth:0,max:0,maxLength:0,media:0,mediaGroup:0,method:0,min:0,minLength:0,multiple:i|o,muted:i|o,name:0,nonce:0,noValidate:o,open:o,optimum:0,pattern:0,placeholder:0,playsInline:o,poster:0,preload:0,profile:0,radioGroup:0,readOnly:o,referrerPolicy:0,rel:0,required:o,reversed:o,role:0,rows:u,rowSpan:a,sandbox:0,scope:0,scoped:o,scrolling:0,seamless:o,selected:i|o,shape:0,size:u,sizes:0,span:u,spellCheck:0,src:0,srcDoc:0,srcLang:0,srcSet:0,start:a,step:0,style:0,summary:0,tabIndex:0,target:0,title:0,type:0,useMap:0,value:0,width:0,wmode:0,wrap:0,about:0,datatype:0,inlist:0,prefix:0,property:0,resource:0,typeof:0,vocab:0,autoCapitalize:0,autoCorrect:0,autoSave:0,color:0,itemProp:0,itemScope:o,itemType:0,itemID:0,itemRef:0,results:0,security:0,unselectable:0},DOMAttributeNames:{acceptCharset:\"accept-charset\",className:\"class\",htmlFor:\"for\",httpEquiv:\"http-equiv\"},DOMPropertyNames:{}};t.exports=s},function(t,e,n){\"use strict\";(function(e){function r(t,e,n,r){var i=void 0===t[n];null!=e&&i&&(t[n]=o(e,!0))}var i=n(24),o=n(169),a=(n(84),n(95)),u=n(172);n(1);\"undefined\"!=typeof e&&e.env,1;var c={instantiateChildren:function(t,e,n,i){if(null==t)return null;var o={};return u(t,r,o),o},updateChildren:function(t,e,n,r,u,c,s,l,f){if(e||t){var p,h;for(p in e)if(e.hasOwnProperty(p)){h=t&&t[p];var d=h&&h._currentElement,v=e[p];if(null!=h&&a(d,v))i.receiveComponent(h,v,u,l),e[p]=h;else{h&&(r[p]=i.getHostNode(h),i.unmountComponent(h,!1));var g=o(v,!0);e[p]=g;var m=i.mountComponent(g,u,c,s,l,f);n.push(m)}}for(p in t)!t.hasOwnProperty(p)||e&&e.hasOwnProperty(p)||(h=t[p],r[p]=i.getHostNode(h),i.unmountComponent(h,!1))}},unmountChildren:function(t,e){for(var n in t)if(t.hasOwnProperty(n)){var r=t[n];i.unmountComponent(r,e)}}};t.exports=c}).call(e,n(153))},function(t,e,n){\"use strict\";var r=n(81),i=n(350),o={processChildrenUpdates:i.dangerouslyProcessChildrenUpdates,replaceNodeWithMarkup:r.dangerouslyReplaceNodeWithMarkup};t.exports=o},function(t,e,n){\"use strict\";function r(t){}function i(t,e){}function o(t){return!(!t.prototype||!t.prototype.isReactComponent)}function a(t){return!(!t.prototype||!t.prototype.isPureReactComponent)}var u=n(2),c=n(3),s=n(26),l=n(86),f=n(15),p=n(87),h=n(40),d=(n(9),n(164)),v=n(24),g=n(38),m=(n(0),n(80)),y=n(95),_=(n(1),{ImpureClass:0,PureClass:1,StatelessFunctional:2});r.prototype.render=function(){var t=h.get(this)._currentElement.type,e=t(this.props,this.context,this.updater);return i(t,e),e};var b=1,x={construct:function(t){this._currentElement=t,this._rootNodeID=0,this._compositeType=null,this._instance=null,this._hostParent=null,this._hostContainerInfo=null,this._updateBatchNumber=null,this._pendingElement=null,this._pendingStateQueue=null,this._pendingReplaceState=!1,this._pendingForceUpdate=!1,this._renderedNodeType=null,this._renderedComponent=null,this._context=null,this._mountOrder=0,this._topLevelWrapper=null,this._pendingCallbacks=null,this._calledComponentWillUnmount=!1},mountComponent:function(t,e,n,c){this._context=c,this._mountOrder=b++,this._hostParent=e,this._hostContainerInfo=n;var l,f=this._currentElement.props,p=this._processContext(c),d=this._currentElement.type,v=t.getUpdateQueue(),m=o(d),y=this._constructComponent(m,f,p,v);m||null!=y&&null!=y.render?a(d)?this._compositeType=_.PureClass:this._compositeType=_.ImpureClass:(l=y,i(d,l),null===y||y===!1||s.isValidElement(y)?void 0:u(\"105\",d.displayName||d.name||\"Component\"),y=new r(d),this._compositeType=_.StatelessFunctional);y.props=f,y.context=p,y.refs=g,y.updater=v,this._instance=y,h.set(y,this);var x=y.state;void 0===x&&(y.state=x=null),\"object\"!=typeof x||Array.isArray(x)?u(\"106\",this.getName()||\"ReactCompositeComponent\"):void 0,this._pendingStateQueue=null,this._pendingReplaceState=!1,this._pendingForceUpdate=!1;var w;return w=y.unstable_handleError?this.performInitialMountWithErrorHandling(l,e,n,t,c):this.performInitialMount(l,e,n,t,c),y.componentDidMount&&t.getReactMountReady().enqueue(y.componentDidMount,y),w},_constructComponent:function(t,e,n,r){return this._constructComponentWithoutOwner(t,e,n,r)},_constructComponentWithoutOwner:function(t,e,n,r){var i=this._currentElement.type;return t?new i(e,n,r):i(e,n,r)},performInitialMountWithErrorHandling:function(t,e,n,r,i){var o,a=r.checkpoint();try{o=this.performInitialMount(t,e,n,r,i)}catch(u){r.rollback(a),this._instance.unstable_handleError(u),this._pendingStateQueue&&(this._instance.state=this._processPendingState(this._instance.props,this._instance.context)),a=r.checkpoint(),this._renderedComponent.unmountComponent(!0),r.rollback(a),o=this.performInitialMount(t,e,n,r,i)}return o},performInitialMount:function(t,e,n,r,i){var o=this._instance,a=0;o.componentWillMount&&(o.componentWillMount(),this._pendingStateQueue&&(o.state=this._processPendingState(o.props,o.context))),void 0===t&&(t=this._renderValidatedComponent());var u=d.getType(t);this._renderedNodeType=u;var c=this._instantiateReactComponent(t,u!==d.EMPTY);this._renderedComponent=c;var s=v.mountComponent(c,r,e,n,this._processChildContext(i),a);return s},getHostNode:function(){return v.getHostNode(this._renderedComponent)},unmountComponent:function(t){if(this._renderedComponent){var e=this._instance;if(e.componentWillUnmount&&!e._calledComponentWillUnmount)if(e._calledComponentWillUnmount=!0,t){var n=this.getName()+\".componentWillUnmount()\";p.invokeGuardedCallback(n,e.componentWillUnmount.bind(e))}else e.componentWillUnmount();this._renderedComponent&&(v.unmountComponent(this._renderedComponent,t),this._renderedNodeType=null,this._renderedComponent=null,this._instance=null),this._pendingStateQueue=null,this._pendingReplaceState=!1,this._pendingForceUpdate=!1,this._pendingCallbacks=null,this._pendingElement=null,this._context=null,this._rootNodeID=0,this._topLevelWrapper=null,h.remove(e)}},_maskContext:function(t){var e=this._currentElement.type,n=e.contextTypes;if(!n)return g;var r={};for(var i in n)r[i]=t[i];return r},_processContext:function(t){var e=this._maskContext(t);return e},_processChildContext:function(t){var e,n=this._currentElement.type,r=this._instance;if(r.getChildContext&&(e=r.getChildContext()),e){\"object\"!=typeof n.childContextTypes?u(\"107\",this.getName()||\"ReactCompositeComponent\"):void 0;for(var i in e)i in n.childContextTypes?void 0:u(\"108\",this.getName()||\"ReactCompositeComponent\",i);return c({},t,e)}return t},_checkContextTypes:function(t,e,n){},receiveComponent:function(t,e,n){var r=this._currentElement,i=this._context;this._pendingElement=null,this.updateComponent(e,r,t,i,n)},performUpdateIfNecessary:function(t){null!=this._pendingElement?v.receiveComponent(this,this._pendingElement,t,this._context):null!==this._pendingStateQueue||this._pendingForceUpdate?this.updateComponent(t,this._currentElement,this._currentElement,this._context,this._context):this._updateBatchNumber=null},updateComponent:function(t,e,n,r,i){var o=this._instance;null==o?u(\"136\",this.getName()||\"ReactCompositeComponent\"):void 0;var a,c=!1;this._context===i?a=o.context:(a=this._processContext(i),c=!0);var s=e.props,l=n.props;e!==n&&(c=!0),c&&o.componentWillReceiveProps&&o.componentWillReceiveProps(l,a);var f=this._processPendingState(l,a),p=!0;this._pendingForceUpdate||(o.shouldComponentUpdate?p=o.shouldComponentUpdate(l,f,a):this._compositeType===_.PureClass&&(p=!m(s,l)||!m(o.state,f))),this._updateBatchNumber=null,p?(this._pendingForceUpdate=!1,this._performComponentUpdate(n,l,f,a,t,i)):(this._currentElement=n,this._context=i,o.props=l,o.state=f,o.context=a)},_processPendingState:function(t,e){var n=this._instance,r=this._pendingStateQueue,i=this._pendingReplaceState;if(this._pendingReplaceState=!1,this._pendingStateQueue=null,!r)return n.state;if(i&&1===r.length)return r[0];for(var o=c({},i?r[0]:n.state),a=i?1:0;a<r.length;a++){var u=r[a];c(o,\"function\"==typeof u?u.call(n,o,t,e):u)}return o},_performComponentUpdate:function(t,e,n,r,i,o){var a,u,c,s=this._instance,l=Boolean(s.componentDidUpdate);l&&(a=s.props,u=s.state,c=s.context),s.componentWillUpdate&&s.componentWillUpdate(e,n,r),this._currentElement=t,this._context=o,s.props=e,s.state=n,s.context=r,this._updateRenderedComponent(i,o),l&&i.getReactMountReady().enqueue(s.componentDidUpdate.bind(s,a,u,c),s)},_updateRenderedComponent:function(t,e){var n=this._renderedComponent,r=n._currentElement,i=this._renderValidatedComponent(),o=0;if(y(r,i))v.receiveComponent(n,i,t,this._processChildContext(e));else{var a=v.getHostNode(n);v.unmountComponent(n,!1);var u=d.getType(i);this._renderedNodeType=u;var c=this._instantiateReactComponent(i,u!==d.EMPTY);this._renderedComponent=c;var s=v.mountComponent(c,t,this._hostParent,this._hostContainerInfo,this._processChildContext(e),o);this._replaceNodeWithMarkup(a,s,n)}},_replaceNodeWithMarkup:function(t,e,n){l.replaceNodeWithMarkup(t,e,n)},_renderValidatedComponentWithoutOwnerOrContext:function(){var t,e=this._instance;return t=e.render()},_renderValidatedComponent:function(){var t;if(this._compositeType!==_.StatelessFunctional){f.current=this;try{t=this._renderValidatedComponentWithoutOwnerOrContext()}finally{f.current=null}}else t=this._renderValidatedComponentWithoutOwnerOrContext();return null===t||t===!1||s.isValidElement(t)?void 0:u(\"109\",this.getName()||\"ReactCompositeComponent\"),t},attachRef:function(t,e){var n=this.getPublicInstance();null==n?u(\"110\"):void 0;var r=e.getPublicInstance(),i=n.refs===g?n.refs={}:n.refs;i[t]=r},detachRef:function(t){var e=this.getPublicInstance().refs;delete e[t]},getName:function(){var t=this._currentElement.type,e=this._instance&&this._instance.constructor;return t.displayName||e&&e.displayName||t.name||e&&e.name||null},getPublicInstance:function(){var t=this._instance;return this._compositeType===_.StatelessFunctional?null:t},_instantiateReactComponent:null};t.exports=x},function(t,e,n){\"use strict\";var r=n(4),i=n(358),o=n(163),a=n(24),u=n(11),c=n(371),s=n(387),l=n(167),f=n(395);n(1);i.inject();var p={findDOMNode:s,render:o.render,unmountComponentAtNode:o.unmountComponentAtNode,version:c,unstable_batchedUpdates:u.batchedUpdates,unstable_renderSubtreeIntoContainer:f};\"undefined\"!=typeof __REACT_DEVTOOLS_GLOBAL_HOOK__&&\"function\"==typeof __REACT_DEVTOOLS_GLOBAL_HOOK__.inject&&__REACT_DEVTOOLS_GLOBAL_HOOK__.inject({ComponentTree:{getClosestInstanceFromNode:r.getClosestInstanceFromNode,getNodeFromInstance:function(t){return t._renderedComponent&&(t=l(t)),t?r.getNodeFromInstance(t):null}},Mount:o,Reconciler:a});t.exports=p},function(t,e,n){\"use strict\";function r(t){if(t){var e=t._currentElement._owner||null;if(e){var n=e.getName();if(n)return\" This DOM node was rendered by `\"+n+\"`.\"}}return\"\"}function i(t,e){e&&(G[t._tag]&&(null!=e.children||null!=e.dangerouslySetInnerHTML?v(\"137\",t._tag,t._currentElement._owner?\" Check the render method of \"+t._currentElement._owner.getName()+\".\":\"\"):void 0),null!=e.dangerouslySetInnerHTML&&(null!=e.children?v(\"60\"):void 0,\"object\"==typeof e.dangerouslySetInnerHTML&&V in e.dangerouslySetInnerHTML?void 0:v(\"61\")),null!=e.style&&\"object\"!=typeof e.style?v(\"62\",r(t)):void 0)}function o(t,e,n,r){if(!(r instanceof I)){var i=t._hostContainerInfo,o=i._node&&i._node.nodeType===H,u=o?i._node:i._ownerDocument;F(e,u),r.getReactMountReady().enqueue(a,{inst:t,registrationName:e,listener:n})}}function a(){var t=this;C.putListener(t.inst,t.registrationName,t.listener)}function u(){var t=this;S.postMountWrapper(t)}function c(){var t=this;A.postMountWrapper(t)}function s(){var t=this;P.postMountWrapper(t)}function l(){var t=this;t._rootNodeID?void 0:v(\"63\");var e=U(t);switch(e?void 0:v(\"64\"),t._tag){case\"iframe\":case\"object\":t._wrapperState.listeners=[k.trapBubbledEvent(\"topLoad\",\"load\",e)];break;case\"video\":case\"audio\":t._wrapperState.listeners=[];for(var n in q)q.hasOwnProperty(n)&&t._wrapperState.listeners.push(k.trapBubbledEvent(n,q[n],e));break;case\"source\":t._wrapperState.listeners=[k.trapBubbledEvent(\"topError\",\"error\",e)];break;case\"img\":t._wrapperState.listeners=[k.trapBubbledEvent(\"topError\",\"error\",e),k.trapBubbledEvent(\"topLoad\",\"load\",e)];break;case\"form\":t._wrapperState.listeners=[k.trapBubbledEvent(\"topReset\",\"reset\",e),k.trapBubbledEvent(\"topSubmit\",\"submit\",e)];break;case\"input\":case\"select\":case\"textarea\":t._wrapperState.listeners=[k.trapBubbledEvent(\"topInvalid\",\"invalid\",e)]}}function f(){N.postUpdateWrapper(this)}function p(t){Z.call(X,t)||($.test(t)?void 0:v(\"65\",t),X[t]=!0)}function h(t,e){return t.indexOf(\"-\")>=0||null!=e.is}function d(t){var e=t.type;p(e),this._currentElement=t,this._tag=e.toLowerCase(),this._namespaceURI=null,this._renderedChildren=null,this._previousStyle=null,this._previousStyleCopy=null,this._hostNode=null,this._hostParent=null,this._rootNodeID=0,this._domID=0,this._hostContainerInfo=null,this._wrapperState=null,this._topLevelWrapper=null,this._flags=0}var v=n(2),g=n(3),m=n(332),y=n(334),_=n(20),b=n(82),x=n(21),w=n(156),C=n(22),M=n(83),k=n(51),E=n(157),T=n(4),S=n(351),P=n(352),N=n(158),A=n(355),O=(n(9),n(364)),I=n(369),D=(n(8),n(54)),R=(n(0),n(94),n(80),n(96),n(1),E),L=C.deleteListener,U=T.getNodeFromInstance,F=k.listenTo,j=M.registrationNameModules,B={string:!0,number:!0},W=\"style\",V=\"__html\",z={children:null,dangerouslySetInnerHTML:null,suppressContentEditableWarning:null},H=11,q={topAbort:\"abort\",topCanPlay:\"canplay\",topCanPlayThrough:\"canplaythrough\",topDurationChange:\"durationchange\",topEmptied:\"emptied\",topEncrypted:\"encrypted\",topEnded:\"ended\",topError:\"error\",topLoadedData:\"loadeddata\",topLoadedMetadata:\"loadedmetadata\",topLoadStart:\"loadstart\",topPause:\"pause\",topPlay:\"play\",topPlaying:\"playing\",topProgress:\"progress\",topRateChange:\"ratechange\",topSeeked:\"seeked\",topSeeking:\"seeking\",topStalled:\"stalled\",topSuspend:\"suspend\",topTimeUpdate:\"timeupdate\",topVolumeChange:\"volumechange\",topWaiting:\"waiting\"},Y={area:!0,base:!0,br:!0,col:!0,embed:!0,hr:!0,img:!0,input:!0,keygen:!0,link:!0,meta:!0,param:!0,source:!0,track:!0,wbr:!0},K={listing:!0,pre:!0,textarea:!0},G=g({menuitem:!0},Y),$=/^[a-zA-Z][a-zA-Z:_\\.\\-\\d]*$/,X={},Z={}.hasOwnProperty,Q=1;d.displayName=\"ReactDOMComponent\",d.Mixin={mountComponent:function(t,e,n,r){this._rootNodeID=Q++,this._domID=n._idCounter++,this._hostParent=e,this._hostContainerInfo=n;var o=this._currentElement.props;switch(this._tag){case\"audio\":case\"form\":case\"iframe\":case\"img\":case\"link\":case\"object\":case\"source\":case\"video\":this._wrapperState={listeners:null},t.getReactMountReady().enqueue(l,this);break;case\"input\":S.mountWrapper(this,o,e),o=S.getHostProps(this,o),t.getReactMountReady().enqueue(l,this);break;case\"option\":P.mountWrapper(this,o,e),o=P.getHostProps(this,o);break;case\"select\":N.mountWrapper(this,o,e),o=N.getHostProps(this,o),t.getReactMountReady().enqueue(l,this);break;case\"textarea\":A.mountWrapper(this,o,e),o=A.getHostProps(this,o),t.getReactMountReady().enqueue(l,this)}i(this,o);var a,f;null!=e?(a=e._namespaceURI,f=e._tag):n._tag&&(a=n._namespaceURI,f=n._tag),(null==a||a===b.svg&&\"foreignobject\"===f)&&(a=b.html),a===b.html&&(\"svg\"===this._tag?a=b.svg:\"math\"===this._tag&&(a=b.mathml)),this._namespaceURI=a;var p;if(t.useCreateElement){var h,d=n._ownerDocument;if(a===b.html)if(\"script\"===this._tag){var v=d.createElement(\"div\"),g=this._currentElement.type;v.innerHTML=\"<\"+g+\"></\"+g+\">\",h=v.removeChild(v.firstChild)}else h=o.is?d.createElement(this._currentElement.type,o.is):d.createElement(this._currentElement.type);else h=d.createElementNS(a,this._currentElement.type);T.precacheNode(this,h),this._flags|=R.hasCachedChildNodes,this._hostParent||w.setAttributeForRoot(h),this._updateDOMProperties(null,o,t);var y=_(h);this._createInitialChildren(t,o,r,y),p=y}else{var x=this._createOpenTagMarkupAndPutListeners(t,o),C=this._createContentMarkup(t,o,r);p=!C&&Y[this._tag]?x+\"/>\":x+\">\"+C+\"</\"+this._currentElement.type+\">\"}switch(this._tag){case\"input\":t.getReactMountReady().enqueue(u,this),o.autoFocus&&t.getReactMountReady().enqueue(m.focusDOMComponent,this);break;case\"textarea\":t.getReactMountReady().enqueue(c,this),o.autoFocus&&t.getReactMountReady().enqueue(m.focusDOMComponent,this);break;case\"select\":o.autoFocus&&t.getReactMountReady().enqueue(m.focusDOMComponent,this);break;case\"button\":o.autoFocus&&t.getReactMountReady().enqueue(m.focusDOMComponent,this);break;case\"option\":t.getReactMountReady().enqueue(s,this)}return p},_createOpenTagMarkupAndPutListeners:function(t,e){var n=\"<\"+this._currentElement.type;for(var r in e)if(e.hasOwnProperty(r)){var i=e[r];if(null!=i)if(j.hasOwnProperty(r))i&&o(this,r,i,t);else{r===W&&(i&&(i=this._previousStyleCopy=g({},e.style)),i=y.createMarkupForStyles(i,this));var a=null;null!=this._tag&&h(this._tag,e)?z.hasOwnProperty(r)||(a=w.createMarkupForCustomAttribute(r,i)):a=w.createMarkupForProperty(r,i),a&&(n+=\" \"+a)}}return t.renderToStaticMarkup?n:(this._hostParent||(n+=\" \"+w.createMarkupForRoot()),n+=\" \"+w.createMarkupForID(this._domID))},_createContentMarkup:function(t,e,n){var r=\"\",i=e.dangerouslySetInnerHTML;if(null!=i)null!=i.__html&&(r=i.__html);else{var o=B[typeof e.children]?e.children:null,a=null!=o?null:e.children;if(null!=o)r=D(o);else if(null!=a){var u=this.mountChildren(a,t,n);r=u.join(\"\")}}return K[this._tag]&&\"\\n\"===r.charAt(0)?\"\\n\"+r:r},_createInitialChildren:function(t,e,n,r){var i=e.dangerouslySetInnerHTML;if(null!=i)null!=i.__html&&_.queueHTML(r,i.__html);else{var o=B[typeof e.children]?e.children:null,a=null!=o?null:e.children;if(null!=o)\"\"!==o&&_.queueText(r,o);else if(null!=a)for(var u=this.mountChildren(a,t,n),c=0;c<u.length;c++)_.queueChild(r,u[c])}},receiveComponent:function(t,e,n){var r=this._currentElement;this._currentElement=t,this.updateComponent(e,r,t,n)},updateComponent:function(t,e,n,r){var o=e.props,a=this._currentElement.props;switch(this._tag){case\"input\":o=S.getHostProps(this,o),a=S.getHostProps(this,a);break;case\"option\":o=P.getHostProps(this,o),a=P.getHostProps(this,a);break;case\"select\":o=N.getHostProps(this,o),a=N.getHostProps(this,a);break;case\"textarea\":o=A.getHostProps(this,o),a=A.getHostProps(this,a)}switch(i(this,a),this._updateDOMProperties(o,a,t),this._updateDOMChildren(o,a,t,r),this._tag){case\"input\":S.updateWrapper(this);break;case\"textarea\":A.updateWrapper(this);break;case\"select\":t.getReactMountReady().enqueue(f,this)}},_updateDOMProperties:function(t,e,n){var r,i,a;for(r in t)if(!e.hasOwnProperty(r)&&t.hasOwnProperty(r)&&null!=t[r])if(r===W){var u=this._previousStyleCopy;for(i in u)u.hasOwnProperty(i)&&(a=a||{},a[i]=\"\");this._previousStyleCopy=null}else j.hasOwnProperty(r)?t[r]&&L(this,r):h(this._tag,t)?z.hasOwnProperty(r)||w.deleteValueForAttribute(U(this),r):(x.properties[r]||x.isCustomAttribute(r))&&w.deleteValueForProperty(U(this),r);for(r in e){var c=e[r],s=r===W?this._previousStyleCopy:null!=t?t[r]:void 0;if(e.hasOwnProperty(r)&&c!==s&&(null!=c||null!=s))if(r===W)if(c?c=this._previousStyleCopy=g({},c):this._previousStyleCopy=null,s){for(i in s)!s.hasOwnProperty(i)||c&&c.hasOwnProperty(i)||(a=a||{},a[i]=\"\");for(i in c)c.hasOwnProperty(i)&&s[i]!==c[i]&&(a=a||{},a[i]=c[i])}else a=c;else if(j.hasOwnProperty(r))c?o(this,r,c,n):s&&L(this,r);else if(h(this._tag,e))z.hasOwnProperty(r)||w.setValueForAttribute(U(this),r,c);else if(x.properties[r]||x.isCustomAttribute(r)){var l=U(this);null!=c?w.setValueForProperty(l,r,c):w.deleteValueForProperty(l,r)}}a&&y.setValueForStyles(U(this),a,this)},_updateDOMChildren:function(t,e,n,r){var i=B[typeof t.children]?t.children:null,o=B[typeof e.children]?e.children:null,a=t.dangerouslySetInnerHTML&&t.dangerouslySetInnerHTML.__html,u=e.dangerouslySetInnerHTML&&e.dangerouslySetInnerHTML.__html,c=null!=i?null:t.children,s=null!=o?null:e.children,l=null!=i||null!=a,f=null!=o||null!=u;null!=c&&null==s?this.updateChildren(null,n,r):l&&!f&&this.updateTextContent(\"\"),null!=o?i!==o&&this.updateTextContent(\"\"+o):null!=u?a!==u&&this.updateMarkup(\"\"+u):null!=s&&this.updateChildren(s,n,r)},getHostNode:function(){return U(this)},unmountComponent:function(t){switch(this._tag){case\"audio\":case\"form\":case\"iframe\":case\"img\":case\"link\":case\"object\":case\"source\":case\"video\":var e=this._wrapperState.listeners;if(e)for(var n=0;n<e.length;n++)e[n].remove();break;case\"html\":case\"head\":case\"body\":v(\"66\",this._tag)}this.unmountChildren(t),T.uncacheNode(this),C.deleteAllListeners(this),this._rootNodeID=0,this._domID=0,this._wrapperState=null},getPublicInstance:function(){return U(this)}},g(d.prototype,d.Mixin,O.Mixin),t.exports=d},function(t,e,n){\"use strict\";function r(t,e){var n={_topLevelWrapper:t,_idCounter:1,_ownerDocument:e?e.nodeType===i?e:e.ownerDocument:null,_node:e,_tag:e?e.nodeName.toLowerCase():null,_namespaceURI:e?e.namespaceURI:null};return n}var i=(n(96),9);t.exports=r},function(t,e,n){\"use strict\";var r=n(3),i=n(20),o=n(4),a=function(t){this._currentElement=null,this._hostNode=null,this._hostParent=null,this._hostContainerInfo=null,this._domID=0};r(a.prototype,{mountComponent:function(t,e,n,r){var a=n._idCounter++;this._domID=a,this._hostParent=e,this._hostContainerInfo=n;var u=\" react-empty: \"+this._domID+\" \";if(t.useCreateElement){var c=n._ownerDocument,s=c.createComment(u);return o.precacheNode(this,s),i(s)}return t.renderToStaticMarkup?\"\":\"<!--\"+u+\"-->\"},receiveComponent:function(){},getHostNode:function(){return o.getNodeFromInstance(this)},unmountComponent:function(){o.uncacheNode(this)}}),t.exports=a},function(t,e,n){\"use strict\";var r={useCreateElement:!0,useFiber:!1};t.exports=r},function(t,e,n){\"use strict\";var r=n(81),i=n(4),o={dangerouslyProcessChildrenUpdates:function(t,e){var n=i.getNodeFromInstance(t);r.processUpdates(n,e)}};t.exports=o},function(t,e,n){\"use strict\";function r(){this._rootNodeID&&f.updateWrapper(this)}function i(t){var e=this._currentElement.props,n=c.executeOnChange(e,t);l.asap(r,this);var i=e.name;if(\"radio\"===e.type&&null!=i){for(var a=s.getNodeFromInstance(this),u=a;u.parentNode;)u=u.parentNode;for(var f=u.querySelectorAll(\"input[name=\"+JSON.stringify(\"\"+i)+'][type=\"radio\"]'),p=0;p<f.length;p++){var h=f[p];if(h!==a&&h.form===a.form){var d=s.getInstanceFromNode(h);d?void 0:o(\"90\"),l.asap(r,d)}}}return n}var o=n(2),a=n(3),u=n(156),c=n(85),s=n(4),l=n(11),f=(n(0),n(1),{getHostProps:function(t,e){var n=c.getValue(e),r=c.getChecked(e),i=a({type:void 0,step:void 0,min:void 0,max:void 0},e,{defaultChecked:void 0,defaultValue:void 0,value:null!=n?n:t._wrapperState.initialValue,checked:null!=r?r:t._wrapperState.initialChecked,onChange:t._wrapperState.onChange});return i},mountWrapper:function(t,e){var n=e.defaultValue;t._wrapperState={initialChecked:null!=e.checked?e.checked:e.defaultChecked,initialValue:null!=e.value?e.value:n,listeners:null,onChange:i.bind(t)}},updateWrapper:function(t){var e=t._currentElement.props,n=e.checked;null!=n&&u.setValueForProperty(s.getNodeFromInstance(t),\"checked\",n||!1);var r=s.getNodeFromInstance(t),i=c.getValue(e);if(null!=i){var o=\"\"+i;o!==r.value&&(r.value=o)}else null==e.value&&null!=e.defaultValue&&r.defaultValue!==\"\"+e.defaultValue&&(r.defaultValue=\"\"+e.defaultValue),null==e.checked&&null!=e.defaultChecked&&(r.defaultChecked=!!e.defaultChecked)},postMountWrapper:function(t){var e=t._currentElement.props,n=s.getNodeFromInstance(t);switch(e.type){case\"submit\":case\"reset\":break;case\"color\":case\"date\":case\"datetime\":case\"datetime-local\":case\"month\":case\"time\":case\"week\":n.value=\"\",n.value=n.defaultValue;break;default:n.value=n.value}var r=n.name;\"\"!==r&&(n.name=\"\"),n.defaultChecked=!n.defaultChecked,n.defaultChecked=!n.defaultChecked,\"\"!==r&&(n.name=r)}});t.exports=f},function(t,e,n){\"use strict\";function r(t){var e=\"\";return o.Children.forEach(t,function(t){null!=t&&(\"string\"==typeof t||\"number\"==typeof t?e+=t:c||(c=!0))}),e}var i=n(3),o=n(26),a=n(4),u=n(158),c=(n(1),!1),s={mountWrapper:function(t,e,n){var i=null;if(null!=n){var o=n;\"optgroup\"===o._tag&&(o=o._hostParent),null!=o&&\"select\"===o._tag&&(i=u.getSelectValueContext(o))}var a=null;if(null!=i){var c;if(c=null!=e.value?e.value+\"\":r(e.children),a=!1,Array.isArray(i)){for(var s=0;s<i.length;s++)if(\"\"+i[s]===c){a=!0;break}}else a=\"\"+i===c}t._wrapperState={selected:a}},postMountWrapper:function(t){var e=t._currentElement.props;if(null!=e.value){var n=a.getNodeFromInstance(t);n.setAttribute(\"value\",e.value)}},getHostProps:function(t,e){var n=i({selected:void 0,children:void 0},e);null!=t._wrapperState.selected&&(n.selected=t._wrapperState.selected);var o=r(e.children);return o&&(n.children=o),n}};t.exports=s},function(t,e,n){\"use strict\";function r(t,e,n,r){return t===n&&e===r}function i(t){var e=document.selection,n=e.createRange(),r=n.text.length,i=n.duplicate();i.moveToElementText(t),i.setEndPoint(\"EndToStart\",n);var o=i.text.length,a=o+r;return{start:o,end:a}}function o(t){var e=window.getSelection&&window.getSelection();if(!e||0===e.rangeCount)return null;var n=e.anchorNode,i=e.anchorOffset,o=e.focusNode,a=e.focusOffset,u=e.getRangeAt(0);try{u.startContainer.nodeType,u.endContainer.nodeType}catch(t){return null}var c=r(e.anchorNode,e.anchorOffset,e.focusNode,e.focusOffset),s=c?0:u.toString().length,l=u.cloneRange();l.selectNodeContents(t),l.setEnd(u.startContainer,u.startOffset);var f=r(l.startContainer,l.startOffset,l.endContainer,l.endOffset),p=f?0:l.toString().length,h=p+s,d=document.createRange();d.setStart(n,i),d.setEnd(o,a);var v=d.collapsed;return{start:v?h:p,end:v?p:h}}function a(t,e){var n,r,i=document.selection.createRange().duplicate();void 0===e.end?(n=e.start,r=n):e.start>e.end?(n=e.end,r=e.start):(n=e.start,r=e.end),i.moveToElementText(t),i.moveStart(\"character\",n),i.setEndPoint(\"EndToStart\",i),i.moveEnd(\"character\",r-n),i.select()}function u(t,e){if(window.getSelection){var n=window.getSelection(),r=t[l()].length,i=Math.min(e.start,r),o=void 0===e.end?i:Math.min(e.end,r);if(!n.extend&&i>o){var a=o;o=i,i=a}var u=s(t,i),c=s(t,o);if(u&&c){var f=document.createRange();f.setStart(u.node,u.offset),n.removeAllRanges(),i>o?(n.addRange(f),n.extend(c.node,c.offset)):(f.setEnd(c.node,c.offset),n.addRange(f))}}}var c=n(6),s=n(392),l=n(168),f=c.canUseDOM&&\"selection\"in document&&!(\"getSelection\"in window),p={getOffsets:f?i:o,setOffsets:f?a:u};t.exports=p},function(t,e,n){\"use strict\";var r=n(2),i=n(3),o=n(81),a=n(20),u=n(4),c=n(54),s=(n(0),n(96),function(t){this._currentElement=t,this._stringText=\"\"+t,this._hostNode=null,this._hostParent=null,this._domID=0,this._mountIndex=0,this._closingComment=null,this._commentNodes=null});i(s.prototype,{mountComponent:function(t,e,n,r){var i=n._idCounter++,o=\" react-text: \"+i+\" \",s=\" /react-text \";if(this._domID=i,this._hostParent=e,t.useCreateElement){var l=n._ownerDocument,f=l.createComment(o),p=l.createComment(s),h=a(l.createDocumentFragment());return a.queueChild(h,a(f)),this._stringText&&a.queueChild(h,a(l.createTextNode(this._stringText))),a.queueChild(h,a(p)),u.precacheNode(this,f),this._closingComment=p,h}var d=c(this._stringText);return t.renderToStaticMarkup?d:\"<!--\"+o+\"-->\"+d+\"<!--\"+s+\"-->\"},receiveComponent:function(t,e){if(t!==this._currentElement){this._currentElement=t;var n=\"\"+t;if(n!==this._stringText){this._stringText=n;var r=this.getHostNode();o.replaceDelimitedText(r[0],r[1],n)}}},getHostNode:function(){var t=this._commentNodes;if(t)return t;if(!this._closingComment)for(var e=u.getNodeFromInstance(this),n=e.nextSibling;;){if(null==n?r(\"67\",this._domID):void 0,8===n.nodeType&&\" /react-text \"===n.nodeValue){this._closingComment=n;break}n=n.nextSibling}return t=[this._hostNode,this._closingComment],this._commentNodes=t,t},unmountComponent:function(){this._closingComment=null,this._commentNodes=null,u.uncacheNode(this)}}),t.exports=s},function(t,e,n){\"use strict\";function r(){this._rootNodeID&&l.updateWrapper(this)}function i(t){var e=this._currentElement.props,n=u.executeOnChange(e,t);return s.asap(r,this),n}var o=n(2),a=n(3),u=n(85),c=n(4),s=n(11),l=(n(0),n(1),{getHostProps:function(t,e){null!=e.dangerouslySetInnerHTML?o(\"91\"):void 0;var n=a({},e,{value:void 0,defaultValue:void 0,children:\"\"+t._wrapperState.initialValue,onChange:t._wrapperState.onChange});return n},mountWrapper:function(t,e){var n=u.getValue(e),r=n;if(null==n){var a=e.defaultValue,c=e.children;null!=c&&(null!=a?o(\"92\"):void 0,Array.isArray(c)&&(c.length<=1?void 0:o(\"93\"),c=c[0]),a=\"\"+c),null==a&&(a=\"\"),r=a}t._wrapperState={initialValue:\"\"+r,listeners:null,onChange:i.bind(t)}},updateWrapper:function(t){var e=t._currentElement.props,n=c.getNodeFromInstance(t),r=u.getValue(e);if(null!=r){var i=\"\"+r;i!==n.value&&(n.value=i),null==e.defaultValue&&(n.defaultValue=i)}null!=e.defaultValue&&(n.defaultValue=e.defaultValue)},postMountWrapper:function(t){var e=c.getNodeFromInstance(t),n=e.textContent;\n",
              "n===t._wrapperState.initialValue&&(e.value=n)}});t.exports=l},function(t,e,n){\"use strict\";function r(t,e){\"_hostNode\"in t?void 0:c(\"33\"),\"_hostNode\"in e?void 0:c(\"33\");for(var n=0,r=t;r;r=r._hostParent)n++;for(var i=0,o=e;o;o=o._hostParent)i++;for(;n-i>0;)t=t._hostParent,n--;for(;i-n>0;)e=e._hostParent,i--;for(var a=n;a--;){if(t===e)return t;t=t._hostParent,e=e._hostParent}return null}function i(t,e){\"_hostNode\"in t?void 0:c(\"35\"),\"_hostNode\"in e?void 0:c(\"35\");for(;e;){if(e===t)return!0;e=e._hostParent}return!1}function o(t){return\"_hostNode\"in t?void 0:c(\"36\"),t._hostParent}function a(t,e,n){for(var r=[];t;)r.push(t),t=t._hostParent;var i;for(i=r.length;i-- >0;)e(r[i],\"captured\",n);for(i=0;i<r.length;i++)e(r[i],\"bubbled\",n)}function u(t,e,n,i,o){for(var a=t&&e?r(t,e):null,u=[];t&&t!==a;)u.push(t),t=t._hostParent;for(var c=[];e&&e!==a;)c.push(e),e=e._hostParent;var s;for(s=0;s<u.length;s++)n(u[s],\"bubbled\",i);for(s=c.length;s-- >0;)n(c[s],\"captured\",o)}var c=n(2);n(0);t.exports={isAncestor:i,getLowestCommonAncestor:r,getParentInstance:o,traverseTwoPhase:a,traverseEnterLeave:u}},function(t,e,n){\"use strict\";function r(){this.reinitializeTransaction()}var i=n(3),o=n(11),a=n(53),u=n(8),c={initialize:u,close:function(){p.isBatchingUpdates=!1}},s={initialize:u,close:o.flushBatchedUpdates.bind(o)},l=[s,c];i(r.prototype,a,{getTransactionWrappers:function(){return l}});var f=new r,p={isBatchingUpdates:!1,batchedUpdates:function(t,e,n,r,i,o){var a=p.isBatchingUpdates;return p.isBatchingUpdates=!0,a?t(e,n,r,i,o):f.perform(t,null,e,n,r,i,o)}};t.exports=p},function(t,e,n){\"use strict\";function r(){C||(C=!0,y.EventEmitter.injectReactEventListener(m),y.EventPluginHub.injectEventPluginOrder(u),y.EventPluginUtils.injectComponentTree(p),y.EventPluginUtils.injectTreeTraversal(d),y.EventPluginHub.injectEventPluginsByName({SimpleEventPlugin:w,EnterLeaveEventPlugin:c,ChangeEventPlugin:a,SelectEventPlugin:x,BeforeInputEventPlugin:o}),y.HostComponent.injectGenericComponentClass(f),y.HostComponent.injectTextComponentClass(v),y.DOMProperty.injectDOMPropertyConfig(i),y.DOMProperty.injectDOMPropertyConfig(s),y.DOMProperty.injectDOMPropertyConfig(b),y.EmptyComponent.injectEmptyComponentFactory(function(t){return new h(t)}),y.Updates.injectReconcileTransaction(_),y.Updates.injectBatchingStrategy(g),y.Component.injectEnvironment(l))}var i=n(331),o=n(333),a=n(335),u=n(337),c=n(338),s=n(341),l=n(343),f=n(346),p=n(4),h=n(348),d=n(356),v=n(354),g=n(357),m=n(361),y=n(362),_=n(367),b=n(372),x=n(373),w=n(374),C=!1;t.exports={inject:r}},function(t,e,n){\"use strict\";var r=\"function\"==typeof Symbol&&Symbol.for&&Symbol.for(\"react.element\")||60103;t.exports=r},function(t,e,n){\"use strict\";function r(t){i.enqueueEvents(t),i.processEventQueue(!1)}var i=n(22),o={handleTopLevel:function(t,e,n,o){var a=i.extractEvents(t,e,n,o);r(a)}};t.exports=o},function(t,e,n){\"use strict\";function r(t){for(;t._hostParent;)t=t._hostParent;var e=f.getNodeFromInstance(t),n=e.parentNode;return f.getClosestInstanceFromNode(n)}function i(t,e){this.topLevelType=t,this.nativeEvent=e,this.ancestors=[]}function o(t){var e=h(t.nativeEvent),n=f.getClosestInstanceFromNode(e),i=n;do t.ancestors.push(i),i=i&&r(i);while(i);for(var o=0;o<t.ancestors.length;o++)n=t.ancestors[o],v._handleTopLevel(t.topLevelType,n,t.nativeEvent,h(t.nativeEvent))}function a(t){var e=d(window);t(e)}var u=n(3),c=n(150),s=n(6),l=n(17),f=n(4),p=n(11),h=n(93),d=n(324);u(i.prototype,{destructor:function(){this.topLevelType=null,this.nativeEvent=null,this.ancestors.length=0}}),l.addPoolingTo(i,l.twoArgumentPooler);var v={_enabled:!0,_handleTopLevel:null,WINDOW_HANDLE:s.canUseDOM?window:null,setHandleTopLevel:function(t){v._handleTopLevel=t},setEnabled:function(t){v._enabled=!!t},isEnabled:function(){return v._enabled},trapBubbledEvent:function(t,e,n){return n?c.listen(n,e,v.dispatchEvent.bind(null,t)):null},trapCapturedEvent:function(t,e,n){return n?c.capture(n,e,v.dispatchEvent.bind(null,t)):null},monitorScrollValue:function(t){var e=a.bind(null,t);c.listen(window,\"scroll\",e)},dispatchEvent:function(t,e){if(v._enabled){var n=i.getPooled(t,e);try{p.batchedUpdates(o,n)}finally{i.release(n)}}}};t.exports=v},function(t,e,n){\"use strict\";var r=n(21),i=n(22),o=n(50),a=n(86),u=n(159),c=n(51),s=n(161),l=n(11),f={Component:a.injection,DOMProperty:r.injection,EmptyComponent:u.injection,EventPluginHub:i.injection,EventPluginUtils:o.injection,EventEmitter:c.injection,HostComponent:s.injection,Updates:l.injection};t.exports=f},function(t,e,n){\"use strict\";var r=n(385),i=/\\/?>/,o=/^<\\!\\-\\-/,a={CHECKSUM_ATTR_NAME:\"data-react-checksum\",addChecksumToMarkup:function(t){var e=r(t);return o.test(t)?t:t.replace(i,\" \"+a.CHECKSUM_ATTR_NAME+'=\"'+e+'\"$&')},canReuseMarkup:function(t,e){var n=e.getAttribute(a.CHECKSUM_ATTR_NAME);n=n&&parseInt(n,10);var i=r(t);return i===n}};t.exports=a},function(t,e,n){\"use strict\";function r(t,e,n){return{type:\"INSERT_MARKUP\",content:t,fromIndex:null,fromNode:null,toIndex:n,afterNode:e}}function i(t,e,n){return{type:\"MOVE_EXISTING\",content:null,fromIndex:t._mountIndex,fromNode:p.getHostNode(t),toIndex:n,afterNode:e}}function o(t,e){return{type:\"REMOVE_NODE\",content:null,fromIndex:t._mountIndex,fromNode:e,toIndex:null,afterNode:null}}function a(t){return{type:\"SET_MARKUP\",content:t,fromIndex:null,fromNode:null,toIndex:null,afterNode:null}}function u(t){return{type:\"TEXT_CONTENT\",content:t,fromIndex:null,fromNode:null,toIndex:null,afterNode:null}}function c(t,e){return e&&(t=t||[],t.push(e)),t}function s(t,e){f.processChildrenUpdates(t,e)}var l=n(2),f=n(86),p=(n(40),n(9),n(15),n(24)),h=n(342),d=(n(8),n(388)),v=(n(0),{Mixin:{_reconcilerInstantiateChildren:function(t,e,n){return h.instantiateChildren(t,e,n)},_reconcilerUpdateChildren:function(t,e,n,r,i,o){var a,u=0;return a=d(e,u),h.updateChildren(t,a,n,r,i,this,this._hostContainerInfo,o,u),a},mountChildren:function(t,e,n){var r=this._reconcilerInstantiateChildren(t,e,n);this._renderedChildren=r;var i=[],o=0;for(var a in r)if(r.hasOwnProperty(a)){var u=r[a],c=0,s=p.mountComponent(u,e,this,this._hostContainerInfo,n,c);u._mountIndex=o++,i.push(s)}return i},updateTextContent:function(t){var e=this._renderedChildren;h.unmountChildren(e,!1);for(var n in e)e.hasOwnProperty(n)&&l(\"118\");var r=[u(t)];s(this,r)},updateMarkup:function(t){var e=this._renderedChildren;h.unmountChildren(e,!1);for(var n in e)e.hasOwnProperty(n)&&l(\"118\");var r=[a(t)];s(this,r)},updateChildren:function(t,e,n){this._updateChildren(t,e,n)},_updateChildren:function(t,e,n){var r=this._renderedChildren,i={},o=[],a=this._reconcilerUpdateChildren(r,t,o,i,e,n);if(a||r){var u,l=null,f=0,h=0,d=0,v=null;for(u in a)if(a.hasOwnProperty(u)){var g=r&&r[u],m=a[u];g===m?(l=c(l,this.moveChild(g,v,f,h)),h=Math.max(g._mountIndex,h),g._mountIndex=f):(g&&(h=Math.max(g._mountIndex,h)),l=c(l,this._mountChildAtIndex(m,o[d],v,f,e,n)),d++),f++,v=p.getHostNode(m)}for(u in i)i.hasOwnProperty(u)&&(l=c(l,this._unmountChild(r[u],i[u])));l&&s(this,l),this._renderedChildren=a}},unmountChildren:function(t){var e=this._renderedChildren;h.unmountChildren(e,t),this._renderedChildren=null},moveChild:function(t,e,n,r){if(t._mountIndex<r)return i(t,e,n)},createChild:function(t,e,n){return r(n,e,t._mountIndex)},removeChild:function(t,e){return o(t,e)},_mountChildAtIndex:function(t,e,n,r,i,o){return t._mountIndex=r,this.createChild(t,n,e)},_unmountChild:function(t,e){var n=this.removeChild(t,e);return t._mountIndex=null,n}}});t.exports=v},function(t,e,n){\"use strict\";function r(t){return!(!t||\"function\"!=typeof t.attachRef||\"function\"!=typeof t.detachRef)}var i=n(2),o=(n(0),{addComponentAsRefTo:function(t,e,n){r(n)?void 0:i(\"119\"),n.attachRef(e,t)},removeComponentAsRefFrom:function(t,e,n){r(n)?void 0:i(\"120\");var o=n.getPublicInstance();o&&o.refs[e]===t.getPublicInstance()&&n.detachRef(e)}});t.exports=o},function(t,e,n){\"use strict\";var r=\"SECRET_DO_NOT_PASS_THIS_OR_YOU_WILL_BE_FIRED\";t.exports=r},function(t,e,n){\"use strict\";function r(t){this.reinitializeTransaction(),this.renderToStaticMarkup=!1,this.reactMountReady=o.getPooled(null),this.useCreateElement=t}var i=n(3),o=n(155),a=n(17),u=n(51),c=n(162),s=(n(9),n(53)),l=n(88),f={initialize:c.getSelectionInformation,close:c.restoreSelection},p={initialize:function(){var t=u.isEnabled();return u.setEnabled(!1),t},close:function(t){u.setEnabled(t)}},h={initialize:function(){this.reactMountReady.reset()},close:function(){this.reactMountReady.notifyAll()}},d=[f,p,h],v={getTransactionWrappers:function(){return d},getReactMountReady:function(){return this.reactMountReady},getUpdateQueue:function(){return l},checkpoint:function(){return this.reactMountReady.checkpoint()},rollback:function(t){this.reactMountReady.rollback(t)},destructor:function(){o.release(this.reactMountReady),this.reactMountReady=null}};i(r.prototype,s,v),a.addPoolingTo(r),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n){\"function\"==typeof t?t(e.getPublicInstance()):o.addComponentAsRefTo(e,t,n)}function i(t,e,n){\"function\"==typeof t?t(null):o.removeComponentAsRefFrom(e,t,n)}var o=n(365),a={};a.attachRefs=function(t,e){if(null!==e&&\"object\"==typeof e){var n=e.ref;null!=n&&r(n,t,e._owner)}},a.shouldUpdateRefs=function(t,e){var n=null,r=null;null!==t&&\"object\"==typeof t&&(n=t.ref,r=t._owner);var i=null,o=null;return null!==e&&\"object\"==typeof e&&(i=e.ref,o=e._owner),n!==i||\"string\"==typeof i&&o!==r},a.detachRefs=function(t,e){if(null!==e&&\"object\"==typeof e){var n=e.ref;null!=n&&i(n,t,e._owner)}},t.exports=a},function(t,e,n){\"use strict\";function r(t){this.reinitializeTransaction(),this.renderToStaticMarkup=t,this.useCreateElement=!1,this.updateQueue=new u(this)}var i=n(3),o=n(17),a=n(53),u=(n(9),n(370)),c=[],s={enqueue:function(){}},l={getTransactionWrappers:function(){return c},getReactMountReady:function(){return s},getUpdateQueue:function(){return this.updateQueue},destructor:function(){},checkpoint:function(){},rollback:function(){}};i(r.prototype,a,l),o.addPoolingTo(r),t.exports=r},function(t,e,n){\"use strict\";function r(t,e){if(!(t instanceof e))throw new TypeError(\"Cannot call a class as a function\")}function i(t,e){}var o=n(88),a=(n(1),function(){function t(e){r(this,t),this.transaction=e}return t.prototype.isMounted=function(t){return!1},t.prototype.enqueueCallback=function(t,e,n){this.transaction.isInTransaction()&&o.enqueueCallback(t,e,n)},t.prototype.enqueueForceUpdate=function(t){this.transaction.isInTransaction()?o.enqueueForceUpdate(t):i(t,\"forceUpdate\")},t.prototype.enqueueReplaceState=function(t,e){this.transaction.isInTransaction()?o.enqueueReplaceState(t,e):i(t,\"replaceState\")},t.prototype.enqueueSetState=function(t,e){this.transaction.isInTransaction()?o.enqueueSetState(t,e):i(t,\"setState\")},t}());t.exports=a},function(t,e,n){\"use strict\";t.exports=\"15.4.2\"},function(t,e,n){\"use strict\";var r={xlink:\"http://www.w3.org/1999/xlink\",xml:\"http://www.w3.org/XML/1998/namespace\"},i={accentHeight:\"accent-height\",accumulate:0,additive:0,alignmentBaseline:\"alignment-baseline\",allowReorder:\"allowReorder\",alphabetic:0,amplitude:0,arabicForm:\"arabic-form\",ascent:0,attributeName:\"attributeName\",attributeType:\"attributeType\",autoReverse:\"autoReverse\",azimuth:0,baseFrequency:\"baseFrequency\",baseProfile:\"baseProfile\",baselineShift:\"baseline-shift\",bbox:0,begin:0,bias:0,by:0,calcMode:\"calcMode\",capHeight:\"cap-height\",clip:0,clipPath:\"clip-path\",clipRule:\"clip-rule\",clipPathUnits:\"clipPathUnits\",colorInterpolation:\"color-interpolation\",colorInterpolationFilters:\"color-interpolation-filters\",colorProfile:\"color-profile\",colorRendering:\"color-rendering\",contentScriptType:\"contentScriptType\",contentStyleType:\"contentStyleType\",cursor:0,cx:0,cy:0,d:0,decelerate:0,descent:0,diffuseConstant:\"diffuseConstant\",direction:0,display:0,divisor:0,dominantBaseline:\"dominant-baseline\",dur:0,dx:0,dy:0,edgeMode:\"edgeMode\",elevation:0,enableBackground:\"enable-background\",end:0,exponent:0,externalResourcesRequired:\"externalResourcesRequired\",fill:0,fillOpacity:\"fill-opacity\",fillRule:\"fill-rule\",filter:0,filterRes:\"filterRes\",filterUnits:\"filterUnits\",floodColor:\"flood-color\",floodOpacity:\"flood-opacity\",focusable:0,fontFamily:\"font-family\",fontSize:\"font-size\",fontSizeAdjust:\"font-size-adjust\",fontStretch:\"font-stretch\",fontStyle:\"font-style\",fontVariant:\"font-variant\",fontWeight:\"font-weight\",format:0,from:0,fx:0,fy:0,g1:0,g2:0,glyphName:\"glyph-name\",glyphOrientationHorizontal:\"glyph-orientation-horizontal\",glyphOrientationVertical:\"glyph-orientation-vertical\",glyphRef:\"glyphRef\",gradientTransform:\"gradientTransform\",gradientUnits:\"gradientUnits\",hanging:0,horizAdvX:\"horiz-adv-x\",horizOriginX:\"horiz-origin-x\",ideographic:0,imageRendering:\"image-rendering\",in:0,in2:0,intercept:0,k:0,k1:0,k2:0,k3:0,k4:0,kernelMatrix:\"kernelMatrix\",kernelUnitLength:\"kernelUnitLength\",kerning:0,keyPoints:\"keyPoints\",keySplines:\"keySplines\",keyTimes:\"keyTimes\",lengthAdjust:\"lengthAdjust\",letterSpacing:\"letter-spacing\",lightingColor:\"lighting-color\",limitingConeAngle:\"limitingConeAngle\",local:0,markerEnd:\"marker-end\",markerMid:\"marker-mid\",markerStart:\"marker-start\",markerHeight:\"markerHeight\",markerUnits:\"markerUnits\",markerWidth:\"markerWidth\",mask:0,maskContentUnits:\"maskContentUnits\",maskUnits:\"maskUnits\",mathematical:0,mode:0,numOctaves:\"numOctaves\",offset:0,opacity:0,operator:0,order:0,orient:0,orientation:0,origin:0,overflow:0,overlinePosition:\"overline-position\",overlineThickness:\"overline-thickness\",paintOrder:\"paint-order\",panose1:\"panose-1\",pathLength:\"pathLength\",patternContentUnits:\"patternContentUnits\",patternTransform:\"patternTransform\",patternUnits:\"patternUnits\",pointerEvents:\"pointer-events\",points:0,pointsAtX:\"pointsAtX\",pointsAtY:\"pointsAtY\",pointsAtZ:\"pointsAtZ\",preserveAlpha:\"preserveAlpha\",preserveAspectRatio:\"preserveAspectRatio\",primitiveUnits:\"primitiveUnits\",r:0,radius:0,refX:\"refX\",refY:\"refY\",renderingIntent:\"rendering-intent\",repeatCount:\"repeatCount\",repeatDur:\"repeatDur\",requiredExtensions:\"requiredExtensions\",requiredFeatures:\"requiredFeatures\",restart:0,result:0,rotate:0,rx:0,ry:0,scale:0,seed:0,shapeRendering:\"shape-rendering\",slope:0,spacing:0,specularConstant:\"specularConstant\",specularExponent:\"specularExponent\",speed:0,spreadMethod:\"spreadMethod\",startOffset:\"startOffset\",stdDeviation:\"stdDeviation\",stemh:0,stemv:0,stitchTiles:\"stitchTiles\",stopColor:\"stop-color\",stopOpacity:\"stop-opacity\",strikethroughPosition:\"strikethrough-position\",strikethroughThickness:\"strikethrough-thickness\",string:0,stroke:0,strokeDasharray:\"stroke-dasharray\",strokeDashoffset:\"stroke-dashoffset\",strokeLinecap:\"stroke-linecap\",strokeLinejoin:\"stroke-linejoin\",strokeMiterlimit:\"stroke-miterlimit\",strokeOpacity:\"stroke-opacity\",strokeWidth:\"stroke-width\",surfaceScale:\"surfaceScale\",systemLanguage:\"systemLanguage\",tableValues:\"tableValues\",targetX:\"targetX\",targetY:\"targetY\",textAnchor:\"text-anchor\",textDecoration:\"text-decoration\",textRendering:\"text-rendering\",textLength:\"textLength\",to:0,transform:0,u1:0,u2:0,underlinePosition:\"underline-position\",underlineThickness:\"underline-thickness\",unicode:0,unicodeBidi:\"unicode-bidi\",unicodeRange:\"unicode-range\",unitsPerEm:\"units-per-em\",vAlphabetic:\"v-alphabetic\",vHanging:\"v-hanging\",vIdeographic:\"v-ideographic\",vMathematical:\"v-mathematical\",values:0,vectorEffect:\"vector-effect\",version:0,vertAdvY:\"vert-adv-y\",vertOriginX:\"vert-origin-x\",vertOriginY:\"vert-origin-y\",viewBox:\"viewBox\",viewTarget:\"viewTarget\",visibility:0,widths:0,wordSpacing:\"word-spacing\",writingMode:\"writing-mode\",x:0,xHeight:\"x-height\",x1:0,x2:0,xChannelSelector:\"xChannelSelector\",xlinkActuate:\"xlink:actuate\",xlinkArcrole:\"xlink:arcrole\",xlinkHref:\"xlink:href\",xlinkRole:\"xlink:role\",xlinkShow:\"xlink:show\",xlinkTitle:\"xlink:title\",xlinkType:\"xlink:type\",xmlBase:\"xml:base\",xmlns:0,xmlnsXlink:\"xmlns:xlink\",xmlLang:\"xml:lang\",xmlSpace:\"xml:space\",y:0,y1:0,y2:0,yChannelSelector:\"yChannelSelector\",z:0,zoomAndPan:\"zoomAndPan\"},o={Properties:{},DOMAttributeNamespaces:{xlinkActuate:r.xlink,xlinkArcrole:r.xlink,xlinkHref:r.xlink,xlinkRole:r.xlink,xlinkShow:r.xlink,xlinkTitle:r.xlink,xlinkType:r.xlink,xmlBase:r.xml,xmlLang:r.xml,xmlSpace:r.xml},DOMAttributeNames:{}};Object.keys(i).forEach(function(t){o.Properties[t]=0,i[t]&&(o.DOMAttributeNames[t]=i[t])}),t.exports=o},function(t,e,n){\"use strict\";function r(t){if(\"selectionStart\"in t&&c.hasSelectionCapabilities(t))return{start:t.selectionStart,end:t.selectionEnd};if(window.getSelection){var e=window.getSelection();return{anchorNode:e.anchorNode,anchorOffset:e.anchorOffset,focusNode:e.focusNode,focusOffset:e.focusOffset}}if(document.selection){var n=document.selection.createRange();return{parentElement:n.parentElement(),text:n.text,top:n.boundingTop,left:n.boundingLeft}}}function i(t,e){if(y||null==v||v!==l())return null;var n=r(v);if(!m||!p(m,n)){m=n;var i=s.getPooled(d.select,g,t,e);return i.type=\"select\",i.target=v,o.accumulateTwoPhaseDispatches(i),i}return null}var o=n(23),a=n(6),u=n(4),c=n(162),s=n(14),l=n(152),f=n(170),p=n(80),h=a.canUseDOM&&\"documentMode\"in document&&document.documentMode<=11,d={select:{phasedRegistrationNames:{bubbled:\"onSelect\",captured:\"onSelectCapture\"},dependencies:[\"topBlur\",\"topContextMenu\",\"topFocus\",\"topKeyDown\",\"topKeyUp\",\"topMouseDown\",\"topMouseUp\",\"topSelectionChange\"]}},v=null,g=null,m=null,y=!1,_=!1,b={eventTypes:d,extractEvents:function(t,e,n,r){if(!_)return null;var o=e?u.getNodeFromInstance(e):window;switch(t){case\"topFocus\":(f(o)||\"true\"===o.contentEditable)&&(v=o,g=e,m=null);break;case\"topBlur\":v=null,g=null,m=null;break;case\"topMouseDown\":y=!0;break;case\"topContextMenu\":case\"topMouseUp\":return y=!1,i(n,r);case\"topSelectionChange\":if(h)break;case\"topKeyDown\":case\"topKeyUp\":return i(n,r)}return null},didPutListener:function(t,e,n){\"onSelect\"===e&&(_=!0)}};t.exports=b},function(t,e,n){\"use strict\";function r(t){return\".\"+t._rootNodeID}function i(t){return\"button\"===t||\"input\"===t||\"select\"===t||\"textarea\"===t}var o=n(2),a=n(150),u=n(23),c=n(4),s=n(375),l=n(376),f=n(14),p=n(379),h=n(381),d=n(52),v=n(378),g=n(382),m=n(383),y=n(25),_=n(384),b=n(8),x=n(91),w=(n(0),{}),C={};[\"abort\",\"animationEnd\",\"animationIteration\",\"animationStart\",\"blur\",\"canPlay\",\"canPlayThrough\",\"click\",\"contextMenu\",\"copy\",\"cut\",\"doubleClick\",\"drag\",\"dragEnd\",\"dragEnter\",\"dragExit\",\"dragLeave\",\"dragOver\",\"dragStart\",\"drop\",\"durationChange\",\"emptied\",\"encrypted\",\"ended\",\"error\",\"focus\",\"input\",\"invalid\",\"keyDown\",\"keyPress\",\"keyUp\",\"load\",\"loadedData\",\"loadedMetadata\",\"loadStart\",\"mouseDown\",\"mouseMove\",\"mouseOut\",\"mouseOver\",\"mouseUp\",\"paste\",\"pause\",\"play\",\"playing\",\"progress\",\"rateChange\",\"reset\",\"scroll\",\"seeked\",\"seeking\",\"stalled\",\"submit\",\"suspend\",\"timeUpdate\",\"touchCancel\",\"touchEnd\",\"touchMove\",\"touchStart\",\"transitionEnd\",\"volumeChange\",\"waiting\",\"wheel\"].forEach(function(t){var e=t[0].toUpperCase()+t.slice(1),n=\"on\"+e,r=\"top\"+e,i={phasedRegistrationNames:{bubbled:n,captured:n+\"Capture\"},dependencies:[r]};w[t]=i,C[r]=i});var M={},k={eventTypes:w,extractEvents:function(t,e,n,r){var i=C[t];if(!i)return null;var a;switch(t){case\"topAbort\":case\"topCanPlay\":case\"topCanPlayThrough\":case\"topDurationChange\":case\"topEmptied\":case\"topEncrypted\":case\"topEnded\":case\"topError\":case\"topInput\":case\"topInvalid\":case\"topLoad\":case\"topLoadedData\":case\"topLoadedMetadata\":case\"topLoadStart\":case\"topPause\":case\"topPlay\":case\"topPlaying\":case\"topProgress\":case\"topRateChange\":case\"topReset\":case\"topSeeked\":case\"topSeeking\":case\"topStalled\":case\"topSubmit\":case\"topSuspend\":case\"topTimeUpdate\":case\"topVolumeChange\":case\"topWaiting\":a=f;break;case\"topKeyPress\":if(0===x(n))return null;case\"topKeyDown\":case\"topKeyUp\":a=h;break;case\"topBlur\":case\"topFocus\":a=p;break;case\"topClick\":if(2===n.button)return null;case\"topDoubleClick\":case\"topMouseDown\":case\"topMouseMove\":case\"topMouseUp\":case\"topMouseOut\":case\"topMouseOver\":case\"topContextMenu\":a=d;break;case\"topDrag\":case\"topDragEnd\":case\"topDragEnter\":case\"topDragExit\":case\"topDragLeave\":case\"topDragOver\":case\"topDragStart\":case\"topDrop\":a=v;break;case\"topTouchCancel\":case\"topTouchEnd\":case\"topTouchMove\":case\"topTouchStart\":a=g;break;case\"topAnimationEnd\":case\"topAnimationIteration\":case\"topAnimationStart\":a=s;break;case\"topTransitionEnd\":a=m;break;case\"topScroll\":a=y;break;case\"topWheel\":a=_;break;case\"topCopy\":case\"topCut\":case\"topPaste\":a=l}a?void 0:o(\"86\",t);var c=a.getPooled(i,e,n,r);return u.accumulateTwoPhaseDispatches(c),c},didPutListener:function(t,e,n){if(\"onClick\"===e&&!i(t._tag)){var o=r(t),u=c.getNodeFromInstance(t);M[o]||(M[o]=a.listen(u,\"click\",b))}},willDeleteListener:function(t,e){if(\"onClick\"===e&&!i(t._tag)){var n=r(t);M[n].remove(),delete M[n]}}};t.exports=k},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o={animationName:null,elapsedTime:null,pseudoElement:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o={clipboardData:function(t){return\"clipboardData\"in t?t.clipboardData:window.clipboardData}};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o={data:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(52),o={dataTransfer:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(25),o={relatedTarget:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o={data:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(25),o=n(91),a=n(389),u=n(92),c={key:a,location:null,ctrlKey:null,shiftKey:null,altKey:null,metaKey:null,repeat:null,locale:null,getModifierState:u,charCode:function(t){return\"keypress\"===t.type?o(t):0},keyCode:function(t){return\"keydown\"===t.type||\"keyup\"===t.type?t.keyCode:0},which:function(t){return\"keypress\"===t.type?o(t):\"keydown\"===t.type||\"keyup\"===t.type?t.keyCode:0}};i.augmentClass(r,c),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(25),o=n(92),a={touches:null,targetTouches:null,changedTouches:null,altKey:null,metaKey:null,ctrlKey:null,shiftKey:null,getModifierState:o};i.augmentClass(r,a),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(14),o={propertyName:null,elapsedTime:null,pseudoElement:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n,r){return i.call(this,t,e,n,r)}var i=n(52),o={deltaX:function(t){return\"deltaX\"in t?t.deltaX:\"wheelDeltaX\"in t?-t.wheelDeltaX:0},deltaY:function(t){return\"deltaY\"in t?t.deltaY:\"wheelDeltaY\"in t?-t.wheelDeltaY:\"wheelDelta\"in t?-t.wheelDelta:0},deltaZ:null,deltaMode:null};i.augmentClass(r,o),t.exports=r},function(t,e,n){\"use strict\";function r(t){for(var e=1,n=0,r=0,o=t.length,a=o&-4;r<a;){for(var u=Math.min(r+4096,a);r<u;r+=4)n+=(e+=t.charCodeAt(r))+(e+=t.charCodeAt(r+1))+(e+=t.charCodeAt(r+2))+(e+=t.charCodeAt(r+3));e%=i,n%=i}for(;r<o;r++)n+=e+=t.charCodeAt(r);return e%=i,n%=i,e|n<<16}var i=65521;t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n){var r=null==e||\"boolean\"==typeof e||\"\"===e;if(r)return\"\";var i=isNaN(e);if(i||0===e||o.hasOwnProperty(t)&&o[t])return\"\"+e;if(\"string\"==typeof e){e=e.trim()}return e+\"px\"}var i=n(154),o=(n(1),i.isUnitlessNumber);t.exports=r},function(t,e,n){\"use strict\";function r(t){if(null==t)return null;if(1===t.nodeType)return t;var e=a.get(t);return e?(e=u(e),e?o.getNodeFromInstance(e):null):void(\"function\"==typeof t.render?i(\"44\"):i(\"45\",Object.keys(t)))}var i=n(2),o=(n(15),n(4)),a=n(40),u=n(167);n(0),n(1);t.exports=r},function(t,e,n){\"use strict\";(function(e){function r(t,e,n,r){if(t&&\"object\"==typeof t){var i=t,o=void 0===i[n];o&&null!=e&&(i[n]=e)}}function i(t,e){if(null==t)return t;var n={};return o(t,r,n),n}var o=(n(84),n(172));n(1);\"undefined\"!=typeof e&&e.env,1,t.exports=i}).call(e,n(153))},function(t,e,n){\"use strict\";function r(t){if(t.key){var e=o[t.key]||t.key;if(\"Unidentified\"!==e)return e}if(\"keypress\"===t.type){var n=i(t);return 13===n?\"Enter\":String.fromCharCode(n)}return\"keydown\"===t.type||\"keyup\"===t.type?a[t.keyCode]||\"Unidentified\":\"\"}var i=n(91),o={Esc:\"Escape\",Spacebar:\" \",Left:\"ArrowLeft\",Up:\"ArrowUp\",Right:\"ArrowRight\",Down:\"ArrowDown\",Del:\"Delete\",Win:\"OS\",Menu:\"ContextMenu\",Apps:\"ContextMenu\",Scroll:\"ScrollLock\",MozPrintableKey:\"Unidentified\"},a={8:\"Backspace\",9:\"Tab\",12:\"Clear\",13:\"Enter\",16:\"Shift\",17:\"Control\",18:\"Alt\",19:\"Pause\",20:\"CapsLock\",27:\"Escape\",32:\" \",33:\"PageUp\",34:\"PageDown\",35:\"End\",36:\"Home\",37:\"ArrowLeft\",38:\"ArrowUp\",39:\"ArrowRight\",40:\"ArrowDown\",45:\"Insert\",46:\"Delete\",112:\"F1\",113:\"F2\",114:\"F3\",115:\"F4\",116:\"F5\",117:\"F6\",118:\"F7\",119:\"F8\",120:\"F9\",121:\"F10\",122:\"F11\",123:\"F12\",144:\"NumLock\",145:\"ScrollLock\",224:\"Meta\"};t.exports=r},function(t,e,n){\"use strict\";function r(t){var e=t&&(i&&t[i]||t[o]);if(\"function\"==typeof e)return e}var i=\"function\"==typeof Symbol&&Symbol.iterator,o=\"@@iterator\";t.exports=r},function(t,e,n){\"use strict\";function r(){return i++}var i=1;t.exports=r},function(t,e,n){\"use strict\";function r(t){for(;t&&t.firstChild;)t=t.firstChild;return t}function i(t){for(;t;){if(t.nextSibling)return t.nextSibling;t=t.parentNode}}function o(t,e){for(var n=r(t),o=0,a=0;n;){if(3===n.nodeType){if(a=o+n.textContent.length,o<=e&&a>=e)return{node:n,offset:e-o};o=a}n=r(i(n))}}t.exports=o},function(t,e,n){\"use strict\";function r(t,e){var n={};return n[t.toLowerCase()]=e.toLowerCase(),n[\"Webkit\"+t]=\"webkit\"+e,n[\"Moz\"+t]=\"moz\"+e,n[\"ms\"+t]=\"MS\"+e,n[\"O\"+t]=\"o\"+e.toLowerCase(),n}function i(t){if(u[t])return u[t];if(!a[t])return t;var e=a[t];for(var n in e)if(e.hasOwnProperty(n)&&n in c)return u[t]=e[n];return\"\"}var o=n(6),a={animationend:r(\"Animation\",\"AnimationEnd\"),animationiteration:r(\"Animation\",\"AnimationIteration\"),animationstart:r(\"Animation\",\"AnimationStart\"),transitionend:r(\"Transition\",\"TransitionEnd\")},u={},c={};o.canUseDOM&&(c=document.createElement(\"div\").style,\"AnimationEvent\"in window||(delete a.animationend.animation,delete a.animationiteration.animation,delete a.animationstart.animation),\"TransitionEvent\"in window||delete a.transitionend.transition),t.exports=i},function(t,e,n){\"use strict\";function r(t){return'\"'+i(t)+'\"'}var i=n(54);t.exports=r},function(t,e,n){\"use strict\";var r=n(163);t.exports=r.renderSubtreeIntoContainer},function(t,e,n){\"use strict\";function r(t,e){var n=l.extractSingleTouch(e);return n?n[t.page]:t.page in e?e[t.page]:e[t.client]+f[t.envScroll]}function i(t,e){var n=r(b.x,e),i=r(b.y,e);return Math.pow(Math.pow(n-t.x,2)+Math.pow(i-t.y,2),.5)}function o(t){return{tapMoveThreshold:g,ignoreMouseThreshold:m,eventTypes:C,extractEvents:function(e,n,o,a){if(!h(e)&&!d(e))return null;if(v(e))_=M();else if(t(_,M()))return null;var u=null,l=i(y,o);return d(e)&&l<g&&(u=s.getPooled(C.touchTap,n,o,a)),h(e)?(y.x=r(b.x,o),y.y=r(b.y,o)):d(e)&&(y.x=0,y.y=0),c.accumulateTwoPhaseDispatches(u),u}}}var a=n(339),u=n(50),c=n(23),s=n(25),l=n(397),f=n(89),p=n(329),h=(a.topLevelTypes,u.isStartish),d=u.isEndish,v=function(t){var e=[\"topTouchCancel\",\"topTouchEnd\",\"topTouchStart\",\"topTouchMove\"];return e.indexOf(t)>=0},g=10,m=750,y={x:null,y:null},_=null,b={x:{page:\"pageX\",client:\"clientX\",envScroll:\"currentPageScrollLeft\"},y:{page:\"pageY\",client:\"clientY\",envScroll:\"currentPageScrollTop\"}},x=[\"topTouchStart\",\"topTouchCancel\",\"topTouchEnd\",\"topTouchMove\"],w=[\"topMouseDown\",\"topMouseMove\",\"topMouseUp\"].concat(x),C={touchTap:{phasedRegistrationNames:{bubbled:p({onTouchTap:null}),captured:p({onTouchTapCapture:null})},dependencies:w}},M=function(){return Date.now?Date.now:function(){return+new Date}}();t.exports=o},function(t,e){var n={extractSingleTouch:function(t){var e=t.touches,n=t.changedTouches,r=e&&e.length>0,i=n&&n.length>0;return!r&&i?n[0]:r?e[0]:t}};t.exports=n},function(t,e){t.exports=function(t,e){if(t&&e-t<750)return!0}},function(t,e,n){\"use strict\";function r(t){var e=/[=:]/g,n={\"=\":\"=0\",\":\":\"=2\"},r=(\"\"+t).replace(e,function(t){return n[t]});return\"$\"+r}function i(t){var e=/(=0|=2)/g,n={\"=0\":\"=\",\"=2\":\":\"},r=\".\"===t[0]&&\"$\"===t[1]?t.substring(2):t.substring(1);return(\"\"+r).replace(e,function(t){return n[t]})}var o={escape:r,unescape:i};t.exports=o},function(t,e,n){\"use strict\";var r=n(28),i=(n(0),function(t){var e=this;if(e.instancePool.length){var n=e.instancePool.pop();return e.call(n,t),n}return new e(t)}),o=function(t,e){var n=this;if(n.instancePool.length){var r=n.instancePool.pop();return n.call(r,t,e),r}return new n(t,e)},a=function(t,e,n){var r=this;if(r.instancePool.length){var i=r.instancePool.pop();return r.call(i,t,e,n),i}return new r(t,e,n)},u=function(t,e,n,r){var i=this;if(i.instancePool.length){var o=i.instancePool.pop();return i.call(o,t,e,n,r),o}return new i(t,e,n,r)},c=function(t){var e=this;t instanceof e?void 0:r(\"25\"),t.destructor(),e.instancePool.length<e.poolSize&&e.instancePool.push(t)},s=10,l=i,f=function(t,e){var n=t;return n.instancePool=[],n.getPooled=e||l,n.poolSize||(n.poolSize=s),n.release=c,n},p={addPoolingTo:f,oneArgumentPooler:i,twoArgumentPooler:o,threeArgumentPooler:a,fourArgumentPooler:u};t.exports=p},function(t,e,n){\"use strict\";function r(t){return(\"\"+t).replace(b,\"$&/\")}function i(t,e){this.func=t,this.context=e,this.count=0}function o(t,e,n){var r=t.func,i=t.context;r.call(i,e,t.count++)}function a(t,e,n){if(null==t)return t;var r=i.getPooled(e,n);m(t,o,r),i.release(r)}function u(t,e,n,r){this.result=t,this.keyPrefix=e,this.func=n,this.context=r,this.count=0}function c(t,e,n){var i=t.result,o=t.keyPrefix,a=t.func,u=t.context,c=a.call(u,e,t.count++);Array.isArray(c)?s(c,i,n,g.thatReturnsArgument):null!=c&&(v.isValidElement(c)&&(c=v.cloneAndReplaceKey(c,o+(!c.key||e&&e.key===c.key?\"\":r(c.key)+\"/\")+n)),i.push(c))}function s(t,e,n,i,o){var a=\"\";null!=n&&(a=r(n)+\"/\");var s=u.getPooled(e,a,i,o);m(t,c,s),u.release(s)}function l(t,e,n){if(null==t)return t;var r=[];return s(t,r,null,e,n),r}function f(t,e,n){return null}function p(t,e){return m(t,f,null)}function h(t){var e=[];return s(t,e,null,g.thatReturnsArgument),e}var d=n(400),v=n(27),g=n(8),m=n(409),y=d.twoArgumentPooler,_=d.fourArgumentPooler,b=/\\/+/g;i.prototype.destructor=function(){this.func=null,this.context=null,this.count=0},d.addPoolingTo(i,y),u.prototype.destructor=function(){this.result=null,this.keyPrefix=null,this.func=null,this.context=null,this.count=0},d.addPoolingTo(u,_);var x={forEach:a,map:l,mapIntoWithKeyPrefixInternal:s,count:p,toArray:h};t.exports=x},function(t,e,n){\"use strict\";function r(t){return t}function i(t,e){var n=b.hasOwnProperty(e)?b[e]:null;w.hasOwnProperty(e)&&(\"OVERRIDE_BASE\"!==n?p(\"73\",e):void 0),t&&(\"DEFINE_MANY\"!==n&&\"DEFINE_MANY_MERGED\"!==n?p(\"74\",e):void 0)}function o(t,e){if(e){\"function\"==typeof e?p(\"75\"):void 0,v.isValidElement(e)?p(\"76\"):void 0;var n=t.prototype,r=n.__reactAutoBindPairs;e.hasOwnProperty(y)&&x.mixins(t,e.mixins);for(var o in e)if(e.hasOwnProperty(o)&&o!==y){var a=e[o],u=n.hasOwnProperty(o);if(i(u,o),x.hasOwnProperty(o))x[o](t,a);else{var l=b.hasOwnProperty(o),f=\"function\"==typeof a,h=f&&!l&&!u&&e.autobind!==!1;if(h)r.push(o,a),n[o]=a;else if(u){var d=b[o];!l||\"DEFINE_MANY_MERGED\"!==d&&\"DEFINE_MANY\"!==d?p(\"77\",d,o):void 0,\"DEFINE_MANY_MERGED\"===d?n[o]=c(n[o],a):\"DEFINE_MANY\"===d&&(n[o]=s(n[o],a))}else n[o]=a}}}else;}function a(t,e){if(e)for(var n in e){var r=e[n];if(e.hasOwnProperty(n)){var i=n in x;i?p(\"78\",n):void 0;var o=n in t;o?p(\"79\",n):void 0,t[n]=r}}}function u(t,e){t&&e&&\"object\"==typeof t&&\"object\"==typeof e?void 0:p(\"80\");for(var n in e)e.hasOwnProperty(n)&&(void 0!==t[n]?p(\"81\",n):void 0,t[n]=e[n]);return t}function c(t,e){return function(){var n=t.apply(this,arguments),r=e.apply(this,arguments);if(null==n)return r;if(null==r)return n;var i={};return u(i,n),u(i,r),i}}function s(t,e){return function(){t.apply(this,arguments),e.apply(this,arguments)}}function l(t,e){var n=e.bind(t);return n;\n",
              "}function f(t){for(var e=t.__reactAutoBindPairs,n=0;n<e.length;n+=2){var r=e[n],i=e[n+1];t[r]=l(t,i)}}var p=n(28),h=n(3),d=n(97),v=n(27),g=(n(175),n(98)),m=n(38),y=(n(0),n(1),\"mixins\"),_=[],b={mixins:\"DEFINE_MANY\",statics:\"DEFINE_MANY\",propTypes:\"DEFINE_MANY\",contextTypes:\"DEFINE_MANY\",childContextTypes:\"DEFINE_MANY\",getDefaultProps:\"DEFINE_MANY_MERGED\",getInitialState:\"DEFINE_MANY_MERGED\",getChildContext:\"DEFINE_MANY_MERGED\",render:\"DEFINE_ONCE\",componentWillMount:\"DEFINE_MANY\",componentDidMount:\"DEFINE_MANY\",componentWillReceiveProps:\"DEFINE_MANY\",shouldComponentUpdate:\"DEFINE_ONCE\",componentWillUpdate:\"DEFINE_MANY\",componentDidUpdate:\"DEFINE_MANY\",componentWillUnmount:\"DEFINE_MANY\",updateComponent:\"OVERRIDE_BASE\"},x={displayName:function(t,e){t.displayName=e},mixins:function(t,e){if(e)for(var n=0;n<e.length;n++)o(t,e[n])},childContextTypes:function(t,e){t.childContextTypes=h({},t.childContextTypes,e)},contextTypes:function(t,e){t.contextTypes=h({},t.contextTypes,e)},getDefaultProps:function(t,e){t.getDefaultProps?t.getDefaultProps=c(t.getDefaultProps,e):t.getDefaultProps=e},propTypes:function(t,e){t.propTypes=h({},t.propTypes,e)},statics:function(t,e){a(t,e)},autobind:function(){}},w={replaceState:function(t,e){this.updater.enqueueReplaceState(this,t),e&&this.updater.enqueueCallback(this,e,\"replaceState\")},isMounted:function(){return this.updater.isMounted(this)}},C=function(){};h(C.prototype,d.prototype,w);var M={createClass:function(t){var e=r(function(t,n,r){this.__reactAutoBindPairs.length&&f(this),this.props=t,this.context=n,this.refs=m,this.updater=r||g,this.state=null;var i=this.getInitialState?this.getInitialState():null;\"object\"!=typeof i||Array.isArray(i)?p(\"82\",e.displayName||\"ReactCompositeComponent\"):void 0,this.state=i});e.prototype=new C,e.prototype.constructor=e,e.prototype.__reactAutoBindPairs=[],_.forEach(o.bind(null,e)),o(e,t),e.getDefaultProps&&(e.defaultProps=e.getDefaultProps()),e.prototype.render?void 0:p(\"83\");for(var n in b)e.prototype[n]||(e.prototype[n]=null);return e},injection:{injectMixin:function(t){_.push(t)}}};t.exports=M},function(t,e,n){\"use strict\";var r=n(27),i=r.createFactory,o={a:i(\"a\"),abbr:i(\"abbr\"),address:i(\"address\"),area:i(\"area\"),article:i(\"article\"),aside:i(\"aside\"),audio:i(\"audio\"),b:i(\"b\"),base:i(\"base\"),bdi:i(\"bdi\"),bdo:i(\"bdo\"),big:i(\"big\"),blockquote:i(\"blockquote\"),body:i(\"body\"),br:i(\"br\"),button:i(\"button\"),canvas:i(\"canvas\"),caption:i(\"caption\"),cite:i(\"cite\"),code:i(\"code\"),col:i(\"col\"),colgroup:i(\"colgroup\"),data:i(\"data\"),datalist:i(\"datalist\"),dd:i(\"dd\"),del:i(\"del\"),details:i(\"details\"),dfn:i(\"dfn\"),dialog:i(\"dialog\"),div:i(\"div\"),dl:i(\"dl\"),dt:i(\"dt\"),em:i(\"em\"),embed:i(\"embed\"),fieldset:i(\"fieldset\"),figcaption:i(\"figcaption\"),figure:i(\"figure\"),footer:i(\"footer\"),form:i(\"form\"),h1:i(\"h1\"),h2:i(\"h2\"),h3:i(\"h3\"),h4:i(\"h4\"),h5:i(\"h5\"),h6:i(\"h6\"),head:i(\"head\"),header:i(\"header\"),hgroup:i(\"hgroup\"),hr:i(\"hr\"),html:i(\"html\"),i:i(\"i\"),iframe:i(\"iframe\"),img:i(\"img\"),input:i(\"input\"),ins:i(\"ins\"),kbd:i(\"kbd\"),keygen:i(\"keygen\"),label:i(\"label\"),legend:i(\"legend\"),li:i(\"li\"),link:i(\"link\"),main:i(\"main\"),map:i(\"map\"),mark:i(\"mark\"),menu:i(\"menu\"),menuitem:i(\"menuitem\"),meta:i(\"meta\"),meter:i(\"meter\"),nav:i(\"nav\"),noscript:i(\"noscript\"),object:i(\"object\"),ol:i(\"ol\"),optgroup:i(\"optgroup\"),option:i(\"option\"),output:i(\"output\"),p:i(\"p\"),param:i(\"param\"),picture:i(\"picture\"),pre:i(\"pre\"),progress:i(\"progress\"),q:i(\"q\"),rp:i(\"rp\"),rt:i(\"rt\"),ruby:i(\"ruby\"),s:i(\"s\"),samp:i(\"samp\"),script:i(\"script\"),section:i(\"section\"),select:i(\"select\"),small:i(\"small\"),source:i(\"source\"),span:i(\"span\"),strong:i(\"strong\"),style:i(\"style\"),sub:i(\"sub\"),summary:i(\"summary\"),sup:i(\"sup\"),table:i(\"table\"),tbody:i(\"tbody\"),td:i(\"td\"),textarea:i(\"textarea\"),tfoot:i(\"tfoot\"),th:i(\"th\"),thead:i(\"thead\"),time:i(\"time\"),title:i(\"title\"),tr:i(\"tr\"),track:i(\"track\"),u:i(\"u\"),ul:i(\"ul\"),var:i(\"var\"),video:i(\"video\"),wbr:i(\"wbr\"),circle:i(\"circle\"),clipPath:i(\"clipPath\"),defs:i(\"defs\"),ellipse:i(\"ellipse\"),g:i(\"g\"),image:i(\"image\"),line:i(\"line\"),linearGradient:i(\"linearGradient\"),mask:i(\"mask\"),path:i(\"path\"),pattern:i(\"pattern\"),polygon:i(\"polygon\"),polyline:i(\"polyline\"),radialGradient:i(\"radialGradient\"),rect:i(\"rect\"),stop:i(\"stop\"),svg:i(\"svg\"),text:i(\"text\"),tspan:i(\"tspan\")};t.exports=o},function(t,e,n){\"use strict\";function r(t,e){return t===e?0!==t||1/t===1/e:t!==t&&e!==e}function i(t){this.message=t,this.stack=\"\"}function o(t){function e(e,n,r,o,a,u,c){o=o||E,u=u||r;if(null==n[r]){var s=w[a];return e?new i(null===n[r]?\"The \"+s+\" `\"+u+\"` is marked as required \"+(\"in `\"+o+\"`, but its value is `null`.\"):\"The \"+s+\" `\"+u+\"` is marked as required in \"+(\"`\"+o+\"`, but its value is `undefined`.\")):null}return t(n,r,o,a,u)}var n=e.bind(null,!1);return n.isRequired=e.bind(null,!0),n}function a(t){function e(e,n,r,o,a,u){var c=e[n],s=y(c);if(s!==t){var l=w[o],f=_(c);return new i(\"Invalid \"+l+\" `\"+a+\"` of type \"+(\"`\"+f+\"` supplied to `\"+r+\"`, expected \")+(\"`\"+t+\"`.\"))}return null}return o(e)}function u(){return o(M.thatReturns(null))}function c(t){function e(e,n,r,o,a){if(\"function\"!=typeof t)return new i(\"Property `\"+a+\"` of component `\"+r+\"` has invalid PropType notation inside arrayOf.\");var u=e[n];if(!Array.isArray(u)){var c=w[o],s=y(u);return new i(\"Invalid \"+c+\" `\"+a+\"` of type \"+(\"`\"+s+\"` supplied to `\"+r+\"`, expected an array.\"))}for(var l=0;l<u.length;l++){var f=t(u,l,r,o,a+\"[\"+l+\"]\",C);if(f instanceof Error)return f}return null}return o(e)}function s(){function t(t,e,n,r,o){var a=t[e];if(!x.isValidElement(a)){var u=w[r],c=y(a);return new i(\"Invalid \"+u+\" `\"+o+\"` of type \"+(\"`\"+c+\"` supplied to `\"+n+\"`, expected a single ReactElement.\"))}return null}return o(t)}function l(t){function e(e,n,r,o,a){if(!(e[n]instanceof t)){var u=w[o],c=t.name||E,s=b(e[n]);return new i(\"Invalid \"+u+\" `\"+a+\"` of type \"+(\"`\"+s+\"` supplied to `\"+r+\"`, expected \")+(\"instance of `\"+c+\"`.\"))}return null}return o(e)}function f(t){function e(e,n,o,a,u){for(var c=e[n],s=0;s<t.length;s++)if(r(c,t[s]))return null;var l=w[a],f=JSON.stringify(t);return new i(\"Invalid \"+l+\" `\"+u+\"` of value `\"+c+\"` \"+(\"supplied to `\"+o+\"`, expected one of \"+f+\".\"))}return Array.isArray(t)?o(e):M.thatReturnsNull}function p(t){function e(e,n,r,o,a){if(\"function\"!=typeof t)return new i(\"Property `\"+a+\"` of component `\"+r+\"` has invalid PropType notation inside objectOf.\");var u=e[n],c=y(u);if(\"object\"!==c){var s=w[o];return new i(\"Invalid \"+s+\" `\"+a+\"` of type \"+(\"`\"+c+\"` supplied to `\"+r+\"`, expected an object.\"))}for(var l in u)if(u.hasOwnProperty(l)){var f=t(u,l,r,o,a+\".\"+l,C);if(f instanceof Error)return f}return null}return o(e)}function h(t){function e(e,n,r,o,a){for(var u=0;u<t.length;u++){var c=t[u];if(null==c(e,n,r,o,a,C))return null}var s=w[o];return new i(\"Invalid \"+s+\" `\"+a+\"` supplied to \"+(\"`\"+r+\"`.\"))}return Array.isArray(t)?o(e):M.thatReturnsNull}function d(){function t(t,e,n,r,o){if(!g(t[e])){var a=w[r];return new i(\"Invalid \"+a+\" `\"+o+\"` supplied to \"+(\"`\"+n+\"`, expected a ReactNode.\"))}return null}return o(t)}function v(t){function e(e,n,r,o,a){var u=e[n],c=y(u);if(\"object\"!==c){var s=w[o];return new i(\"Invalid \"+s+\" `\"+a+\"` of type `\"+c+\"` \"+(\"supplied to `\"+r+\"`, expected `object`.\"))}for(var l in t){var f=t[l];if(f){var p=f(u,l,r,o,a+\".\"+l,C);if(p)return p}}return null}return o(e)}function g(t){switch(typeof t){case\"number\":case\"string\":case\"undefined\":return!0;case\"boolean\":return!t;case\"object\":if(Array.isArray(t))return t.every(g);if(null===t||x.isValidElement(t))return!0;var e=k(t);if(!e)return!1;var n,r=e.call(t);if(e!==t.entries){for(;!(n=r.next()).done;)if(!g(n.value))return!1}else for(;!(n=r.next()).done;){var i=n.value;if(i&&!g(i[1]))return!1}return!0;default:return!1}}function m(t,e){return\"symbol\"===t||(\"Symbol\"===e[\"@@toStringTag\"]||\"function\"==typeof Symbol&&e instanceof Symbol)}function y(t){var e=typeof t;return Array.isArray(t)?\"array\":t instanceof RegExp?\"object\":m(e,t)?\"symbol\":e}function _(t){var e=y(t);if(\"object\"===e){if(t instanceof Date)return\"date\";if(t instanceof RegExp)return\"regexp\"}return e}function b(t){return t.constructor&&t.constructor.name?t.constructor.name:E}var x=n(27),w=n(175),C=n(405),M=n(8),k=n(177),E=(n(1),\"<<anonymous>>\"),T={array:a(\"array\"),bool:a(\"boolean\"),func:a(\"function\"),number:a(\"number\"),object:a(\"object\"),string:a(\"string\"),symbol:a(\"symbol\"),any:u(),arrayOf:c,element:s(),instanceOf:l,node:d(),objectOf:p,oneOf:f,oneOfType:h,shape:v};i.prototype=Error.prototype,t.exports=T},function(t,e,n){\"use strict\";var r=\"SECRET_DO_NOT_PASS_THIS_OR_YOU_WILL_BE_FIRED\";t.exports=r},function(t,e,n){\"use strict\";function r(t,e,n){this.props=t,this.context=e,this.refs=c,this.updater=n||u}function i(){}var o=n(3),a=n(97),u=n(98),c=n(38);i.prototype=a.prototype,r.prototype=new i,r.prototype.constructor=r,o(r.prototype,a.prototype),r.prototype.isPureReactComponent=!0,t.exports=r},function(t,e,n){\"use strict\";t.exports=\"15.4.2\"},function(t,e,n){\"use strict\";function r(t){return o.isValidElement(t)?void 0:i(\"143\"),t}var i=n(28),o=n(27);n(0);t.exports=r},function(t,e,n){\"use strict\";function r(t,e){return t&&\"object\"==typeof t&&null!=t.key?s.escape(t.key):e.toString(36)}function i(t,e,n,o){var p=typeof t;if(\"undefined\"!==p&&\"boolean\"!==p||(t=null),null===t||\"string\"===p||\"number\"===p||\"object\"===p&&t.$$typeof===u)return n(o,t,\"\"===e?l+r(t,0):e),1;var h,d,v=0,g=\"\"===e?l:e+f;if(Array.isArray(t))for(var m=0;m<t.length;m++)h=t[m],d=g+r(h,m),v+=i(h,d,n,o);else{var y=c(t);if(y){var _,b=y.call(t);if(y!==t.entries)for(var x=0;!(_=b.next()).done;)h=_.value,d=g+r(h,x++),v+=i(h,d,n,o);else for(;!(_=b.next()).done;){var w=_.value;w&&(h=w[1],d=g+s.escape(w[0])+f+r(h,0),v+=i(h,d,n,o))}}else if(\"object\"===p){var C=\"\",M=String(t);a(\"31\",\"[object Object]\"===M?\"object with keys {\"+Object.keys(t).join(\", \")+\"}\":M,C)}}return v}function o(t,e,n){return null==t?0:i(t,\"\",e,n)}var a=n(28),u=(n(15),n(174)),c=n(177),s=(n(0),n(399)),l=(n(1),\".\"),f=\":\";t.exports=o},function(t,e,n){\"use strict\";function r(t){return t&&t.__esModule?t:{default:t}}var i=n(41),o=r(i),a=n(182),u=r(a),c=n(183),s=r(c),l=n(181),f=r(l),p=n(180),h=r(p),d=n(179),v=r(d);(0,s.default)(),window.SHAP={SimpleListVisualizer:f.default,AdditiveForceVisualizer:h.default,AdditiveForceArrayVisualizer:v.default,React:o.default,ReactDom:u.default}}]);</script>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_dataset(X, y):\n",
        "    plt.plot(X[:, 0][y==0], X[:, 1][y==0], \"bs\")\n",
        "    plt.plot(X[:, 0][y==1], X[:, 1][y==1], \"g^\")\n",
        "    #plt.axis(axes)\n",
        "    plt.grid(True, which='both')\n",
        "    plt.xlabel(\"$x_1$\")\n",
        "    plt.ylabel(\"$x_2$\", rotation=0)\n",
        "\n",
        "def plot_decision_boundary(clf, X, y, alpha=1.0):\n",
        "    axes=[-1.5, 2.4, -1, 1.5]\n",
        "    x1, x2 = np.meshgrid(np.linspace(axes[0], axes[1], 100),\n",
        "                         np.linspace(axes[2], axes[3], 100))\n",
        "    X_new = np.c_[x1.ravel(), x2.ravel()]\n",
        "    y_pred = clf.predict(X_new).reshape(x1.shape)\n",
        "    \n",
        "    plt.contourf(x1, x2, y_pred, alpha=0.3 * alpha, cmap='Wistia')\n",
        "    plt.contour(x1, x2, y_pred, cmap=\"Greys\", alpha=0.8 * alpha)\n",
        "    colors = [\"#78785c\", \"#c47b27\"]\n",
        "    markers = (\"o\", \"^\")\n",
        "    for idx in (0, 1):\n",
        "        plt.plot(X[:, 0][y == idx], X[:, 1][y == idx],\n",
        "                 color=colors[idx], marker=markers[idx], linestyle=\"none\")\n",
        "    plt.axis(axes)\n",
        "    plt.xlabel(r\"$x_1$\")\n",
        "    plt.ylabel(r\"$x_2$\", rotation=0)"
      ],
      "metadata": {
        "id": "ruWWKAR_Mz5P"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Voting classifier"
      ],
      "metadata": {
        "id": "LTrx75jqLcLg"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`Scikit-Learn` provides a `VotingClassifier` class that’s quite easy to use: just give it a list of name/predictor pairs, and use it like a normal classifier, that’s it! Let’s try it on the moons dataset (this is a toy dataset for binary classification in which the data points are shaped as two interleaving crescent moons). We will load and split the moons dataset into a training set and a test set, then we’ll create and train a voting classifier composed of three diverse classifiers:"
      ],
      "metadata": {
        "id": "PvL__u5fL5Sw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X, y = make_moons(n_samples=500, noise=0.30, random_state=42)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)\n",
        "plot_dataset(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "-AB6jsQHLfpk",
        "outputId": "f4b4489f-8b3b-4f2e-942a-49641835aa4a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here we use [`SciKeras`](https://github.com/adriangb/scikeras) to wrap kerase model into `Scikit-Learn`."
      ],
      "metadata": {
        "id": "qmJ0_x9ezBAO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_model():\n",
        "    model= keras.models.Sequential([keras.layers.Dense(30,activation='relu',input_shape=[2]),\n",
        "                      keras.layers.Dense(20,activation='relu'),\n",
        "                      keras.layers.Dense(1,activation='sigmoid')\n",
        "                           ])\n",
        "    model.compile(optimizer='NAdam',loss='binary_crossentropy',metrics=['accuracy'])\n",
        "    return model"
      ],
      "metadata": {
        "id": "DSARLteqwsY-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dense_model = KerasClassifier(model=get_model, epochs=200, verbose=False)"
      ],
      "metadata": {
        "id": "mufRkMoGxOuu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "voting_clf = VotingClassifier(\n",
        "    estimators=[\n",
        "        ('lr', LogisticRegression(random_state=42)),\n",
        "        ('rf', RandomForestClassifier(random_state=42)),\n",
        "        ('svc', SVC(random_state=42)),\n",
        "        ('dense', dense_model)\n",
        "    ]\n",
        ")\n",
        "voting_clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z2-c2OA5M4aG",
        "outputId": "133a4343-5ec0-45d7-e7f3-7b8330d9a83d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "VotingClassifier(estimators=[('lr', LogisticRegression(random_state=42)),\n",
              "                             ('rf', RandomForestClassifier(random_state=42)),\n",
              "                             ('svc', SVC(random_state=42)),\n",
              "                             ('dense',\n",
              "                              KerasClassifier(epochs=200, model=<function get_model at 0x7f9198a875f0>, verbose=False))])"
            ]
          },
          "metadata": {},
          "execution_count": 109
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "When you fit a `VotingClassifier`, it clones every estimator and fits the clones. The original estimators are available via the `estimators` attribute, while the fitted clones are available via the `estimators_` attribute. If you prefer a dict rather than a list, you can use `named_estimators` or `named_estimators_` instead. For example, let’s look at each fitted classifier’s accuracy on the test set:"
      ],
      "metadata": {
        "id": "HMfYwrvVNeH3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "for name, clf in voting_clf.named_estimators_.items():\n",
        "    print(name, \"=\", clf.score(X_test, y_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Y1o7PFQNIPg",
        "outputId": "7853f87b-55f2-4ab1-8482-33fbef160914"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "lr = 0.864\n",
            "rf = 0.896\n",
            "svc = 0.896\n",
            "dense = 0.896\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "When you call the voting classifier’s `predict()` method, it performs hard voting. For example, the voting classifier predicts class 1 for the first instance of the test set, because 3 out of 4 classifiers predict that class:"
      ],
      "metadata": {
        "id": "pcfc8mtjNywf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "voting_clf.predict(X_test[:1]),  [clf.predict(X_test[:1]) for clf in voting_clf.estimators_]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2TrG5WLoNvGl",
        "outputId": "26dedde2-204a-469f-d971-b87ad8dc555d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([1]), [array([1]), array([1]), array([0]), array([1])])"
            ]
          },
          "metadata": {},
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now let’s look at the performance of the voting classifier on the test set:"
      ],
      "metadata": {
        "id": "FfwmQiZiN-7P"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "voting_clf.score(X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R-4gZYXnN3Q0",
        "outputId": "c64c50e5-7c31-4e8b-d8c3-6ca1e6fb6021"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.88"
            ]
          },
          "metadata": {},
          "execution_count": 112
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_decision_boundary(voting_clf, X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "vQznlFPfU6aY",
        "outputId": "0fe2bafe-b182-4970-83fc-4ed8decdaf39"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "If all classifiers are able to estimate class probabilities (i.e., they all have a `predict_proba()` method), then you can tell `Scikit-Learn` to predict the class with the **highest class probability, averaged over all the individual classifiers.** This is called **soft voting**. It often achieves higher performance than hard voting because it gives more weight to highly confident votes. All you need to do is set the voting classifier’s voting hyperparameter to \"soft\", and ensure that all classifiers can estimate class probabilities. \n",
        "\n",
        "This is not the case for the SVC class by default, so you need to set its probability hyperparameter to True (this will make the SVC class use cross-validation to estimate class probabilities, slowing down training, and it will add a `predict_proba()` method). Let’s try that:"
      ],
      "metadata": {
        "id": "0OCmMXDAOEsI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "voting_clf.voting = \"soft\"\n",
        "voting_clf.named_estimators[\"svc\"].probability = True\n",
        "voting_clf.fit(X_train, y_train)\n",
        "voting_clf.score(X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tex4mcrIOBQ2",
        "outputId": "5dde7cd6-8849-41c6-e165-5a21e7e2aa00"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.912"
            ]
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We reach 91.2% accuracy simply by using soft voting, not bad!"
      ],
      "metadata": {
        "id": "I__SqQ0lOdco"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "For stacking neural ntwork model, you can refer to https://ensemble-pytorch.readthedocs.io/en/latest/ for more details."
      ],
      "metadata": {
        "id": "R7wMvsUO8t9V"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Stacking"
      ],
      "metadata": {
        "id": "PBHcV002qZ1u"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`Scikit-Learn` provides two classes for stacking ensembles: `StackingClassifier` and `StackingRegressor`. For example, you can replace the `VotingClassifier` you used on the moons dataset with a `StackingClassifier`:"
      ],
      "metadata": {
        "id": "DP5DoUCdqbaN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "stacking_clf = StackingClassifier(\n",
        "    estimators=[\n",
        "        ('lr', LogisticRegression(random_state=42)),\n",
        "        ('rf', RandomForestClassifier(random_state=42)),\n",
        "        ('svc', SVC(probability=True, random_state=42))\n",
        "    ],\n",
        "    final_estimator=RandomForestClassifier(random_state=43),\n",
        "    cv=5  # number of cross-validation folds\n",
        ")\n",
        "stacking_clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1rqpE6FZqlN1",
        "outputId": "6de014a5-d7c7-4bac-d60b-6c42c6644494"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "StackingClassifier(cv=5,\n",
              "                   estimators=[('lr', LogisticRegression(random_state=42)),\n",
              "                               ('rf', RandomForestClassifier(random_state=42)),\n",
              "                               ('svc', SVC(probability=True, random_state=42))],\n",
              "                   final_estimator=RandomForestClassifier(random_state=43))"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "For each predictor, the stacking classifier will call `predict_proba()` if available, or it will fallback to `decision_function()` if available, or as a last resort it will call `predict()`. If you don’t provide a final estimator, `StackingClassifier` will use `LogisticRegression`, and `StackingRegressor` will use `RidgeCV`."
      ],
      "metadata": {
        "id": "08TA3D47qsWN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "stacking_clf.score(X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7ms6ZeGCq4_W",
        "outputId": "6147afb3-260d-491c-d36c-dbe82cd0c620"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.928"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You get 92.8% accuracy! which is a bit better than the voting classifier using soft voting, which got 92%."
      ],
      "metadata": {
        "id": "pIeOuOsOq_N-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Baaging and Pasting"
      ],
      "metadata": {
        "id": "I97bOx8CTlf3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`Scikit-Learn` offers a simple API for both bagging and pasting with the `BaggingClassifier` class (or `BaggingRegressor` for regression). The following code trains an ensemble of 500 **Decision Tree classifiers**:⁠ each is trained on 100 training instances randomly sampled from the training set with replacement (this is an example of bagging, but if you want to use pasting instead, just set `bootstrap=False`). The `n_jobs` parameter tells `Scikit-Learn` the number of CPU cores to use for training and predictions, and –1 tells `Scikit-Learn` to use all available cores."
      ],
      "metadata": {
        "id": "QzERUKToTov_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "bag_clf = BaggingClassifier(DecisionTreeClassifier(), n_estimators=500,\n",
        "                            max_samples=100, random_state=42)\n",
        "bag_clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x43B6MDAObRX",
        "outputId": "5e4c6379-1a36-497d-a99f-f949e3e1c8fa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BaggingClassifier(base_estimator=DecisionTreeClassifier(), max_samples=100,\n",
              "                  n_estimators=500, random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Notice that the `BaggingClassifier` automatically performs **soft voting** instead of hard voting if the base classifier can estimate class probabilities (i.e., if it has a `predict_proba()` method), which is the case with Decision Tree classifiers."
      ],
      "metadata": {
        "id": "BF7-HM3KT99Q"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We compares the decision boundary of a single Decision Tree with the decision boundary of a bagging ensemble of 500 trees (from the preceding code), both trained on the moons dataset. As you can see, the ensemble’s predictions will likely generalize much better than the single Decision Tree’s predictions: the ensemble has a comparable bias but a smaller variance (it makes roughly the same number of errors on the training set, but the decision boundary is less irregular)."
      ],
      "metadata": {
        "id": "4lhj-r7nUMn3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "tree_clf = DecisionTreeClassifier(random_state=42)\n",
        "tree_clf.fit(X_train, y_train)\n",
        "\n",
        "fig, axes = plt.subplots(ncols=2, figsize=(10, 4), sharey=True)\n",
        "plt.sca(axes[0])\n",
        "plot_decision_boundary(tree_clf, X_train, y_train)\n",
        "plt.title(\"Decision Tree\")\n",
        "plt.sca(axes[1])\n",
        "plot_decision_boundary(bag_clf, X_train, y_train)\n",
        "plt.title(\"Decision Trees with Bagging\")\n",
        "plt.ylabel(\"\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "8wOuoYCmUMG2",
        "outputId": "cb3cf50f-611e-477e-cf95-6c10cd6c1df5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x288 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmUAAAEYCAYAAAAd24GlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydeZhU1Zn/P6eWXmm62cFmkX1xaVDUoEZFJIqKGBNNZkwmJmbPTPKLiZlMnGyTZCaJM5qZyWQzmkWjJiiCqK0CCoiKCkID0rKD2GwNdFfvXdv5/VELtdx7695bt5buPp/n4aGr7nbu9q33vOc97yuklCgUCoVCoVAoCour0A1QKBQKhUKhUCijTKFQKBQKhaIoUEaZQqFQKBQKRRGgjDKFQqFQKBSKIkAZZQqFQqFQKBRFgDLKFAqFQqFQKIoAZZQpCoYQol4I8SkT63UIISblo00KhaIwKD2wjhDiN0KI7xos/4EQ4pF8tsku6r5GECpPmcIIIcRBYBQQBELATuDPwO+klOECNi0rhBAdCR8rgF4i5wfwBSnlX/LfKoWiuFF6ULwIIa4CHpFSjk347gfAFCnlJ0zuQwJdgAR6gFXAl6SUrY43WKGJ8pQpzLBYSlkFTAB+Cvwz8GBhm5QdUspBsX/Ae0TOMfZdXICFEJ7CtVKhKEqUHvRv6qLXYRIwBPhBYZszsFBGmcI0UkqflPJp4GPAp4QQ5wIIIUqFEP8phHhPCHE86lIvj20nhFgihNgqhGgTQuwTQlwX/X6tEOKz0b+nCCHWCSF8QoiTQoi/JmwvhRBTon9XCyH+LIRoFkIcEkL8qxDCFV12hxBiQ7QtLUKIA0KIRVbOUQhxlRDifSHEPwshjgF/EEK4hBDfjrb9lBDib0KIoQnbfEAI8ZoQolUI0RDtsSoU/RqlB9b1INqm/UKI9mh7btc4ZpkQolsIMTz6+R4hRFAIMTj6+UdCiF9E//6jEOLHQohKoB44KzoM2CGEOCu6y5LoNWoXQrwjhJhr5tyllG3A08CshLZ9WgjRGN3XfiHEF1La/i0hxFEhxBEhxGdT7tUwIcTK6H1/K9ruDQnbJq77RyHE/wkhno0e6w0hxOSEdT8khNgVfT5+FX1WPmvmvIodZZQpLCOlfBN4H/hg9KufAtOA2cAUoBb4HoAQ4mIiwxt3AzXAFcBBjd3+CHiRSM9sLPC/Oof/X6CaSC/uSuAfgE8nLL8E2AUMB34OPCiEEBZPcTQwlIgn4PPAPwE3R493FtAC/F/0/GqBZ4EfR7f5JvCkEGKExWMqFH0SpQfm9CBqOP0PsCjqabwU2Jp6MCllD/BWdP9E/z8EXJbweV3KNp3AIuBIgofvSHTxTcDjRK7308AvzZy0EGJI9Dw3Jnx9ArgRGEzkOt8vhLgguv51wF3ANUTu+1Upu/w/oJPI9fxU9J8RHwd+SOQZ2Av8JHqc4cATwL8Aw4jc30vNnFOfQEqp/ql/uv+ICOY1Gt9vBO4BBJEXbXLCsnnAgejfvwXu19n3WuCz0b//DPwOGKuxniTykrsBPzArYdkXgLXRv+8A9iYsq4huO9rsORIREj9QlrC8EViQ8HkMEAA8RIZuHk7Z3wvApwp979Q/9c/pf0oP7OsBUAm0Ah8ByjO04UdEDDgPcAz4GhFjtwzoBoZF1/sj8OOEtr6fsp8fAKsTPs8Cug2OK4G2aDtDwLtArcH6y4GvRf9+CPiPhGVTUu5VAJiesPzHwIbU+5pwXr9PWHY98G70738AXk9YJoDDsWenr/9TnjKFXWqB08AIImK3OequbwWej34PMA7YZ2J/3yLycr0ZdbF/RmOd4YCXSK8xxqFoW2Ici/0hpeyK/jnIxPETaZaR3mqMCcBTCefXSESwRkWX3RpbFl1+ORGhVigGCkoPMuiBjHizPgZ8ETgaHZqboXPMdUSMrAuA7UQC7q8EPkDE0Dxlof3HEv7uAsqEcWzcBVLKGiIG4K+BV4QQZQBCiEVCiI1CiNPRc7ueyH2AiNfwcMJ+Ev8eQcTA1Ftupt2x+5Z0HBmxzN7PsK8+gzLKFJYRQlxERPg2ACeJ9NzOkVLWRP9Vy0igKERensk6u4ojpTwmpfyclPIsIr3dX8XiCxI4SaS3NSHhu/FAU3ZnlN6clM+HiQw51CT8K5NSNkWXPZyyrFJK+VOH26RQFCVKD8zrgZTyBSnlQiKdtneBB3SO+RowHfgwsE5KuTN6bteTMnRp0M6skFIGgN8DE4FzhRClwJPAfwKjoobbc0SMZ4CjRIaaY4xL+LuZyIxdveVWSDpOdDh6rP7qfQtllClMI4QYLIS4kUh8wiNSyu0yMg3+ASKxBSOj69UKIa6NbvYg8GkhxIJogGytVu9QCHGrECL2YrUQEZikKfZSyhDwN+AnQogqIcQEIjEMuc7D85voMSdE2zpCCLEkuuwRYLEQ4lohhDsapHtVwrkoFP0SpQfW9EAIMUpEJjlUEkm50ZF6Tgnn1gVsBr7CGSPsNSJeNj2j7DgwTAhR7cRJCiHcROLGuoH9QAlQStTAEpFJEx9K2ORvRO7tTCFEBRDPnxa9V8uAHwghKqL3/B9sNu1Z4DwhxM1Rj99XiMSp9QuUUaYww0ohRDuRXuA9wH0kB9P+M5FAzI1CiDZgNZFeHjISBPxp4H7AR0RQEnu2MS4C3hCRfEFPE4lT2K+x3j8RiVnZT6Rn/iiRWIZc8t/RNr0YvQ4biQQQI6U8DCwBvkNErA4TCWJW75aiv6L0wJ4euIgYjUeIDPVeCXzJ4DjriAzPvpnwuQpYr7WylPJd4DFgf3To9Cyt9UzQEL3uLURi4T4spTwtpWwHvkrE+GoB/p7IdYgdv55IHNzLRO9/dFFv9P9/JDIp4xjwcLStsWWmkVKeBG4lMnHjFJE4uU129lWMqOSxCoVCoVAoHEUIMRPYAZRKKYMay39GZNJFxioOGY7jIhJTdruU8uVs9lUMqN68QqFQKBSKrBFCfFhE8tQNAX4GrIwZZEKIGUKI80WEi4E7gadsHudaIURNNM7tO0Ti2jZm2KxPkFejTAjxkBDihBBih87yq0QkGdzW6L/v5bN9CoVCoVAobPMFIrnM9hGZkZo4PFtFJK6sE/gr8F/ACpvHmRc9xklgMXCzlLLb5r6KirwOXwohriAS3PhnKeW5GsuvAr4ppbwxb41SKBQKhUKhKALy6imTUq4nEuCoUCgUCoVCoUigGGPK5olIvbB6IcQ5hW6MQqFQKBQKRT4otor3bwMTpJQdQojriZRwmJq6khDi80RqkFFZWX7h9Bln57WRivzQ1dVDe1sHoVAIt9tN1eBBVFSUAXD82ElCoVDaNm63m1Gjh6d9P3AQhIIeWlpaaDndQkVZCWNHD828WR9j8469J6WUfba+qNKw/o+RfoHSsMwIggE3p0+30NrayuDKMsaMqCl0oxzBSL/ynhJDCHE28IxWTJnGugeBudG8JJpcOHeWfP3NvzjWPkVx0LC1kZXLVxEInJlJ7fV6WHzzQupmz8y4fODixndqJEuXPskTS5dywfSx3PvNjxW6UY7jmnrTZinl3EK3wwmUhvU/zOiT0rBMlHDqeBWPP/4Ey596igWXzOC7X7ip0I1yBCP9KqrhSyHE6GjJBKJTZl1EksMpBhhrVm1IEiuAQCDImlUbAKibPZPFNy+kuqYKgOqaKiVmCoWiKMikX6A0TKFNXocvhRCPESmyOlwI8T7wfSIZi5FS/gb4KPAlIUSQSGmHj0uV3bbP0LC1kTWrNuBrbae6pooFCy+3LTC+1vaM39fNnqkETKFQOEK+9QuUhinSyatRJqX8uwzLfwn8Mk/N6fM4KSJOtGX5k88TDkdsaF9rO8uffB7AVpuqa6o0hS3Wq1QoFH0bpV8KRTpFNXypME8sHiH24vta21m5fBUNWxsL0p76Z16KC1qMcFhS/8xLtva3YOHleL3JfQav18OChZfbbqNCoSgOlH4pFNooo6yPYiZmIZ90d2vXgtX7PhMq3kKh6L8o/VIotCm2lBgKk5iNWejLqHgLhaJ/ovRLodBGGWV9lGKLWSivKKO7q0fze0Vmiim+RqHINUq/+h9Kw5xBDV/2UYotZmHRDfNxu5MfJ7fbxaIb5hekPX2JYouvUShyjdKv/oXSMOdQRlkfpdhiFupmz2TJLdcmtWfJLdeqnpIJii2+RqHINUq/+hdKw5xDDV/2YYotZqHY2tNXGAjxNQpFKsWmF8XWnr6E0jDnUEaZIi+oeAN9ii2+RqFQJKP0yxilYc6hjDKHGUgvr9lzTa3x5mttZ9nSeuqffZlFN8zPeH36+zVdsPByzRp4KqeRohD09/ctRrb6tWxpvenr09+vqdIw51BGmYNovbwrl68C7GWFLmasnKtWvAFAd1dPxuszEK5p7Dz6s2gr+gYD4X0DZ/Qr03Z2jtVXURrmHMooi+JET8Yo2LG/PZxWztUoriDT9Rko11TFsyiywSlPzEB535zSL6Pt7ByrL6M0zBmUUYZzPZm+EOzolHhbOVe9eINM+7J6HIViIOKkJ6bY37di1C+j/Vk9lkKhUmLg3HRevaDGYgl2dDKXjJVz1cpJZGZfVo+jUAxEnExHUMzvW7Hql9H+rB5LoVBGGc71ZIotIWIqToq3lXON5SQqLy9NW5bp+hT7NVUoCo2Tnphift8KrV96RpTSMIWTqOFLnJvOW+zBjk6Kt9VzjcUbWB1+KPZrqlAUGifTERTz+1YM+gXWh1CL+Zoqig9llOHsdN5iDnZ0OpeMnXPN1zYKxUDB6XQExfq+FYN+2d2uWK+povhQw5cUX8mPXKHc6ApF/0Ppl9IvRf9BecqiDISejHKjKxT9E6VfCkX/QBllA4y+Lt79PTO2QqHQR+mXor+jjLJ+QrG/7E60byBkxlYoBiJKv4rnXBWFRcWU9QOczN+TC5xqn5NT4hUKRXGg9Evpl+IMyijrBxT7y+5U+1RmbIWi/6H0S+mX4gxq+LIfUOwvu1Pt05sSL4Tg+/fcV5TDHgqFwhilX0q/FGdQnrJ+QLGX8dDK5A/W26dX7kRKCRTfsIdCociM0i+lX4ozKE9ZP8Dp5JGQObDVbOBrw9ZG/P5A2vcul8jYPq1jLL55Yfw7IURc0GLEhhVUb1Oh6BsUQr+srKP0S5FPlKesH+B08shMga1ay5ctrWflitVp+1qzagOhUDjt+9KyUsP26bUB4K67P8cPf3JXmqDFKJZhD4VCkZl861dsnRXLXkhaZ8WyF9K8VEq/FPlGecr6CU7m7zEKbK2bPVNzOcCmN7cxfkJtUjv0BKa7qyerNoDzZVcUCkVhyKd+AdQ/+3KasRUKhal/9mWlX4qCooyyPkiuc/pkCmw16smlut7tCo+Z4NpcDHsoFIrcUmj9An2jKvV7pV+KfKOMsgKQjSjlIwGhnhCVV5QZLo+1JxEt4QHw9/pp2Nqo22YzYqjKrigU+aev6pcQwlBztFD6pcg3yijLM9mKkhm3eLYsWHg5K5a9kObe7+3ppWFrI1OnTWTTm9s0t03tQSYOFyT2Qru7ew3P22wvsq+XXVEo+hJ9Rb+0DCkpZbytXq9HMwQjdXak0i9FvlGB/nkm20SE+cjpUzd7JiUl3rTvw2HJmlUb2LP7gO62Wq53vf0ZnbfTwb8KhSJ7+op+Lb55IUKItGWxtno8bs1ttb5X+qXIJ8pTlmeyEaWGrY2a06jB+eDQ7u5eze8ztTMmOqlDHHbOW/UiFYriIlujqryiTDOey2n9qps9k2VL6zWXGbU1pntKvxSFQnnK8ozdRImxYQMtgywXwaGx+DGt7zOdg9Z0cD3UTCOFou+QTaLXhq2N9Pakd/bcbpfj+mWUgLW6psrwPJR+KQqJ8pTlGbszbvTSUAghTLnFLQfn6uTQQcqM56DX1lTUTCOFom+RzYzBNas2EA6n60pJideUR8mKhhkNp8baqnceSr8UhUR5yvKM3VgDvd6alNKUQZYpmWIqesOX3d29Gc/BTM9SxVj0b/xdbeyo/xX+rrZCN0XhINnESunm/NLRmkSsapiZYUW981D6pYDCaVhePWVCiIeAG4ETUspzNZYL4L+B64Eu4A4p5dv5bGM+sBNrYBTXcN+9D2TsNVqd8ZRpSrfRORhte9fdn4t/btjayH33PqCmg/dDDjespu34QQ43rGbyvFsK3RyFg9iNlcomVYVVDTObksLqton6BbnPuaYoHIXSsHx7yv4IXGewfBEwNfrv88Cv89CmPoFeMVuw32s06hHqHW/qtIm22prq6rfjvVP0DfxdbZzY+xYgad67SXnLFIBxQe5M775VDcu1foHSsP5MITUsr0aZlHI9cNpglSXAn2WEjUCNEGJMflpX3KS621Mxmp5tJzi3bvZM6ubMSvu+YcvOjKJjZogj26n1iuLlcMPqeEyilOHIZ8WAx0yqCj2saliu9QuUhvVnCqlhxRboXwscTvj8fvS7o4krCSE+T8STxvjxo/PWuEITc7d//577NJf7Wts1hwHsBudq5SMzm+gx0xBHrvIVFXo4odDHLzSxHqYMhwCQ4RDNezcxru4aSioGF7h1xcNA1jCjVBV6w5h2NCyX+hVrr5XvzVBo/Sj08YuBQmtYsRllppBS/g74HcCFc2fpTBPsvxjFl2llmbZbziPbnGpGx8tFMV4nS7jYEad8lJApdhJ7mDFiPU0VW3aGgaxhVvUr8bOVdzKX+mV0HnY1TOlXcVBoDSs2o6wJGJfweWz0O0UCemVEQL8naCc4Vy9RrdbwQyJaL/eypfUsW1ofF4hcFON1qoSLXXHKRwmZYqf9xKF4DzOGDIdoP3GoQC1SFBt29Ausa1gu9atu9kzHNUzpV3FQaA0rNqPsaeAfhRCPA5cAPinl0QzbZEVfdNfG2mcnY7UVtATN6PsYRnl+YgKx+OaFLL55YdbXPvH+6WH1etgVp3yUkCl2Zi/5eqGbMKBQ+qVPLvULnCkorvSr+Ci0huU7JcZjwFXAcCHE+8D3AS+AlPI3wHNE0mHsJZIS49O5bE+27tpCCGLiMXNdcsmuez7TSxwTiLvu/lxW1yv1/ulh9XrYFSez16sv/pAqig+lX8bkWr9inju710zpl0KLvBplUsq/y7BcAl/JU3OyctcWYvw99Zi5Lrlk1z1vFDMSw4nel5nM23auh1UxN+rt6qUCGehxG4rsUfpljNKv9O+1UPpVXBTb8GVeycZda0cQs+1hGJVaklI63mux6543ihmJ4URv2EzPz871sCLmRr3d8vJSEIJlS+tZs2qDbgmXgRi3ocgepV/GKP3KTr+qa6qYOm0ia1ZtiMfT+f0BpV85ZkAbZdnMnrEqiE70MIxKLf3wJ3eZ2odV7LjnU8UwFad6w1Yyb1vBipjr/dCUV5QRDATT7rdRrIpCYQWlX5lR+mVPv2LbpN5zPZR+OceANsqymT1jVRCd8JBkU6Yk3ySKYa5iEKzePyvtMCvmemLU3dWT9l0gEMx5HE1/w9/Vxu51jzDtyk+oPGcp9Bf9gsyl4vJNMeqXlbZkq1++1nbThdlB6ZcedvRrQBtl2cyesfpCOTGzRc+tHitTAsU5rp9NMGym/YK5+5erWAgz8SeJSCnxej2OpgLpz6gamvr0F/2K7adYNawY9Atyo2FGxrnZe6v0Sx87+tXvjDKrvRq7L5zVF8qJRIOxfT/1xPNp3pZsxvXzOZvG6WOZvX+5iuXS+3HzeNx0d/emrR87ZzV7KTOp9ecGSlWAXHh0tbaDwuiX3rCg3fdxIOgX5EbDjIxzvftUXl5KSWmJ0q8M2NWvfmWU5XtmiJUXyujhtyrCTub3sXrNshGlQs7ccToHT+J1KC8vxeP10N3VE78mgO79zlXPu7+hVX+uv3vL8vmOOKVfsXZbGVYzKhVnhYGiX7HjWfneiEz6FTsfrXu+6MarlX6ZwK5+9SujrFhntsVegMSYIr0fbzMvupleq1nxsXLN7IhSprxEZu6PE71TJ0uipF6H7u5evF4Pt9y6KK1dyiNmj0LXnysUxahhRvpVN3umLV1wSsMGin6BcxpmVr+yGR4f6GSjX/3KKHPaG+IEWrl5Ej0m9937gGURzuR1q3/25aRA85j4vHeoiT27DyS9YFaumdUfDDN5ifSOlbiP5U8+Tzgs4+suf/J5wFrv1MmSKGavg/KI2afQ9ecKRbFpWCb9AnuGpFMa1hf0S6tk03uHmli85Brd7bRwSsOsXAelYfbIRr/6lVGWiyLX2aL3AtQ/+7JheQ2jF12vBwPp7ubEY256c1vS/lcuX4UQac8OoF0fzmpbzc7eMapFV//MS3GDLEY4LKl/5iVTYpHkpq8oi8d6ZdPrK7Yfzv5IoevPFYpi0zC7+gX50TA9ikW/9Pax6c1tjJ9QWxANU/qVe7LRr35llOWiyHW2GKVM0EqbECOTCGv1YLS8bkYYrZvaK2zY2qi7rl5bzb7kRrXotILljb5PJM1N39WjO8xohWx/OFWZkswUuv5coSg2DbOrX5AfDdNDS1PKK8o021xeUaa5Dyf0y2gfdiovOKFhThj+SsOMyUa/XA62o+DUzZ7J4psXxh+u6poqFt+8sKAPi50erl0RttvTKcXPPFcjpQTi36W2e82qDbrb67XV7Lnnygtg5KbPhgULL8frTe7PmL1nMZGN3auYx9LI6FUMHIpNw+y+m/nUMDP6BWgPCRh874R+GS0zc6650LBs9AuUhuWafuUpg+IbA586bWKSyz0TuQgEzcRUcYShdDBVNLFDno3b7Up7Qc0MRSArCAVrgIg7f8E1C1m54mnDnm8mMbDauzXTZkdmXNocRijGQG5FcVFMGmZVvyD/GpaqXxBpdypWve5myi1l0q8FCy/XnS2fi8oLRjg1DKo0LLf0O6Msn5hx4e7ZfcD0/rItr6EnIuXlpYRCYfz+QNo2pfgZJ04iBIzjJHtkLa6SQWnnkcnlHQ5Vs7vRx65dDXF3/oTx47lx8S289FJ9/BpNnTYxbbKB0Yu86Ib5rFj2AqFQOP6d2+1i0Q3zM16PnM64tDmMoOI5FMVCsekX6GuYXuyrln714tVst1U90Ip7s6pfdbNn8t6hpjTDNleVF/RwchhUaVhuUUaZTcxOr85FVuStr79J84sP8ZZ/ImU1Q9NyX2kJrV5eoKniSMrnJnZ0n522nlGsSzg4jIYtTfz5zw+ze/eeuHqeVXsWt99+O1/96nfxlB4zdW6ppJ6TEIJQKBx33xsJSiFmXGai2AK5FQOTQupXw9ZG1r+whqkdDewZVMcV1y6IH1NPw/S8TZr6Jc/WbLcdPXDCa7l4yTWMn1CbpGGJQ5D50DAnvVuF0DCjuL3+hjLKbGL2Idd7gO1mRW7Y2si++kcZJ30RAWr1JompnohotSPWy3SLyAPvFpJxnORY1bS07Y0Mvk5fOe++u4s9u/cQ7Glnwqgq3m/u4MCBA7zzzjucd95Mho0qAfwZz08LrUSGZnIMOZlnx6neYbEFcisGJoXUr5XLVzE9uJehooPR7btZuTziBU80zPTy/SWip197ZC1lNUPTjl3IvFuF1jAnvVv50DAp3XS1D2Ht2rWsW7eeQMDPuFFDHNt/MaOMMpuYfcj1HmC7WZHXv7CGi+SJZHd9IPNMHq12THMfhdQOiIBLR2rHWOgZfFJKwuEwEsnkMVU88O9f5Se/W8mLr+9ESulIL8duT8+p+ByneodmRFbNbFLkmkLp15pVG3AFuhjnShhyDNTa0q/p7mO4NMY1p7uPMmXhTZr7KWS8XiE1zEnvViYNy1a/pPTS0lzDsmXLWbFiBW2+Vm6afwGfXHyZ5bb2RZRRZhOzD7nTvbPR7btjcfSAsbs+UzsmuAT4kgXNjaSsuxkoLuPAbk/PqXNwsndoJLKFLuWiGBgUSr98re2cqzXk2Oo13E6rHeNdgC+ctJ5bSMZXR9YvJv2CwmqY094tPQ1zRL/Cw9m3by9vvvkGvtZWRg4dxLc+vQiXq18li9BFGWU2sfKQO9U787e3MM51Cjfm3PXZtKPYjAM7PT0nzyFfQx92so4nt+kKzh430tE2peLvamP3ukeYduUn+nXJo/5MIfQLYPjgEsZ1mAuZyKYdxaZfUFgN6zv6dTnn13mYPmMi8+dfzalTpznRfIJ//tnv+ck3P0OJN3uTpdj1SxllNilEfMKRtctwuwQkdw4N3fV2sfJy7dy5mf37X+PCC0cTDobYstd81nWzvUA7PT2np27nY+jDSm9aW7Bf4Jprcis0hxtW03b8YL8vedSfKVR81aUjewh0pnxpEDJhF6s1MbMpUm5220JrWN/Qr5jBOYsP3/IhampqeOyxx3j1nQM8tGw9X/zY1Vm3sdj1a0AZZU67s/Mdn9B5eA+klG5IdNc7idmXq2FrIy+8sJpgMIAQArfXw/INmxAis6vZSi/Qzo9ILqZu53pIxEpvWk+wN2yop6Zmhq3jZ+pFxgrtghwQBcKLib6uXwBl3ScJox8y4RRW9MuuJ8rqtgNBw5zQr5jBWVZ5mosvnsP+/fs4eqSJ5taOjMfvD/o1YIyyYnRnW+WcL/80b8cy+3KtWbWBYDA5/1kgFEKIMKFwmNOnW2hr66KsfETavla/8AfNl3L1CxuZPDG9WO/0KbXU1c0BYVzexeo56JEqXlOnTaRhy86cPkNWetN6wtze3kpNjb3jZ+pFJhbaHQgFwouF/qBfkD8Ns6Jfdj1Rdra1aghno2FaxhdYnwFqBSf0KxuDsz/o14AxylQWYmuYfbn0XiApJF5XmHXr1tHT0824sePS1ultO8481z7eDk+hlzOBvm1tLTz8p0fT1i8rL+eKKy5j4pRBuNyZe03ZBLc2bG1k+ZPPx4uh+1rbNTObO/0MWelN6wl2VZU9iyxTLzK2PFZoV4ZDRdvb7G/oFgZ/5iWlXxpkq19mDIOe1lOa+uVkElW7GqalX8uffJ7S0pKc/g46oV928531F/0aMEaZykJsHY/XE3+By8tLNafB671YNZUV/Pp7n+Er//YgG17RrtP2pSs9DHUll0cB6O4O8Nhjj6Wt73K7OXDgAJ/61CeYPLUaXD7D9mcTN1P/zEtxQcuEr7Wd799zn2NDAWZ703qCffnli9ixw3wm9hiZepGJy2MUa2+zv6GnU47CT70AACAASURBVN3dvTRsbVSGmQbZ6JcZw+CcsmaG+tP1y8kkqnY1TEu/wmGpW1LK19rOffc+UBT6ZXdGaH/RrwFjlAkhNPNlCSE01i5O8jXFO3WoBCAYDGmuu2Dh5Tz91OqkIUyv283CuedRWV6Cy+0G0retKgkzwXMqrTxKKBRm//6TmscKhUIEAn6CwRASgZk7ZzduRk+8jMj3kJK2YF/B2eMusGyUmelFtp84FF8eQ4ZDtJ8wP7FDYQ+jmpB9ydufDw2zql92DAN/ewtnhU5Ain7lIhG0HQ3ru/pl73noT/o1YIwyvQSmTiQ2LYTQ5PIFsjLUWzd7Jr1dQ3jhhScJBHoIB0PceuUlVHhK+cy//pZBNVVceMloXC6JEF7KykZSUlLD5J7XcQf3EJtKOlU08Q5TGTRoJJdeep5mu8pKS7lq/hVMnVGFcLXqtr+Q+YnyPSSeLthufKes78dML3L2kq9n0VJFNhiVGsrW21+ozl6uNMyqfsW2sZII9cjaZZF6nETqck4VTRyuPs+Ra5fr++FN8CCmUnj9skd/0q8BY5SZcVPbeRmKUWiyxepQ76xZF7Jnz1EeffRRJgwRzJkygZ8++AyDqgcxY8YohIi5lAMEAie46tLZeF55FBk1yNxCMtHbxtWf/gaeymG67XK7XXjZx64/3kXvOdfz0itb0+6VU/ejvKKM7q70CQUlJV7KK8oMfwj74pB4X+lFDlTqZs+k/tmXNZ/JbDSsWDt72WBVv6wmQpXd7QS3rEOGIt+7kEws8XHzF2/FayKe09/ewr6//Y+mhoEzgfh6+lVeUcaiG+Zrlq2KofSrsAwYo0zPTT112kTuu/eBtAfR7MugG4D77MsFFZpscCIAMxSSTJw0jNTR4WAwwPFXf0dtarI1GaJ96284e/FnDPd7cOVf6Di0i/cOteALjQeS75VTwr/ohvmsWPYCodCZdrrdLm5cck18P1rPDUTiV/oafaUXOZBZdMN8xzVM73156onnDbezQ740zKkAcr1rc2j1X6lNHWGRYZrWLsuoXxDJN6mnYR4NL5aT+rXohvlxI1RPv4QQfS5OsT/p18CoW0BEXBbfvDD+YlbXVFE3ZxYNW3bqikLsZTBCNwC3q4eGrY3ZNToBPUFxMqg0xoKFl+NNyZxsJ1aitFTb5q/0t8Z7mTFkKEjn4d2G+/O3t3ByyzpAUitPUMqZOLbYvXJK+Otmz2TJLdcmPS9Lbrk2SagWLLwclys9ss3vDzh67xUKyI2G6c6elpKVy1f1SQ1zSr/0ro1d/YLMGqbl3TJqix5m9Sv1OkFu7r11BIGeGrZva+Ttt7fg9/upHlRewPbkjwHjKYN0N/V99z6gO7YeI9PLkK8AXKdnqhjhVABmb2+QsrL0mnavhM8BIu1ffPNC0/s9snZZUtxA6qynWFudmmadKd5Bb0gpFAr3qeBrRd/BaQ0z0i+nhxbzpWFO6ZfetYnpFzivYUZtsYoZ/QJ46onn02KrC5kuSkoXXe0jWPXiBh5//HGONDVRN2kId95yRd7bUggGlFGWipneR6aXIZcBuInUzZ7Je4ea2PzWdqSUCCGomzMrZy+NEwGYhw62MH3GKEB7MoWVsiczJ4/mZEIcR2Ldz1iOoNi6VoU/m8DabHq2xVYwWdH3yFbDtN4Xq/s3Sz41zAn9ynRtwHkNKy8vJRgM2cpLZkdL6mbPtP37lTP9Co9i+7ZdPPPMMxw90sTCeefwvS/dhMftzn7ffYABbZQZ9RLB+GVIfCCFSJv4Ed+/UzRsbaRhy854j0ZKScOWnYyfUFu0P+QnT3Yyt2oc4KO9XTunmNmyJ8EJfs2LHOtpxu6V1V5ytoHOdj1z/SVDu6Kw2NWwRP0qryjTNTwGsoalaokeTmrYohuvTjpmPiZrFLJQuhZSuunp6aG3t4cyr+Abd1w3YAwyGOBGmVFPyOhlSH0gtQwyK5njzbyAVgPYC+2FESKSV6yzU3L77V9ixYrf0NaWnsZi8OAaOlrPin9e/cJDmufpO/AuVSR/7xaSobRzuWcP1Vf+PyafPZ+OVph89llM/tyCpHU7dDJo6B3PrOve7pCMqjChcAI7GpaqX91dPbjdLqSUSQlHnfbQ5KtIuJMketz0AuPNlm7yHXyXKpmuYWM5xbGqaVxx7YL4sayca7ZaUuhC6ckIerrDnD59mq6uLgBcfSiXqBMMaKPMbuyB1gMJZxLU5iKdhpUA9kJ6YcrdIXbU/4obLr2a1a9tY/269XR39zB8eDXgI3koUyBlNX/601/i32gZbgDrQrNSvhFUVY1nRugdRvY2cuK1v/HWtiOW26t3PF9rO1J6EMI4XsfuM9SfpqMrCoed509Lv0KhMOUVZZSUeHPmoTH7zBfaixxLWTHltq8lpbjItnTTumCyhnm9HhZN8OM5+DaLZ1Vwts1zy1ZL7DxDudAvKT20tw7lmZX1PPnkkzQ3N/OhOaOpqiyzvc++yIA2ysBe7IHRjKUf/uQuICIssZ6V3kNupbdhxcVcSC9M3TAfbce7GFWzk9/+4E6++MPf89qrrwIwclQVkyYNp6zMQ09PkP37T3Li+K6k7T8wbyLl5emTA9KRtDU3MrSsEeGGIV07+fWLh2n3W5tQrHc8j6eU9pZqBg9tJbUigVYv/q67P2fpuE7XfVMMXKxqmNGM8W/f8+X450waZlVn8lEk3AmOrF1Gx3u70lJcmDVeMg0px3AFuggc2I4byckt66i96hZTec5SsTv8mI2GOa9fbtpbqlm58kUee+wx2tpauf2GS/nyx+bb3F/fJa9GmRDiOuC/ATfweynlT1OW3wHcCzRFv/qllPL3+WyjGTI9kGZ7elZ6G1ZczPnwwiS+1IMHD6GiYjSDvCGmDu4EoHnvJkaMugAhzpRZOnG8nRPHtduQaLDFgoAzcW75CaJ5aXEB10zq5al3rU2b3r//JDNmjMLtPmPMhUJhRo6cGB2WTm6HU734fM6mVSgSMZtIO9NzblVn8lEk3ApGwfhIbUPJKNFsPEavvBS325WUI0yLqeJIJPZFYCnPWSpWtcQJDcuFfkkZCXmRUiKAslJvnyqD6BR5M8pE5Nf5/4CFwPvAW0KIp6WUO1NW/auU8h/z1S47ZHogzfb0rPQ2NHtp11xB3XmXIUPJD251dQ0+X/qwXHV1DTI0wsKZarOtYQsrn15NIBDJsdPW1kJ7u48PTeqMrxMOh3j68QfoDVZx8SWXMHbsWN399fa20NXVhN4sTS1K8TNenCRmS3ncMG9ciJKZiwi4KiydT29vCz09xwmHA7hcXsaOncFHb/07Bg/1QUocm1O9eKem7SsUVjHzg2rmObfqLcnW0+T0pIOMwfgmDaW0GL3uXlwuoZtVHyL6NU6cxB2rdhIK2vaWWdUSJzTMef0KMXhoG4tvuhYpJU899RQPLVvPiVNtfOdzN9rcZ98kn56yi4G9Usr9AEKIx4ElQKpRVvSkPpDlFWUgJcuW1ltKYKoXpOtrbee+ex8wrsF2zVWMGXUBG9bvIhRO7pHV1p5De/tGwgllJ1wuN7W157B+bfaX+9VXn40bZDFKZA+zqtvxxOxDGWbOKD+vNVdw/vnnMW7c+KT1hd9HaeOf6J15B69t3o+WQVZaWglAb29n2rKpIj1+zCVgdtVh/FNus3VesTZVzbmdwdUCIQJp61iN7TMSrVivO7besqX11D/7MkhJd3dvvzPU/F1t7F73CNOu/ES8SLAi/2j9oE6dNjH+DBoNvyV+n41+3XLrIt3nOh9eZC3D5MxwYkRPMxlKsdiz9SeHpe0rHJaUlHj59j1f1pwgoKVfMhyy7S2LaUk8Hm7yGN11rcT25VO/hAgwuBrOOWcWb7/9NqdPn2bTtkZCoeuTRjIKRb70K59GWS1wOOHz+8AlGut9RAhxBbAb+LqU8rDGOgUn8YFM7XHpkdrTM5pynehShvR6aCueqsfteott2/cSCiXHPAEMHuxl1KgyvF5BICA5fryL7dufs3eyKcyaNTjNrTxVHCHV0exxCy4eepzHHnucysrKpGULzzrO7GE+dj37n/TUTNV0U/f0dPD++93U1panZc4fIjrjvcwYQoZo2b+JP77QbOu8Ym16e+mPON11D/Mum4W39ETSOmZ78WaHCLRmwsXob2kyDjespu34waQiwUYoIy53JA7D2dWwbPTL6LnOhxdZ6xyThhNjGHjLYrFno0PDOcnZusfQMjK19ItwyFRVACP04uEScWr4Wmu9bPQr0DuSja/t5OGH/8KuXe8yZmg5v/zXO4rCIIP86VexBfqvBB6TUvYKIb4A/Am4OnUlIcTngc8DjB8/Or8tTEFvJqYWWj29mDhq9aYSS6Skz5YK0t19hObjRxhUln4bT3bASeuTEU0R8FdQUpocHK8lMi4kF0+qYP1RP/6OMy/rIG+Y84b4cAk4b4iPo73dyNL0IceAP8jJI8cI9VZy1riheEsi5ymESMqqDZFJFlvfPBD9dMryOSW2aXLZMd5+8xWmTJlE7fhyJN3x9cz24s0OEWR6fvpLmgx/Vxsn9r4FSJr3bmJc3TUZhcqqCPYl+ouG2dWvTM+1E8lfjdAyTLQ0TK98UrxckpSMc51iT+hMAtjEY4C28ZqqXzF++OW77J1QSpuMPHxODV/rrZdpGy0E5TQf7+Gttzaze/cuJo6s5Df/9gWqKoujtFI+9SufRlkTMC7h81jOBPQDIKVM/DX9PfBzrR1JKX8H/A7gwrmzzAci5QArwadOTzH2egVDBpXw9K++ZboNqdix5rfsPcTyDZsIJHjoNorzufnyucyZMiFt/StvT/687/VlnNjzJjIcosTr5qZawXPH3En787rd3PqhS5jz5eT9/fzxZ2jt7Eo7xpBBldT/7tum2q9FYptcwLjerYRCnyS1PKzZXrzZ+2nm+XEqwLmQnqfDDavjsTpShjMKlR0R7Ev0Nw2zo19OPNd66SsyoWWYbHTXmS6ZlFguye0STOco20JnQjRSjRw7+c6sklTCycDDZ0bD8q9fLsLhcCTQPxzm6ssu0DTICqVh+dSvfPoF3wKmCiEmCiFKgI8DTyeuIIRIHAi/CSj6qs5mX6RM6xkV69VbFggYz+4xQ6I1b5Y5UyZw8+VzqamMeLdqKit0DbJUYg+rjMa7yXAI2byLJRfPNLW/hXPPw5uS3dnrdrNw7nmm25+pTR43jA7tI9TVorl+3eyZ3HX35/jhT+7irrs/pzv9X4vU7808P04FONu5106gdc+b927C39Wmu42WCCpygxMaZke/nHiuE4frrFA3O724u1mDLOaRihclD4cY7zrF8MElpvblVMF0ozbF4uEC7dp5GDNpWDHqFxRGw/KtX3nzlEkpg0KIfwReIJIS4yEp5TtCiH8DNkkpnwa+KoS4iciUt9PAHflqn1lSgx+nTptIw5adhi7c2AuXOm0aIeju6tHdT2y79w41senNbUn7dLncHDvWkdW5ZGPNz5kywZQRlkriwxpDyjDVvnf51sczu3hjx1y1aTutnV3UVFawcO55ttpi1CaBpGPr43DOF23tc+q0iWn3TEt4M9XXcyrAuZCeJ717rtfb1BPB/uYtKxROaVhismwj/QJYseyFpBQRbrcr6+fa7HCdHnaHSFOLikOkgsniWRWcvThz4oBcxMxptSnbNBvLn3w+qcKDyyUKpl9QOA3Lt37lNaZMSvkc8FzKd99L+PtfgH/J9ji5KtGhFfzYsGUndXNmsWf3gSSRS/wceyhTp03H0NtPbLuGLekzJsPhEKNHl9PVUco//cdf0pabYe7gw0yqCOIWEAgGePQPv2Jz27jMG2bBtcPfZag3eWKCDIdo3LGFX6zt1tzGUyooHeRGuECGobcjRLA38pK04+OhAxuADY62yUWYQLM9R22sxl8qWsWX9WbyOj370qr73UnaTxyKC1QMGQ7RfuJQ0nexoYnSQUMtiWB/I5clhpzUsFgNSyP9ik2Gkin3MxQKU//MS4D9SSxmh+ucpvPwnjNesih6sWegfz+djJmz2iYzRCZfyZTPyeRLv6BwGmZWvyCiYQ0rf5GVfhVboH/W5LJEh17w457dBwyzITdsbeSpJ55PEyYz+7nv3gd0eyElJW6mzRjN/n2nOXVK26DRY5A3xEcvPIU7+p65BUwsO8VT77jpCOSu+OvO/UMMlh5P+2bYsHImThqCK9pQ4YaSQS6aTrRYPmezbRo6bBi33XYb1153JZKjlvenF/y6Z/eBpM9W0gRkQ6E9T7OXfN3UerGhCU/rcdMi2N/IdYkhOxqW+JzqYbSPNas2JHlcYnR397L8yecB6+emN1xnNyu+Fc758k8zrxQlXyWjrLTJDGtWbUhLfhsKhZOC9vOlX1BYDTOrXxDRsEB3+ntiRb/6nVHmVHJPrd6NnWDW2EtpZJAZ7SdTkKTLJZgxs5apU6y5iIcffx532ykSSwi53W6+uHAsJ0dea2lfuWTP3g0Eg8kJGN1ul+45+3xHOdG8j2CwB4+njJEjJlNdrZ+zR4sxZ43hA/POpbTiKFYS2sbbYOI5yWd9P6vu90KQODQRDgaYe9t3B+RQZS71q272TMsalvqcGmFHH8NhSf0zL1l+5p0erssVdu5nMRRjz3Qv812ftG9pGLjcXi74yLdtaVi/M8qcKNGh98CVl5cmDTvGMApktDLdXGs/ZuqoBYM93HLrIlPHiLHrwaX0tKUO2YUYVdbGFRb3lUt+8iPtAEmtc96xfTPPPruOYDAQX+fEyd1cdEkd5553oeljejw9uDzpXjswJ5hm8gDls76fFfd7oSjk8GoxkUv9AuvZ8rPVL6NjxtDS1EzkYrguF2RrBDtt7Jg1+DI9J/muTzqQNKzfGWVOlOjQe+A8Xg9er8dSpmmzYqq3n0yBlBA5t5KyJlMvnJSl+H1BPGUV1H1jKd5BQzX2mJ7J3mn87afY/+SPmfzR71JSVQkun+Z6RvezpCwho4qsZO3LzxFMqTQQDARYu3YlF1yknwtKSg+Eh3Ama6QbGRqZtl5qeamIYK6G8GDOr5sTX2/BgutZ+fSypKoHXq+XBQuuj+/XSKy1jm2eaM29aKfySHML3QsX454oOGdKLYMqyrLYd2bsTFn3d7VxYo8K7Ifc6teaVRssZ8vPVr/AnIaBWYNB4PdJXKXVuvol03NpO06ifmlrqARXiyNGsFPGjhWDL9Nz4kTnAYBwNVKWAh7CYV9SFZpErAwhZosdDes41cTxXRuJCW82GtbvjDInSnToPVjdXRHvjBXXsp53DYjXRjPaT+w7o5i02KyoTC+cDFVy6ECQplX/jffodt569Bf4p9yq2/ZcUrJ3abwNFXM/yznnjaOkLD0Tv5n7GQoMZVfjSc16nxC7n8lBqzGkLKX5iJd3dzVqVkZI5PXXV6aVlwoEAjz7zEpOn0p8lTxMmXIR+/dvpbe3i9LSCiZNms3pUx7WvrQdgNLSCnp70/OtlZZWxNexSzAQpHJQBeXlZexvauYb9z4GwIiaCv7nO59iwlnDs9q/EXYSJh5uWJ3eCx6g3rJc6pevtd3yzD8j/UqcfWm0jzMaVp824ggRHTSlX9JDu28YO//6fVxNxaNfWm1wuVzMmDGRBdfMZ+WK57I2gp3I6WbF4Mv0nDjRefD3jGDnjvdpaYno9q5du9mw4VWEkEwam32NZrvY0bA96x8j9ffFrob1O6PMienGMbHR+t7yrBmdKvflFWV8+54vm9pF7Hhavc25F58fz6ht9MLJYA3v7DjNE488yIfKXkO4JKLpNf6wuonOoPXHoNITZMmEY6w4NNry9pWeIF+ceSjehr9sOs2Hp3dz/qf+i6pRARIf7kz3M1Kao5HHHnuc6uoQJSXpkxS83nI620ZROfgEkBC8Gh7Eof0BHn30L2zZskW3lxbjnHNqdMpBdfLLX/7SYMsWNm9OypNMdbWX2trKpPJR4bBk794TbN5stK/MCOHiwropfPjmJTTu2gtI2ts72PnOO9z53d9x7923M2eG/RQietidst52PL32abENTeSL3OrXmWOY3p8D+hU7Jminxlh0w3x9g+HF16iri1Tjaz4qeXrpn5jVvAF3EenXH1ZH3u3E/blcLs4/v46///uPs/gmN2vWrM7oAUSW57QYu1WDz+g5ya7z4KK7YyRrVr/GsmVPxTvT7e0dBPzdfO0T1zH/4sJUMLGjYf6uNrp96eEudjWs3xllkH2JDj2PlJlg/VQSa4GZ+V6PTGJt/MIJOjoq2bJlDSPa3kCURc5DIJk37CgvHLT+ws87u52xlb2mtx/kDXPz1Dae2jOYeaM6EZxpwwcH78LbHubAi7/j/E98DUTyUKbe/RSUc6rZz+bNb9PYuJMxoyqYPGUUroRaaaFQGL/fTdP7zUyfWY0UZxLC+nuHsHPnK7y9eTOnT51gcIU37RiJBPyD0spLRb4PEurW9tLpcbobwv4uxowdgrfEQ8Af5Oj7LbSeTi++bpWObj8bN/m5ffoF3HnnHZSVeejs7GXp0mWsffllfvmn5Tz4H1/L+jip2I2pGDxqEj1tJ5HhEMLlZuTUiwechyyR3OlXZAjLyr6d0i8w1rBlS+s1t/H5fGzf2kI4HOa55+oZdGg5YlRx6de8YZEZ2on783UHeOmll/D5fHzyk3/P17/+LYRHOxm1lG46fcM5evQ0V151PfXPPpmTYuxOGnx2Ow9SevCdHsrKp59j2bJlnDp1ksHlbjxuF0MrPNz1lY9x2ZypltvjFHY07HDDaoTL7Zh+mTLKhBAvA/8upVwlhPgxUC2l/CfbRy1yjDxlVnH6RdB76I2PI5BhCb0+5ozoxBO1WTwuuLg2zBe/+iVL497+rjY2P/kfyJD57fe9vozjuzbyk9smcvJgQzzuw+OCMYPCCAFde1bx7/96nLKaYSxYeCV1s88h4kEJaMabzJ59YVJpjmsuOp+L50yKJ5at9sKE7u28Hb6EcDiMlK6kYsNSQigUIhQOMayqlOX/d7fhOWiVl9IrB1VIPnn3/TR3Ra7pmLOGUzn4GF3to5k+fRpvvLERv8nAbSvYnbJe6HQd/RGjwHqrsUlOe270NEzvOMEg3H//LwiHw7QcO8g/X9KNJ+oMLxb9uuis2LMLc0YHaBkxFm95Fe82NvHWW2/S3t7ONdfM473D22nztSYZMFKWcOr4IJ566im2bt3K1VfP59prP8Irr9TT4zvJRSUHGXHtnY4EzzsxNJ6I1c6DlGUcb/KwdOnj1NfX09XRzj/cdBlfvPUqW7+tTmNHi3KhX2bLLH0fuEcIcTswB/h/to7WR3DSU5aLkhp2jzPCtzFtNMJOCRurJSSSXML7tyDDOuWhZJipoglfaxvPL3ua1//zGxx8p5NNG4+ycvmquGjH4k0atu5I28WcKRP41sdv5N/vvI0PDu1kbKWfc0r3ax5ux45NbNv2IuedN5Qp59SyZa+xqzmb8lJ9HX9XGzvqf6VbWsRoyroRdrdT6OPExKPEfRVKv8Jhye5dR3lv/27eP7iXG6Z04/UkhycUg37JcBAZDsYaxFTRREd3D9Mn1/DNy3ro8R1k+/ZXaIsO08X1a8t+Dh9w8Yc/PMzf/vY3tm/bRn19PaHgIL7+je/x9x8YQU24jZpm7RjThq2N3HfvA3z/nvu4794HaNhqnOw6m9JSTuDvGsHGjVt46aWX6Olq5zufX8yXbpufF4Msk36BPS3KhX6Z8pRJKdeLyJW7C7hKShkSQtwM3AAMBh6UUr5ouxVFRjG4eXNxnAr/sbiXLIbVcW87PYOkB1emG2QiIYHtOE6yR9YyKdyE29fM6498l70l4wkG0+NNVq96mY9/bLZuOzm1C5eAiSVHCXe3AmdmSTVsbeS5Z1cTDAYQQlBS6mH5hk0AhkaW3fJSfR2j4Fd/VxvN+zbbmrLeF6a69zXqZs+k/pmXLKfv0dsX5F+/SkoqOHDgJJ2+Tpb+IjLUfuSV39PVkpzMuRj0KxG3kHENmxhuYlRZD9dN8fNuit0RCAR5/rk1+Hyv88bGjSADlHoFvb299Pp78bedNiwhZTd1htOVBKwQCkl6e3sIBgKcPaqSG66oy9uxMwXv29WwXOiX2eHL84AxwCkpZTuAlHI5sFwIMQT4T6DfGGX5dPNmmyjQyvb7Rt/O3/76V86dNJpffPvvbZ1LtnXA4ExivcMNqzmx582Uh1oyQ7zHWaIFIWCC6z2GhE7yNtPoJTmey9eqnUYj3s6EuA//zifhwjPDk2tWbYjnM4sRCIVYtWn7gDS6jMgU/Hq4YTXhYIBR0+dZjqXI51T3gcSiG692TMPyrV+zZn6INS++SkPDQ7iEYOTQyLM2csldltueipP6VVIxmD2vPE7zvs1p28U1DJjsPUVTeDTtVCSt09nZzmuvbaW6XPCjr32Cu+89Uy7vyPo/GZaQyneeMKfJ53ClmeB9uxqWC/3KOHwphBgD/AVYAnQIIa5LWeVfgf9zvGUFJF9u3lhvJ31YzlzNxWy3t4PVnoGRCGrtyy1gZEKgv9sFQ0UXU0Xy7EWA6ppqzWPGX8Jor9YtJIGD6wi0n46vE7tmpfiZ52qkNJqbrbUzPU2FHbbsPcTPH3+G7zz4N37++DMZh0aLGaPhnlTBMxoeUOSPfGhYrvRrx/ZNjrUxFSf1C6Dl/fRzdQuZpGEAc1z70tbr6QkwfkQ5f/rpV5gyftSZBb0+TjU8l1ZCKtB+ZjKRnn45kToDrA+NFjOZhquLTcMMPWVCiApgGfANKWWjEOJHwM+A56PDmT8F6qWUb+e+qfklF27e1F6h3x/IqrdTiN6S1Z6BkQjOWnhnNOD2zDmEJHgI4xaRlyiWMWIczeyRtXFvmdfr4ZqF8zWPqSWkoWCA5/77m5x902eomz0zniNuqjjCUDqYKprYIc/GLVz88rHs4pm6Ar34Ap3xJA+tnV08se4NVr22gwpvqal9jBw6mFuuuRCPO3d1SM3g72rj+J43QWe4tWsTOwAAIABJREFUJ/Fah8MhGlb+grrF/08F6RcBTmtYvvTr5ZeeY9bMBY61OxEn9cvf1UYo6E9aFpKCDeFZXO7aGdcwIaBK9lBFV9xbFgqFOXr4FBedP4O/Pv8GnV29RDKFCLyHXkjTr2AgwHO/+DFnL77TUL+cSJ2RjxJKvg4/gUAwLY7QacwMVydpWCjAwc3PMe2DH89pu4wwvCJSyi5gXsLn9Qmf/wm4BqgWQkyRUv4mZ63sB2g96HqY7e3kMtFgtsSyIs9aeKfuD/S+19Pr1wlAiPQJFS4Bs0pOssU/JmGY5Fya3kvfr7YHTlIZOM3K5at471ATvT29lOJnnDiJiMay7QqNoeHdFl46oV2qZcTIQUyaNIzSUg+9vUH27z9F84mOtPUu+cAEysqSh1olcKy9lTc2JvfI9fYZlmFefnUT//Uvd1JRbs6QywWHG1bHDbIYsd7muLprkod2ZJhAdzuHNj/H1AKKmsJ58qpfPu3UEfnG39WGp6RUtw6rpn6JiFdMa3Sujr28Is+jpyfI/v0naW7u4PCRM/4Mt9tDbW0tpd3vEA4lh1a4haTS32KoX/tcYzMOT5sZbrbS2bcyfF1aFmbMmDGMGDmCvXv28sXv/4b7v3MngweVG7Y5GzINV2sNT5/c9zZnX3h9wTqWts1UKeX/AP/jYFv6NU7UkNNaL1eJBrPFTFZkLePJpRNqIICxopkbvv09vFXG5xfrDf/o4eV0+5N7soSDbH5rO1JKpoojSYumu49Rc+USzX329Jymvf0wsTi1sjIvs2aNoeqScZSVJZdZaW7eqrmPsjIvH731o6b22dj4Hlv37OHOe37FAz/+Uk7LIxmVFWk7lj5zNeYp0BI8gOb9W5hQQFFTOE9e9at6iKW25YpMGqapX0iqPUFkKN1Yq3b1UjtkBgFXBePHQ2rC5NLSMi6/fB7TZ1Xjcrdx370PpF8fI/1yHTX0ZJn1gJnt7Fv1qLlLjvCBS2cixKd5+OFHaNz1Lp+559f8/kdfoGZwpW67M2GkX5mGq7U1TBbUW9Yvk8cWI07UkEvF6QkJTmE2K7LRUMK+15elTQKQ4RBNa5/k7MV3mGpHmkEW24+U8V5mbIghMmuqmYW3LsJTOSxtm9/+9sekl2mSdHQc5oorLmPWrAsT1j1IW1t6MtnBg2u4445PZNynEG189rN38tBDf2DPrka27T7MpbNzl1DR6Mdn8OhJdPtORNspGDX9A/F1tq64P03wIqcwMEsk9WfyqV/zr76e40etFyl3EjMaZlW/wsLFOWUH6J38UcrKSpk6dQoVFWXENMDlElRUtYKIxDTpXXM9/RoTOk6gvTVplmYieh6w+mdfTjKizHb27YTPeEs6GT9+LJMnT+bAgQOcaO3k+Om2rIwyI/2aveTr8RxzWhqmZbQBtBwuXAydMspyTMy9q0d5eSklpSW2Zi9Zma7e2LiZw4c3cdHFtchwJBA9V7MM7WZ2T0TzZQkH6Ty8y/Q+yktKNA0zIQRTOaLxPbS+/eukWU4xtIwsiAjkiy8upbSiNX7dr7l2nuaPzTXXzmPfwTXx+6VHW1srQ4YMoSY2kcFifrwKT4h/mH4KdzB9aDUVox+fWKHwM4aj5Piu1xk97QNUDjsr/qN0JhnnmfNVCWD7B4XQr1kz53L86Ktp62/ZeyieGLqmsoKFc88rWg3TDKEgzIl9b/Kb5QepqhrMDTfcwE1LFlE99DRCpHsh9YwjI/1KnaWZiFFN58RKD0ad/cThSj30loVDg9m7q4OHH/4Lr732Om6C/PirtzH97DFJ61kpCJ7JeDajYVr6JUMB/F1tBdEvZZTlkFT3biper4dFN16dUcSMxu3NBPM2bN3Jiy+uIhSK5OQSbkzl5DJD6gvkVIbj2A/+t+9fypvvHOIjH/0ot33sVmqGNQPmhlFm1p7FW3v3404ou+T1eqibM4uKzTtwp3ipXEg6D2vHkxllSU/tHer92IB2/dK0Y1Vrzyo1y4XD2xlf1UtH8ytEJk3rY/Tjo1UoHGD3+keZ8+Fvau4jxkAtKN6fKJR+BTScZKnVNFo7u4paw1K9aA8tW8+Dy9ZFFaed9vYOGhoauGjuhdQMGQbiVNo+9IwjO/oF5is9ZKtf5ToxsD1d1Wzf/gYNDdtwSz+//eFnmTZhdNp6VgqCZzKezWhYsemXMspyiFEchtlepb2ZMB4Ss52sf2ENF4W38TZT4rMXncrJlfoCFcMDHhPZsSM/yCO7jjFjRi0ul6S6uoYFH5pH3eyZNEyozehhTPwxKa8ow+UShMPaXqtUsdP6sdEqGq+5L5+PFSt+h9ttvRySv6uNGUO6cAkY5Gsg2Hk6kt5ZZ12jHx+tQuEA3b7jSb1IOwkUrfSGFYWhcPqVztq3NjFX7uhzGial5PfL1vPH5a9QXlnF2LG1AFRUVLJo0XVMnlqLcJ1CUhLfxt/ewr6//Sezbvsa3LxQU6es6ld1TRVTp01k05vbNNvppH51d/dy370PpLdJSoLBIDIcZtzIQZoGmZWC4GaMZzMaZjcBbK40TBllOcTIxXvX3Z8ztQ+r4/ahwHCOH+0lGDzzkI1u38VQcWbadIxsc3JpvUDFkKE9JrKEKmg+0c748YP53Oc+y4xZQ8EV6ZFm8jCm/ph0d/VEPW7aRpmZ4GYrs2K7utooKYERI61N2jjcsDpe3lPIMM2v/Znqj2gHrGb68UksFJ6MSPqBspNA0UpvWFEYCqFfyHJOHDvFzsZGfD4fFdESJKO69/VJDdv73nEeXrEej7eM6667jksvjSQvcLs9DB1aw7GjLUQ60Gc60W2vP0H3oV00rV1O3eI7NK+TVf3ytbbTsGUnXq+bQCDdc+S0ftk1vq0MG5sxns1omN0EsLnSMGWU5ZDy8lLNUid67l0tzKe9EPi7R/LK+q289NLL9PZGjlsmerlx0Jlp04m5vmI1HO2i9QIVOkN7oshy6l2qSuxNt9b6MQmFwpRXlBEMBG1NrjAaPtBCCJg0aYTp9WPnHhutFYQ5ufVp6vee5ANXmps9JsMhmvdt1v1xiq6VVcyYld6wonDkV78gHKpi3+4u/vKXx9mwYQOE/Xz9zg/j72pjnOsUgr6nYRPHjuDKi89l9cZ3WLt2LYcPHzZcv0z0csOgV/AISfPbL3PWFX9HSbX1lDh6xnB5RRkg8qJfVnNmanm+jry7kacPwlUXXZjmEc2kX3pesOiaRathyijLATG3sZagAWglsdGLuzA7E0aGRtGwZS9Llz7Bznfeief6+vD0bmSlJOY+mSHeo0L42SamsXDuebbP0anYMadJ7j1JFkzsxVgGtTEKir3l1kW2SstoxYhkorTM/CuqN717dPtunq9/ktGjZyQt0frxic1USvxx0pwJm8WQtBMTQRS5oxD6hSznWJNgxYpnWL9uHeVeyX3f+RTnThnLvteX4RZEndSSD7re4ZXwOYTdZUWvYR63m3/7ys2MHTWER57ewFunmg3X//D0bqgIgxvCoSD7X3iCGbd9GtCeSa5HseiXFSNOT79Gde9l+YbI94mGWSb9SvSC9SUNU0aZw2QKjoXIi2G0TaLr12zaCym9tLV30N7eRkWp4Ptf/ijl7iCBbY/ER9zcQlLLaYSABSM68e55Fv9Z9sbDiyF2LBVPuCctmelFY8I0d6b/uGRKemj0Y2K3FmBqAG15eSkIQXdXD0IIpMYsy94e8wKoV7ZqCJ0EgwGOH99juH3HqSaO73odSJ496cRwTiz+YuLFN3NiT/EZ84oIhdIvKKG7u5u2Nh+hUIAvfvI6zp0yVqNkGrhkgPO8Rzl77vV9QsOEEHzh1qtYOG8Wp32duutJf2eSXrsI0974IoH2W/FWlaStb6Q1+dCvWJzant0HdI0vKzkzjfTLTPygnn7p7duKhiXGjwE51TBllDmMmSSLVvK9xGI3rPZs5p47iaNbnuGESNadWHLWcPNu2sC2ABVD7FgqIwN700TWBZxTmpwA1UzwsZ0ccGb2qyeIWj+GUsL+/c1wmanT1ylbJXgzPA2AQKBHb1MA9qx/LOHYYVsxF3rBr7H4i93rH01/bpS3rGgoFv0qK4kMT2oZTkLAaNlM1altNGcR05NvDZs0diSTxuov3/f6sjS9RoY5vHo5tdd+HY8niLskkjMwk9YUg355vR4WLFhIb1dkckN3t5+uri6CoSDxoZsEMulXpvhBPf0Ca3GvWhqWGD8G5FTDlFHmMJnctVovRqa4C7s17PTH0yHWHbNr4Rc6dkyLilArUiafr8cNw2VycWAzwcdWcsBZ2a8eqcerqBhMS0uQ5hPZuv+JB0d7vfpVATpONdHtOx7/bLf3pxX8mhh/kXiMxGMV0phXnKGY9AsMNEyGad7/NtnE9BSbhunlZmx6ZwOv+aYwfPhwLrv8IgZVn8yoNYXWr+qaKq6evwh/Tw1PLH0WKcP4fG28+OKLdHV28kEN4zCTfhnFDzqlX7F2JGpYavxYSWU1qRO+nNQwZZQ5jFEwpN6LkatySbHEeLFho23P/W9SLwT6hpdi2LAKjh3byv33vUl1zWAWLLxMUyT2VlzOF249U6T8jW37uOveR7lw7sUkzhUzG3xs1CvUEjtbtUhlOTI0GBCcf+5ozj93PpIS3t35Hg8++KD+dhro1fwcQicet5dRo6YC2sKR2MuMN83is6EX/JostvHgIITLzcipFxf1szfQKCb9goj3RFe/+lBcopnEt4lG4o697/ONnz1MR6+kpMQNPMygQZU0N59kyZLrTWlNXvQrgcTjBXvHsOGVBv7859/S1PQ+UkIoGCQcCvClj13NJ26cl7a9kX55XW7D+EEn9Au0NSw1fkwIN8LlRoZDOdEwZZQ5jJ7bePHNC3V7G7ksl5Q4bKTVCyn6mB6PZNLkoYRCkUBXX2tb3KU+e/aFRlvqks2PiJGL3+p+e1tDvPPwv9Ay4TbC3uRr/8Ybb7Bt23ZKPIKzRpqrBRgT9U/efT/B0kFMmzaKQKCH6poqrvvgDRw8oF3o2d/V5ogHSyv4Na1oeUIPM5tnz+PCm3kthVX6mn5B8WuYncS3504Zyx///Yv8+q9r6erp5fCx07x37CS7dr3LiRMfoLq6Gp/Pl7ZdJg1zUr8i+dT+hym3fS2ltJObzg7Yu3cvTU3v45G9nD9jEkIIlsyfw+UXTNPcX6JRmmrE3mxQvcEp/YJ0DTu4+TlOHWxIih9zwiNnpF/KKHMYO25jO9uYIdOwUQy7PYp8JP8UZeByuZK+i7nU7Rpl2fyIGLn4rew3FBjBtqU/Qp5oZPfO+3jxcHLqC5/PR6knzM++9QnOrjWfFiNGS0sP06dfyfU3LKBy8DG62kdz8MAazXUPN6yO9/xiWO0B6s1kCwf9uj+mYN/LMaLKOybzWgqr9EX9Ante3XwlL161aXvcIIthJnB9zIgh/Ns/fhiAF1/bzg9+tTy+bMHCa1i5YoVlDXNKvwCOrF1Gx3u74qWdpPQiqEDgirzy0fd+9szJ/PwbHzNsVypzpkwwnRTYCf0CbQ07ue9tRMrvTyp2NMxIv5RRlgPsxFBkE3ehR6LVH3tI208coqsluW6a3R5FXpJ/pseDAtamWqeSzY+IkYvf7H4btu5mXf2vubjrFdwC6oa182ZTKZ1Bd3ydKWMq+OHXbmf8mEhx9Fz+iDgR8Kw3k63l/UaDuEb7sRhlJWKQ5Y0Upsi3fknpofWUl1dfXcP27dsRAoZUV2rq1+R5t7B1xf1Za1g+kxfrBahbSXw7ZHAlLgE7drzDq6++yk1LFrF4SS9rVq+zpGHO6FdjpEpMx2u4haR588ucdcUnOOmr5PjxEwA0N5/k9Y1v0NPdzdCayKuaKw1zasKGXkoOI/2yeywj/VJGWT8l0K3tubjgI9/O+oXIa/JPiaZhpuVST32ftFJMxLD7I5LJxW820/b04N74ebmE4J+uG80l131Kd7tsf0SMSpo7EfCsJ4wlFTVc/Hc/zHr/qRxo9u90fKeK/CPLOHbYzdIn/kp9fT3dXR18askVXDBlJG8v0047kO3zmu/kxTWVFZoGmJXEt3PPmcinP/z/2zvzOKmqM+9/T1V309AszSabCKKIoogoQTBEjYKvmoVEo4kmmZiQySR5zcwkk7xJPnlN5s0yWWayTDKJiTFOTDJqXIigISqoSIwg0EILyN6CzdbdQNM7VHfVef+ohVtV91bdqrpbdT/fz6eh+tate05V1/nd5zznOc9zDQ/8aS2PPPJHWltbuf329/OFL86BrGLmUazqA7uhX9FYlHUP/pTVR87mrbfeAjSdnZ20tLQw57xR/OOdiwD3DGGnNmxYbSoZMnKi45tCcumXGGX9lKPbX3QtB4+XyT8romEiujerqHjcpR4jFApRUREmFA7zh6f+xhPPrU+dF4vFiGlFZWUloVAYpWI5jRM7lBo/8/yqlwn1djM5dIxwIsFvWGlOH9nOLZ/9DtdcNY/P3Xl92pJtoTeR6sow3V3dbNu6jfPOO5/Ro4fR2XmYbdu20d3VTfWI4qoc5CJoO9mE8uB0z1g21a1hzZo1nO7p5J5PL+HGt89i37pl/UK/ABbPnZUWUwZQGc4duJ6JUoqlt1zNORNG8e1fLefpp//MyZNt3HbbBxg6NN24q6ysYsLEClRFdgypK/qFZlDLa2xev5eeXo1ScXvt3VfP4ctLb6IiHC6LKh5B0TAxyvopXccaXcnB43Um/8/ceh1f/ekjjB4/nEGDKhk8uIb/deO7mH3Z2WhOMfasYVw5bx6trSfZuXMnsVgs7fWThg1j4cKFnDNlDFo1l9yfUuNn2k52cIk6nHVcAddM7eORZ17lwP4G/u1Ln6B6UHauJjs3kX/+xC184fu/Z+3atRxtaqKmpoauri5279pFzaD480FEipQPPKJRiEQi9PX1cvaYwdz49rih4lYOMT8qkSRjo/LtvrTD4gWXMGFsLV/4/u95ac0ajhw5Qk1NTdo5Q4YM4cYbb2De/Aupqk6vIOCWfgHccnEfN9/xOYYPHUw4HGLokDMpeAZCFQ+n9EuMsn7KhTfezZDq7CzQpeJ1Jv+JY0fysy/fxT995352vtXGZXPmEFZnQXQshFsIV7Uwb8EUpkxdSkPDgSyjbPjw4Vx08USqqo861qdS4mdG1A5jZHtXapaZJKw0l50zjFc7xrNuZwM/+MXv+PrnlxZ1E7n4/LO5/+sfYd3ye3ngta10ROKxahPHDOGnX/s4E8ae2c0Z6W5n5wu/BaW48J0f871ElhQpH7goQ/kmt7wWflUiKSRwPR+XnH82D/7bp/nH7/w3u3akr4Il4+v373+TtrY7uO76BQwe2gyc0cVi9SvWN4phw2oZ2WWuX/POG87kRAyskWINYbtGTufxQ2x/5l4uufGz1IyeWPD7cgqn9EuMMiEnmQPD6RmsnYE3fOhgvvXPH+YjX76Xzs4OOrs60VSlQs1CFSeZMLmCCRPPJysALXQaVO56c14SXz7oyVg+qOQ9730/M6ZeyaLly3n44WYOH4/HoBR7E4kc3MjZQ/v45m0XUnXuNSiluGDqeAYPSjfUG+tX03msMfXYL2OoHJY3hPLEqDFueOD88PBOGDuS333/bnbvP5oWO6u15qEnn+Pl7W/xm988wMmTbbz73TcwfNQJVFbsmX0ip8ay/pUd9PXV8NfYJWRGqX7g6nnMmT7V9LXFaphdI2fP2oeJ9p5m99qHmPP+L9p+T07ipH6JUSbkJHNgGAu8Nu1az7gZ80u6kdsdeJUV4YS5ZRUV1gfh40X3wyuslw8m03ESQKEMhmUxNxGjQESP7SAUPcmF191FVYZBFulup2nPhtTvzXs3+mYMDYTlDcEfjBrjtH5lXt/L7+zgQVXMnnFO1vHZMz7JvY+u4aGnX+HRRx8lHA7x7ndfx9DaExg9ZvYI0d0xltWr/sYjj/yRgwcPctZZQ5k2bQzV1RUMqqhkycIrcnoBS9Ww5r0b6Tp+MK5hGdpkzOTf09ZE1/HDvnjLnNQvT40ypdSNwH8CYeB+rfX3Mp4fBPwOuAI4DnxQa73fyz5aka+AdX/Eyvq3OyvIN4McqN6RQpYPilnGSROIaJTOY42mItFYvxoMYqmjUV+MIT/ifAYaA1G/wFxjAEf0y+r6fn9nQ6EQ//tD13G0pZU1m3bT1NREV1cfw2qr0OSuf2sknqqklqdW/Jlly/7EiRPHmX/pecy5KG4Izp05lZnnTcp7HTc1LDOTvx/eMqf1yzOjTCkVBn4OLAYOAhuVUiu01sZF8aVAq9b6fKXUh4DvA4VlnXMBO4Va+yNW1r/dWUG+GWSu6wQ56NvqBheEG1+mQCQ9i8170r1gmV6y5Ll+eMv8ivMZKAxU/QJzjUn8knasGP2yun5QNKyy0vz2bku/Rgxn/vwbeOONQzzxxBN0d3UwY8p4brzq4nhFeOJhJW5gV8My612CP94yp/Urd6paZ5kH7NVaN2itI8AjwJKMc5YADyYePw5cr4zRnz6RKwtyf8XK+u86ftj0eKS73fT1yRlk1/HDbPvLL1LnWV0/+bxREINE8gaXzPWTvME9tXy16fH6LTs87Z9VUV8d60v7LDO9ZKnzEt4yJ4h0t6f9za1wa6edEGcg6heYa0zz3o007dlQsn5ZXT8oGvbHZzbw3CtbGTykhpG1Ixk8uAJNr339amvn+eeXc/Tom5w6dQqNYteBJv71lyv413uX86/3LuejX7mXFzc4r292Ncys3iXEvWVO4Jd+ebl8OQloNPx+ELjS6hytdZ9Sqg0YDRwznqSU+hTwKYBzzhnvVn9TlFqotRyxsv7NatCZzQoyZ5C71z5ET1uzqbct8zrJeomZSw5BwOoGV7dxa1ay2uSNz0tvhFUCRCDNpW4tGNoxY8hurE1Q8gN5iZcaNhD1Cyw0LBolMy61GP2yvH4ODfPKW/bDB5/liVUbGT6ilve85z3c/K4bGDayHYgWpF99fb3U1HSzZMkSWlvTc55FIhE2b97MPT97grvvWMyHbs68lRePXQ071WkeQ2x1vFD80q+yDPTXWt8H3AdwxdyZpeYDzUspBazLFSvr/1Tn8byzArMZZNLNnBxUuWYXZksCQy+43vH3CIXH2ljdyKyqB3h94zMGMjcbPAJAqsBupPMEMxcvdfUmEcRYmyDhpYYNRP0Cq5u7mQemcP3KtxPdy40rxuLdI2oGs/mN/ahQiNmzZ3P11e9g9Ng+lIoAhetXV1c7N998PR0d6a87fTpCT08Pr9XV8fgzf+WDN83DqUUtuxp2xa1fdU1T/NQvL42yQ8Bkw+9nJ46ZnXNQKVUBjCAe8O8rpWZBLkdKsf6t3M9wRqCsrh/pbqfuie9mLQlUn5M5EwuhqM6+QAHUb9mWFWvz7J/+TO9LDzL7E9+gctgIIJL2GqsbnFLKVNjcvvFZxa1Y3TBaG3cQjfS4Hq9Vyk3J71ic/sZA1C8oXsPs6JdxJ3omVhrmxo19894DadUCTnW18ZELW3k8PIxXXnmF3t5ePvKRD3PJrDGEKnoYUTuctpPZy3FW+jVo0BB++ctf0dbWlnY8Go1y+PAhRg2r4Ptf+mjRBlmuse6nhpVqVJeiYV4aZRuB6Uqpc4kbXx8C7sw4ZwXwMWAd8AHgBZ2rgKFHlJoF2Q65PDZBCCAvhFzu53wCZbUk0L73ZZSC5uYWtm9/g3HjxhMOlzYze2blmixX/rnRt4gdb2HH8kepveozTJhYS2V1C8kZttUNbvacmdRvfsPzG5+Vi93shpG8Wbg9+yt1N5IkkXUWL/QL+o+GlaJf4O3GlVWbtqaVb5quDjOSDt41o4bft1Wwfv16OjraufPOO5k0aQKXzrqWV15ZSTR6RqfC4QrOP/9S9u59Pe24UiHefPMEu3a9xaCKbK09f8JQfvjVTzCmhIlnrrHul4Y5sZuyFA3zzChLxIjdDTxLPCXGA1rr7UqpbwKbtNYrgN8Av1dK7QVOEDfcctJ09Bjf+NqPXB/opWRxz0eu3VFAwTunhlbF2PP8r7nonR/1xdNgHExWLmirL6vV7Oj0ycN86vbF/NfDz/HEE8vYtm07oVBpRtnw4Z0YJ3iDiDBZHUMBXbue49G6Ht729utZdMNChgxrQalYzhvcOVMm5bzxOH1jKtTF7tWSSik3pYG47Hn4UBM/+vdfl61+QfEaNv28yVnX8ttTWop+gbcbV4yFzlP6pWB8rIkHvvEFvvAfj7Jt2zbuvfdexo4dC0BlZZjq6ihKabRWdHeH2bBhd9bxt946yf43j3LTwkv5pw8vYkfjYdbU76C9u4fhQwZzw9xZJRlkxYx1LzSsVKO6VA3zNKZMa70SWJlx7OuGx6eA2wq5ZjQxS3B7m7ebM718u6OsnrNq/9pzTtHVcsCxL2wpIlmoQOVacrgMmHjWCH7460e5KvxX/mfbEDoixW8gvnL+uVRXV6Z+n26s6RaLMfrkOp5YdpIRtbW845rLCFfGV9utbnC5bnxupCUoRKC8zAVWyk1poCaR9SJNRRA1bPp5N2Rdy2lPqZf6BfmXTZ00OmtrhqQMM6N+KQVt+17h19/+DF/7j/9m/Y5GDh1stLqMKWEF/3D7O/m791zFln1v8ZcN9SmvXHt3D8v/VodSqujyUYWOda80rFSjulQNK8tAfyvc2u3mdp6fYnZHWT03JNzL5ePicVBOfWFLEclSYtPMxOvqK2YwumM6bfvr+NYHplJ93juLvv7+5hY27GkgGoulZpnJmm4VYbhyYoQtu9rp7uoi2hciXJnngjnIddMq5jtUqEB5uaRi/JsXcgMa6Elk3dytWy4a5oanNEj6VWp/Mlk8dxZPvryJULQnTb9Chs/vB1/5JHveaiISKazMUu3wGiaPHwVkL5MC9EajrNq0tSijrJix7pWGFatfyfNL1bB+ZZSBO7vdnL6hZpJvd1QhO6euHN2SWpJz4gvr1nKSnS+7mXhFuttpb6wHIHpsJzOueV/R/Zl1wWTOHj+aVZuS+iHrAAAgAElEQVS2cnb3fjJjVZWCq846UdS1M3E6LUGhAuV1LrDk37dq6CjbNyBJIuvebt1AatiI4UQivXR0dNDX2weDnPeUBlG/nOxP0iDat+4JyLC5jJ/fjKkTim4D0pdJ7RzPRzFj3UsNK0a/wBkN63dGmRu73dzO85Nvd5TdnVMq0sbFI1qpSKzoOeFpcGs5Kd9s0Uq8nO7PnPOnMOf8KWxZvofu1vTBVBGCyTU9RV/bSClpCcxuAE4uC7tB8u9L037s3oAkiax7u3WDqGHz59/A448tY8WKp+jq6uTORW+jee8aRz2l/V2/IK5harvK0i8di9J+tKGkaycxLpNmHrdDpoa5sSzsJMXoFzijYf3KKHNrt5vbeX7s7I6yEwtS1biKzDw8pZT+cGs5yShYTbvWM/6C+dSMnpjWNzPxSiZkdGN5K3PA71u3jCM719HYNZjsUOTCKSUtgdkNIMgJV41/3yR2bkBBfk9e4OZu3UBp2IjhzJu3mC1b3mTFiuWc7unm7juu58pRh2neU176ZSyWDQRKv5p2rWf4+GklXTdJcpnUuIRZGQ6zeO4sW6/P1LAgj/Vi9Quc0bCyN8rC4TCAq7svp19wLps2vJ52zGkBzRUkbnfnVEX7fsIZce+ZVnoh8QxuLSelX1enisgm+3agbiXH9tdniVe0L+LJ8lZyUIYUzBrVTltv6d6EfDet+i07WP3sA7S3tzLr0rH0tnWl9cXL3YilBiKbfm8GWHxYobipX/VbdhA5Hck67puGxcbyxrZmdu16llPd3XzuzkV86Ob5bFn+47xehqDpl7FYNhAo/XJSM5LLpMkktbU1Q1g8d1ZaPJkxia3xea81rNz1q+yNsnHjx/D/vvMF165fv2UH9ZvfyDo+e87MwOXZ6ZrzRR544DccbzrIU7/4EkOqq9KeL3RwuLGclF1sNl5E9uSh3Wf61rA563WxWJRjDZtBxxztjxnGQanQ1Bx+Hri5sIvowaDTk9vOvvQqZl96FZGO4+x7/Lucf+6FEBtJfX09Ty1fRW9vLwCDBlVQNWYYm/ceYHhLHToWf8+xWNST+KpSApHN/r5JBlp8mF0mThrHF770965cOzPAP8ngIdXc9K53+qxhGqU0F0yNl5mys2sxePoV14mm3RsSCVSDpV9OjrlkqEfK6Jl4deq5zCS2J7u6efLlTQDUHFmHTuQ/80IDyl2/yt4ocxuzAFmAPbvf9KE3pVHoYHXDxWyVLXvXS384czxDuIzHxs1Y4OqgyByUFSEIH6+jt/M4VYPtXSPWN4qGvR20th40P2HbH6BxO1seuw8u/gjPP/9MyiBLokIh1mzcxPy+TWc+Dx1zfbZW6qw2Zzb0ARYfFgSs9KuqqjJwk8p8BFm/0FG0Tu6wCo5+ueHhMTN6rHZnrtm4iStPv5Y65rbHqT/olxhlCaxy+PSXYr5BSTdglS07GkkPqA+FK7n81q9QNWT4mbIl0T5a9m5i3AVXsn/DcleSS5oPSk3T337LtPfdnvf1kVNjeXXdTh577HFOnMjeuTkk3MvHz91JRUjTu/8lfvtiE+On1JiWKRnXsw8dThd4t71lpc6yrf6+Q0ZOzFmWRsoqlYbolzfkyvZvjEHKpV+TZy8CcOU77/YOZiujx2oX5sSe3RDybkd1f9AvMcrIncOnvxTzDUq6gcuWfN40S3Ymxr5lDrTdL/2BU+3HOFC3kunvyFv0oSDMBqXSUboObgPyGGV6GIcPdvLSSy9Rv2ULVRWQaWq9Z3o3SfFWaGZV7+Pg6ZlpSWyTjAl3Z8+6dSznbK0UgXDixleMd8LpZKEDDdEvb/ULzDP9W/UtU78O1K3k5OHd9PZ0Ot5/t3cwWxk9Vrszx4Wy62zm60+xGtZf9EuMMnLn8OkvxXyDlG4g92wzTrJvZgPtVPsxAFoaNjPlipttDzg7g904KN/1D9+jomYUn/j4J1h0wzuAfBmxQ0Sj0URQdYx//MhN3PSOS1PP9va0s/2pH6ITb70iBAvOiRGafSkrX9udtbNpzNtu53Td71PxGBCfgc9cvNSyB6UIhNmNLxbtdcX4TTIQyyo5jehXMPpjJJd+tex7jeTErJDvfKH65TS5jB6z3Zk14ShVIQ2GeaXRg2hFsRrWX/RLjDJy5/Dxqpiv2wRpC3Ihfdm3bpnlGj+JWafdAee1R2ZwdRWDB1WlxHTQ0FHZ70VrxnTt4X0L52btXBreUkdzAd6BUgXC6kbTenCH7WsUiltByQMJ0S/vsdsfc/0qPNUC+OtRjnS3U//UTyy9lXMMsWVJDbum9jixZtKSNNktB1eMhvUX/RKjjPwufreL+QrW5JuR2vWW+emRSYppxckmy9n+nAW3ZJUr2bJ9WUHegVIFIrO8SDIOJtbXS6S73fHPKyhxQuWO6FdwseNRs/Od99uj3Fi/mt6e7O+YUY+SuzOTbFn+Y7oL9G6WomH9Rb/EKKO0xJ6CuxgH2p6/PkLLvrq053UsxoqH7mNndEbO61wQ3sOEcJSwgt6+Pp586H52R6fnfE1PJMqwGrIDwwrAKKaxvl7m3n6PK/ENTguEFx6soMQJlTvlrF9aKxoOtnD5zKl+d8UVMg2FTY9+m3Tfkb3vvJ8e5TMaZm/5MUmh3k0nNayc9UuMMuxloxb854SJG1opGBFr5rl11rPRYVUx3nHlCcIJ4yqsNGepI/y27jSdkZDl60KhKs49dzyvv76KV19dVtT3wisxtSMQdgNovfJgBS1OqFwpN/3SqoPRo2s5//zp7Nyxk//8w7MoFeLWxVf43TVXiSeYzQ7FsBP4Xux4tEroWnC/y0jDyl2/xChLIC7+wvA6jUEsFqOlvZdRg7KfU0NGc8edd1i+dtLJFwl3n8QYcVoRDvHpRZM4VHut5etO9bRyonUfPT2ngPRdbXa+K6qvm+Y3vVmesyMQdmNSvPJgBS1OqJwpJ/1S6hRnTdR88INLqKys5Omnn+aHD66ko7uHu5Z4493zIw2L1c06V7oFKH485kroatcw8zLEwCkNK3f9EqNMKAq3gk6txLKltYPH9o/hkxccJaw0Ua14ITabaMUQbrrhNj58sfUse+9v/8Kp7vTUEiFinF3Tw7Ufvs3ydb/4xbeJRs13tdm5AQ46sdUTcYh0t1NRNSjn0qhVTIoTxc4FoVCUOs3YCVFu/+CtVFZW8Pjjj/Pc2g2eGWVe6xeAvvgWXnz5Fa7Wm1MatlbN4eaLr8p5zWLHo1VC11Wbtto2yrwwcJKf2czFSwuOq4PsfG/lrl9ilAkF42bQaS6xXDQtvX7fdHWIbX1TWfvXp5g7f7zlNWd97ls5Wjxk+Ux7e6vp8XyJNwdxis9c0UW4xzqw30kKnT0mcyWd7jzBoKGjyqrYudB/UKqP6upB1NbWUllZCWTX53QDv/Rr1aatnBtrTItRPTfWmNdIKnY8WiV0tTpuJGko9Z7qdl3DivHg58r3Vu76JUaZUDCFxBgUskyQSyz7TnUwteIEYRVvN6w0kznGHj3JtezkxSbenB7axZTaKL2Dx3H1h/+vK31LYucGY7oE0fBa/G/YvD/nawWhv+GHfgH0dJ1kcuhYtoZ1tTnzxjKwSuhaWzMk72uThtK4GfOZ8/4vutE9wL6B7GS+t6BjHeUsCCZYxRhEurMzN0P6LCjXNbf95RccqFuZJZZJWne/jFmQ7HR1yLXs5NMvOLeg4wCxnlbOCb1FSEFVxz7Lz8UpzG4wuc5Jkaozmvu1gtCf8Eu/AC6uajZ9vdXxUrlg8oSCjifJNJTc1DA7+pV53hmy8731B8QoEwoiV4xBJsbB3bxnI6//+WemAzwufG/S0vCapVieaj2Y2j2ZJKw0o0Jdrm39tyo6n6sYfc/Wx0iJhdauCoXdG4zdCgpuC7AgGIlFh7P/zUY2b95MV2cXg6vcX7jxS78AJlX3prxkScJKM6nanWXb3Y1HCjqexK6hVCqFGMh28731B/0So0woiEKCKNMGd6yPzpa3sga4MQdOLrGcfM3f87U1tTzYfAWrK6/l6ejb+Ouw6xhzy5eYfdlM4oEazv7kLuacfX5vRyunG15MCa8iVpRQJGfe+V5n9wZz2ZLPM/f2e8iXcK0/zTaFYBPtHcOmVw/xy1/+ipdf/htDB2m+/Ok8tWUdwC/9Alhw25cZvPBuXq6+hqejb+Pl6msYvPBuFtz2ZSfeWhbFxJQV6km0uoaT+gUkSssNDP2SmDKhIOwGUWYO7iTNezemrf2bu6XjGOvH7X3xfubPmwRVimg0yqLFt3HhjMsBaDkCfV3HaVn1bcbecA8VQ0aV8A7PMGxYLR0dJ02PtxwZk3X82Nr/RMeiadJRzE4lu4Gvhdxg9tetxGz5185r7eJHmgGhDNEj2N/QzooVT7NlyxYmjx3Cz7++lLEj3f/O+KVfu1/6A5Ep17L69QbLnGFOj59iYsqc2G1ZrvoFwdAwMcoEV7ASKx2NpgarqfCpEHNv+1ragHj1mQeh8yizqqNs01Pp7T3F6lWP88LzLxIOx+PJLorVMZl9vPLgV9kRyk6PUaV7mK3XU6/mE1GDbT0fjVYQn50Z34eiq6uCX/3qvqxrLIj9leEUv1Mp0t3Ozhd+S9eJI9gJvi/kBnNs32s5z8mXK8kOftbmE8oIXcXp0yfp6GgnrGL8y8ff64lBVghO6te+dctob3qTt5q6ORk9BzDPGZZv/OQzGDKfNysSXhkOs3juLMv3XUo6CT/1C/qPholRJriCdQzAmcFqHoCeXmQ80t1O5OgbhIHJKr7b8jSVgKar6wjr1/+NYVUxrnt7JyoM46P7eGBtEx0ZmfpvmdFD7dm9VDY+w9O7so0yq+fPOmsY06aNpbq6glOn+mhoaKG5eafpe34OUAznqpln8Z0vfoJBVZUFfGJxQeg81kjSTe9UPiCrTOLgjJCB/7X5hPKlsiLsdxeycFK/ksubk3Qzu5iQ0K/0nGF2xk8+gyHz+aSxV0hG/1K0oJz1C4KjYWKUCa6QHCj71i2jec+GNIFLDlYr4Ws1lFNKCV9iTXC6OsQ2PRWAwYMr+T8fv5lhJzZR0dUAxKgMh/g/N42nY9Tc1DVC0R7GHH4aBbz9nBgzrnwnsfBg288bufqiC9NeN+L4K7SNvip1/qgRNbzj8gsIhQoL14x0t9O0Z0PyE4r/G4vSvHcjXccPcuF1dxUtEFaz3MEjxjkmaH7W5hMEp3FcvxIY9QvOxHflGz/5DAar5zOLhBtxcqnOSr9a9m5i/AXzeXPDk0W344V+QXA0TIwywVVyucMvW/J5It3t1D3xXbQhc36srzcVJNq8d6NpbrLTVFJbM4R3LZhO3eOPoxMllBQxhp46wNUL7kgJwL51y2gOgY5BSMElw5rTBlu+563Yt24ZTYeP2T4/F431q8FE4HU0SuexxpIEIlO49q1bRtOu9QwfP62o62XiZSkWQfASJ/Qr+fpM/YJ4fFeku53mPbnHTz6DoRiDwsmlOkv90jF2r32InrbmottxW78gWBomRpngKvlmMnl34JjEdUxXh9gVOo/Fc2fRWL86WzQNotR5/BBNu9aTOXszlhkqZjA66epO28GVhbPJEd1w0XtVa04QvMYt/dqmp6biu/zQMK/0S8ei9LQ1Ac5omFtLjEHSMEmJIfhKrpmo2XNhpRkT7uZ9C+cy5/wptDc1kBlvYAxM3bP24eznDaJZyLZsI07m8jHrgwqFGTxiHCoUdqQNs7acuma515oThGIpRr9Gqi5qa4b4qmFe6Ne4GQsYN2OBoxrmVg61IGmYeMoEX5m5eGmW+z8UrsxbnDbJ8HHTONV+LJ6KIhTmrOnzUjObSHd7apZmxDjYihmMTru6rfpg7LsT7nS3XPTlXmtOEIqlWP26yfDYaw3zSr/ajzZwqvO4Y+24ucQYJA0To0zwFbNZVizay/66lVyQ2MFkRb5B2li/GhUKp4nd5NmLUsGtUNxgzOfqLjSA1qwPuQKMi3WnB8lFnyQIeYEEoVhK0S/wR8O80C+Ia9ipPccs2ymUgaJfsnwp+Er70QbzHUyNO0zOTifXILUSuwN1K7Nq2dnNQJ0k38zUTr28Utuw22/jOW676Av9HMGZz0oQ/KIU/QJnNCyI+mWnnaDpl90+GXFDv8RTJvjK8PHT6GlrZtyM+UyevSi1FKCj8R1MuWYfuQap6Qw2FqWl4TUyg0QL3YWUa2bqVCCqndmvnX4bz3HbRV/o5xiUvECCUCyl6Bc4o2FB1K987UDw9Mtun5K4pV9ilAm+kfmljvZFCtrWnWuQbln+4+wZrI6deZi4/uTZi2wNLLtuaq9y3dgRBC+NnmLaCkpeIEEohlL1C0rXsAN1Kzm2vx7RL2/6ZMStz0qWLwXfyPxSH9v3WkmFcI1ctuTzXHXXv6d+5t5+Dyp8Zg6SvP7+upW2dvPYcVP7VczXqt9u7VQqtj9GnPisBMFP3NQvsKlhDZvRsViqD6JfxVNIe27qlydGmVJqlFJqlVJqT+L/kRbnRZVSWxI/K7zom+APZl/qXNu+S8UqdsOOkGbOoKwGXrHpNTKv4YR4emn0FNOWE5+VIPiF1/oFFvU4dSzlPRP9Kp5C23NTv7zylH0FeF5rPR14PvG7GT1a68sSP+/1qG+CD5gKTAbxgVHnyEC0it3IJaTJWd8Bm960UgNRnRLPzuOHqHvs39DRqOU5TlKMQAUpL5AgFIrX+gW56nEa2swYd5Huduqf+kng9SsWi6bp7pYVP0pLM5Kv76VSqIa5qV9exZQtAa5NPH4QWAN82aO2hQBiJTDGArPJchpOrNWbxW5s/tN/ZOUASubXgeSs701o3n9G1HLkxik1ENVujEI+Qdiz9mG0Nil54pLRU4xABSkvkCAUitf6BdljZsvyH9PdejjtmFG/APbXraS3pyPt+SDqFzqWtvuz71RX1uvcnLQVqmFu6pdXRtk4rfWRxOOjwDiL86qVUpuAPuB7WusnPemd4Dn5vtReBHkad06dt+CWtJpqaaVDTGZQ++tWEuk84Vh+mkISI2bW3AuFK7n81rjzuX7FT9IMzdnv+Tw1oyeW3L9ciIElDDSCoF+XLfl8SrPGzZgPkFYTMtLdzrF9r2W9Lgj6lXyNUcNmLl6aKCm1LnWuCldwxa1fdX1XdpA0zLHlS6XUaqXUNpOfJcbztNaazDWjM0zRWs8F7gR+opQ6z6KtTymlNimlNh1raXXqLQg+khkg6naQZ6Zodh0/nPa7cQNAJjoWpbVxh6P5aQp1n5t9Po31q+k6cSjtvN1rH3Kkf4KziIb1L7zWr2SbSc1q3ruRpj0bMC4dxtvM1rAg6Ffma5LnxktKGa4Rjbq2ZBlUHDPKtNaLtNaXmPwsB5qUUhMAEv83W1zjUOL/BuJLnHMszrtPaz1Xaz13zFjTPQNCmWEMEPUiyDNTEHavfchyJxXES6fMvf2e1C6oWDRCvtiJQrAsVdLUkHWu2efTvHcjTbtfzTq3p62JruOHs44L/iIa1r/wWr+SbaY0KxqFZHuJVBmZRcKTGualfrXsM4+pM9WwPRtNSkrFDc6BtCvbq+XLFcDHgO8l/l+eeUJiR2a31vq0UmoM8HbgBx71T/CRTK9VzJDvJ4kxr1ipZS3MBCGzzmQmxhgJN/LTZLrPU0up46ZlnWs6K41mb1pIsnvtQ8x5/xdL6p8gCOYUol/FlDHK1eYZrTrTXjJVRiZpniuX9Qtyx9SZalgsPbA/dTzhLRsoOQy92n35PWCxUmoPsCjxO0qpuUqp+xPnXARsUkrVAy8Sjyl7w6P+CT7SWL86lWsnFovSenBHzizXpbrd7eycyiTZvhez4Hy7mMyDjK3fz6nO4471TRCEdArRr+T5rmuYIVWGsQ/tRxs88eIVp2FW6AG1K9sTT5nW+jhwvcnxTcAnE49fAWZ50R871G/ZwfOrXqbtZAcjaodx/eKFzL7sIr+71e9IDd6kgOgYsb5e5t5+j2mG57onvoud4NlIdzs7X/gtKMWF7/xY2nl2BcG4kyrJvnXLcs6CSyE5g64aOirnTNasT5nFy5PFiwfK7FLIRjTMfQrRr7TzbWhY5/FDbH/mXi658bNZm3XsaJiVfjlZJNyI0QOYbzVBNMwaKbNkQv2WHTz15Cp6e+Pu1LaTHTz15CoAETWHMc4ykyRz1uRyeecTksb61XQea0w9ziUIheBmfprkDJqm/SQ9X7l2MXnRJ6E8EQ3zhkL0K3m+XQ3bs/Zhor2nTcMPitUwL/QrWfrJzk5ML/pVbohRZsLzq15OiVmS3t4+nl/1sgiaw3Q0H8hysxtz1iQpZMt1pLs9sRMpTvPejY5sSY90t1NRNchyFpzvtbniSIwz6EzyiXeQtnMLwUA0zBvs6hcUpmGdxw+l4lyTm3WcSG0zc/HSouLZCtGvXPFsomH5kdqXJrSd7CjouFA8MxcvTavnBqRy1hgpZMt1Y/3q1E4ksL+tOl/dtlJiQfK9NleMyECdMQrFIxrmDXb1CwrTsMzUEHZT27ilYQXpl0U8m2iYPcRTZsKI2mGm4jWidpgPvenf5BIq46zKrns700uWONOWt8woPJkzulKSQeZ7bfZOKlIJYd1Omij0T0TDvMGufoF9DTN6yZLY9Za5oWGiX94iRpkJ1y9emBaPAVBZWcH1ixf62Kv+iV2hsuvezvSSpa6ZZ1t1PuEpJg2G3cD9QoRdEOwgGuYNhcRC2dWwTC9ZknypbZzWMNEvfxCjzIRkzIXsXLJPsbl3nI4lsHaR595WnUuwCokFybymncB9CXIVnEY0rHCK0TA3YqGsUtjkS23jtIaJfvmDGGUWzL7sIhGwAsjlNveSYkQyn2AVMxMsJHBfglwFNxANK4ygaNiCj3634NdEuttp3uOchol++YcE+gslky9RYNDJF4BbzEzwQN1KdNQiQ3WRs8h8QbyCIBRHf9CwLI0qQcPc2Hgk+mUP8ZQJJeNG2SEvySdYhc4EI93ttDS8lnbMicDXoMzkBaG/Ue4aFq+Rm1m2qDgNcytwX/TLHuIpE0rCq+K7bnLZks8zbsYCVCgMxDNJj5uxoGi3fMOrT9re+m6Xcp/JC0JQ6Q8aNnzctCz9uuqufy9Kw8wS4op+eYcYZUJJFJJ7J6g4LcqtjTuyjpUa+Go2k/cKWXYQ+jPlrmFO65dZQtxy1i8oLw0To0woif6w88ZJUe48fggdS48lC4UrmXv7PVmzVrtC4fdM3okCyoIQVMpdw5w2Ks9feHvWMbOEuOWiX1BeGiYxZUJJlLrzpthUGk7ipCib5RiyilGxG2PhZx6gUpLmCkI5UIqG9Tf9AvsaVg76BeWnYWKUCb4ShOBPp7Z0R7rbszJxg3XlAbtC4edMvtwDoAXBTfqTfoF9DSsX/YLy0zAxygTfCMoMxqnZbmP9alQonCZAKhTmrOnzTL1kdoXCrzxAxSbNFYSBQFD0K9kXLzWsHPQLylPDJKZM8A07wZ9eBGgeqFtJe9ObHKhbWVK7hdTn9DvGwg7lHgAtCG4SFP2KdLdT/9RP0mKmim3XjoaVi35BeWqYeMoEX7A7g3F7ecCYU6ylYTNTrrg5lQG70HYLqs/pUoyFkzEufi87CEJQCYp+AeyvW0lvT7z4fLIPxbZrR8PKRb+gPDVMjDLBF+wMbC+WBw7UrTzTDx3jQN1Kplxxs6vt5hIKoygBBQuUkzcBK4FOzsL9DG4WBD8Jin5Futs5tu9Momqd0LBj++tda9euflUNGV6wkeW0EWumYck+RbrbA6lfYpQJvmBnBuN2gKZZ5v2Whs3xvNgutptrNrpv3bK0ZYhCBMqrGJcgBDcLgp8EQb8g7iUzZvLXsSgtDZvP/O6jfp234JaCtEL0K44YZYIv5HOTexGgmeYlS6JjiZmndq1dK4yi1Lx3I1prChEoL24CQQpuFgS/CIJ+ZXrJUhgSv/qlXy17NzH+gvkFaYXoVxwJ9BcCiRcBmq0HszPvJ1pytV0r0kQpGoWkoNto36vgW78zcwtCOeCFfsWvZV403M12c/bHoA271z5kWytEv84gRpkQSNwI0MzckVQ1pNbW6+y0W+ouq+wiwBlLEnkEyoubQDntuhIEP/FCv+xey267pWiYmTb0tDXZ1grRrzPI8qUQSNzIbZMZS2BW9qjYnT+lximYiZKRfO58L3YZ+Z2ZWxDKBT/0C/zTsHz6Bbm1QvTrDGKUCQMCO7EExYqSE3EKZqJkJJ9AeZGgsRy3lwtCf8CuxvilYfn0C3JrhejXGcQoEwYE+YJISxElJwJU/cx6bZdy6KMg9EfsaIyfGlYO2lAOfQSJKRMGAHZiCYoNAC2XOAVBEMoTuxojGtY/EKNM6PfkCyItRZScDlD1oiyLIAjlgx2NEQ3rP4hRJvR72psacsYSlCJKTscpGGNCBEEQ8ukXiIb1JySmTOj3DB83jZ6TzYybMd/xnT9OximUQ2JDQRC8JZ9+gWhYf0KMMqFfY0ckghIA6kVGa0EQyge7Ro5oWP9Bli+FsqOQmIVyyOAMEmwrCAMJuxpWLvoFomFOIUaZUHbYjVkwE4nmPRsDKRJeZLQWBCEY2NEwM/1q2rWeruOHvepmQYiGOYMYZUJZkenOL7j0UKwvkCJRLokNBUEoDbsaZp4lX8drSgYQ0TBnkJgyoawoJGbBKst0+9EGV/tYDEGJCREEwV3sapiVfvW0NRHpbg9cAL1omDOIUSaUDVYxC/mCX/etW0bzng3oWBQVCjN8/DRP+y0IggCFaZjRyMnUMAmg7794snyplLpNKbVdKRVTSs3Ncd6NSqldSqm9SqmveNE3oXwoJmZBgk8FQQgKomFCPryKKdsG3AKstTpBKRUGfg7cBMwE7lBKzfSme0I5UEzMggSfCoIQFETDhHx4snyptd4BoJTKddo8YK/WuiFx7iPAEuAN1zsolAXFxCxI8KkgCEFBNEzIh9JZuztcbEypNcAXtdabTJ77AHCj1vqTid8/Clyptb7b5NxPAZ9K/K85kVIAAAcxSURBVHoJcU+cn4wBjg3g9oPQB9/arwhROWlk1bRDrZGTfTGa/OhDAr//Bl72YYrWeqwH7bhCwDRsIH1vgtq+r32oCFE5cWTVjMOtkZ19Mfr86EMCv/8OvuuXY54ypdRqYLzJU1/TWi93qh0ArfV9wH2JdjdprS3j1LzA7z743X4Q+uB3+0Hog9/tB6UP5UCQNMzv9oPQB7/bD0If/G4/CH3wu31w0CjTWi8q8RKHgMmG389OHBMEQRAEQej3BCl57EZgulLqXKVUFfAhYIXPfRIEQRAEQfAEr1JivF8pdRBYAPxZKfVs4vhEpdRKAK11H3A38CywA3hUa73dxuXvc6nbheB3H/xuH/zvg9/tg/998Lt9CEYfyg2/PzO/2wf/++B3++B/H/xuH/zvg9/texvoLwiCIAiCIJgTpOVLQRAEQRCEAYsYZYIgCIIgCAGg7IyyAko27VdKbVVKbVFKZeVF86gPrpSNUkqNUkqtUkrtSfw/0uK8aOL9b1FKObJpIt97UkoNUkr9MfH8q0qpqU60W0D7dymlWgzv+5MOt/+AUqpZKWWaV0rF+Wmif68rpS53sn2bfbhWKdVm+Ay+7nD7k5VSLyql3kiMg38yOcf1z6Fc8VvD/NavxLV90TC/9ctmH/q1hol+5UFrXVY/wEXADGANMDfHefuBMX71AQgD+4BpQBVQD8x0qP0fAF9JPP4K8H2L8zodft953xPwWeCXiccfAv7ocft3Af/l4vfvauByYJvF8zcDfwEUMB941Yc+XAs87eJnMAG4PPF4GLDb5O/g+udQrj9+a5jf+pW4vuca5rd+FdCHfq1hol+5f8rOU6a13qG13lUGfUiVjdJaR4Bk2SgnWAI8mHj8IPA+h66bDzvvydi3x4HrlcpdX8vh9l1Fa70WOJHjlCXA73Sc9UCtUmqCx31wFa31Ea31a4nHHcR3S0/KOM31z6Fc8VvDAqBf4I+G+a1fdvvgKn5rmOhXbsrOKCsADTynlKpT8ZImXjMJaDT8fpDsP3yxjNNaH0k8PgqMszivWim1SSm1XinlhOjZeU+pc3Q8zUkbMNqBtu22D3BrwuX8uFJqssnzbuLm370QFiil6pVSf1FKXexWI4nlnTnAqxlPBeVzKGf81DC3/35+aJjf+mW3DyAaNmD1y5OC5IWinCnZtFBrfUgpdRawSim1M2Ghe9mHosnVvvEXrbVWSlnlNZmS+AymAS8opbZqrfc53deA8RTwsNb6tFLqH4jPeq/zuU9e8xrxv32nUupm4ElgutONKKWGAk8A/6y1bnf6+uWM3xrmt37l64PxF9GwLAa6hg1o/QqkUaZLL9mE1vpQ4v9mpdSfiLuNbRtlDvShpLJRudpXSjUppSZorY8kXKrNFtdIfgYNKl4Mfg7xeIZisfOekuccVEpVACOA4yW0WVD7WmtjW/cTj13xEt/LhRkFRmu9Uin1C6XUGK21Y4V2lVKVxAXtf7TWy0xO8f1z8BO/Ncxv/crXB580zG/9stWHga5hA12/+uXypVKqRik1LPkYuAEw3enhIm6WjVoBfCzx+GNA1sxXKTVSKTUo8XgM8HbgjRLbtfOejH37APCC1tqpDMV5289Y938v8XgBL1kB/F1i9858oM2wTOMJSqnxyTgYpdQ84uPcsRtL4tq/AXZorX9kcZrvn0M5EwANc7vsnR8a5rd+2erDQNewAa9fhe4M8PsHeD/x9d3TQBPwbOL4RGBl4vE04rta6oHtxF32nvZBn9nBsZv4zM6xPhCPcXge2AOsBkYljs8F7k88vgrYmvgMtgJLHWo76z0B3wTem3hcDTwG7AU2ANMc/uzztf/dxN+8HngRuNDh9h8GjgC9ie/AUuDTwKcTzyvg54n+bSXH7joX+3C34TNYD1zlcPsLicc7vQ5sSfzc7PXnUK4/fmuY3/qVuLYvGua3ftnsQ7/WMNGv3D9SZkkQBEEQBCEA9MvlS0EQBEEQhHJDjDJBEARBEIQAIEaZIAiCIAhCABCjTBAEQRAEIQCIUSYIgiAIghAAxCgTBEEQBEEIAGKUCWWDUupFpdTixONvK6V+5nefBEEQ7CIaJuQjkGWWBMGCbwDfTNQCnEM827UgCEK5IBom5ESSxwplhVLqJWAocK3WuiNRqPhrwAit9Qf87Z0gCEJuRMOEXMjypVA2KKVmAROAiNa6A+KFirXWS/3tmSAIQn5Ew4R8iFEmlAWJIr3/AywBOpVSN/rcJUEQBNuIhgl2EKNMCDxKqSHAMuBftNY7gG8Rj80QBEEIPKJhgl0kpkwoa5RSo4HvAIuB+7XW3/W5S4IgCLYRDROMiFEmCIIgCIIQAGT5UhAEQRAEIQCIUSYIgiAIghAAxCgTBEEQBEEIAGKUCYIgCIIgBAAxygRBEARBEAKAGGWCIAiCIAgBQIwyQRAEQRCEACBGmSAIgiAIQgAQo0wQBEEQBCEA/H8ZEEhP9d6AVAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Out-of-Bag evaluation"
      ],
      "metadata": {
        "id": "9HiUsZa_VGBw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "It can be shown mathematically that only about 63% of the training instances are sampled on average for each predictor.⁠6 The remaining 37% of the training instances that are not sampled are called out-of-bag (oob) instances. Note that they are not the same 37% for all predictors. The following calculate this number when `m=1000`:"
      ],
      "metadata": {
        "id": "LXRxTOnAa2Ev"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(1 - (1 - 1 / 1000) ** 1000)\n",
        "print(1 - np.exp(-1))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "t_pOXuBaVRwI",
        "outputId": "1104123d-7f8b-4160-f868-f6b17a2ee36c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.6323045752290363\n",
            "0.6321205588285577\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "In `Scikit-Learn`, you can set `oob_score=True` when creating a `BaggingClassifier` to request an automatic oob evaluation after training. The following code demonstrates this. The resulting evaluation score is available in the `oob_score_` attribute:"
      ],
      "metadata": {
        "id": "aMg1xzaMVHfr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "bag_clf = BaggingClassifier(DecisionTreeClassifier(), n_estimators=500,\n",
        "                            oob_score=True, n_jobs=-1, random_state=42)\n",
        "bag_clf.fit(X_train, y_train)\n",
        "bag_clf.oob_score_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MpOmKptvT65-",
        "outputId": "418d2dc7-8f3e-4379-cf07-be3a423772cb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.896"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "According to this oob evaluation, this `BaggingClassifier` is likely to achieve about 89.6% accuracy on the test set. Let’s verify this:"
      ],
      "metadata": {
        "id": "qjCSrZaaakSn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = bag_clf.predict(X_test)\n",
        "accuracy_score(y_test, y_pred)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C5GIYPd1an13",
        "outputId": "65ebf350-b3f7-4885-cd6b-3ef10aa70a76"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.92"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We get 92% accuracy on the test. The oob evaluation was a bit too pessimistic, a bit over 2% too low."
      ],
      "metadata": {
        "id": "OmiahqHWaqqo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The `BaggingClassifier` class supports **sampling the features** as well. Sampling is controlled by two hyperparameters: `max_features` and `bootstrap_features`. They work the same way as `max_samples` and `bootstrap`, but for feature sampling instead of instance sampling. Thus, each predictor will be trained on a random subset of the input features.\n",
        "\n",
        "This technique is particularly useful when you are dealing with high-dimensional inputs (such as images). **Sampling both training instances and features is called the Random Patches method**.⁠ Keeping all training instances (by setting `bootstrap=False` and `max_samples=1.0`) but sampling features (by setting `bootstrap_features` to True and/or `max_features` to a value smaller than 1.0) is called the **Random Subspaces method**.⁠"
      ],
      "metadata": {
        "id": "rva-jGH6bOmw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Random forest"
      ],
      "metadata": {
        "id": "9oWyFDYpdIwf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Random Forest9 is an ensemble of Decision Trees, generally trained via the bagging method (or sometimes pasting), **typically with `max_samples` set to the size of the training set.** Instead of building a `BaggingClassifier` and passing it a `DecisionTreeClassifier`, you can use the `RandomForestClassifier` class, which is more convenient and **optimized for Decision Trees** (similarly, there is a `RandomForestRegressor` class for regression tasks). The follwong `BaggingClassifier` is equivalent to random forest\n",
        "\n",
        "```python\n",
        "bag_clf = BaggingClassifier(\n",
        "    DecisionTreeClassifier(max_features=\"sqrt\", max_leaf_nodes=16),\n",
        "    n_estimators=500, n_jobs=-1, random_state=42)\n",
        "```\n",
        "\n",
        "The following code trains a Random Forest classifier with 500 trees, each limited to maximum 16 nodes, and using all available CPU cores:"
      ],
      "metadata": {
        "id": "j_TkqrLMdMT4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rnd_clf = RandomForestClassifier(n_estimators=500, max_leaf_nodes=16, n_jobs=-1, random_state=42)\n",
        "rnd_clf.fit(X_train, y_train)\n",
        "y_pred_rf = rnd_clf.predict(X_test)"
      ],
      "metadata": {
        "id": "rBo1gB1MVWwI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = rnd_clf.predict(X_test)\n",
        "accuracy_score(y_test, y_pred)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X5-UVNb5jk-H",
        "outputId": "63b8fb73-d95f-47f4-dcb5-fe0150c9e8d9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.912"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can also create an Extra-Trees classifier using `Scikit-Learn`’s `ExtraTreesClassifier` class. Its API is identical to the `RandomForestClassifier` class, except bootstrap defaults to False. Similarly, the `ExtraTreesRegressor` class has the same API as the RandomForestRegressor class, except bootstrap defaults to False."
      ],
      "metadata": {
        "id": "9x22GreAfOSv"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## AdaBoost"
      ],
      "metadata": {
        "id": "Ro3Mdc0UjuAe"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Scikit-Learn uses a multiclass version of AdaBoost called `SAMME`⁠ (which stands for Stagewise Additive Modeling using a Multiclass Exponential loss function). When there are just two classes, SAMME is equivalent to AdaBoost. If the predictors can estimate class probabilities (i.e., if they have a `predict_proba()` method), Scikit-Learn can use a variant of SAMME called SAMME.R (the R stands for “Real”), which relies on class probabilities rather than predictions and generally performs better."
      ],
      "metadata": {
        "id": "bS8rugGHkMov"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The following code trains an AdaBoost classifier based on 30 Decision Stumps using `Scikit-Learn`’s `AdaBoostClassifier` class (as you might expect, there is also an `AdaBoostRegressor` class). A Decision Stump is a Decision Tree with `max_depth=1`—in other words, **a tree composed of a single decision node** plus two leaf nodes. This is the default base estimator for the `AdaBoostClassifier` class:"
      ],
      "metadata": {
        "id": "Pmgz2OtNkX-n"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "ada_clf = AdaBoostClassifier(\n",
        "    DecisionTreeClassifier(max_depth=1), n_estimators=30,\n",
        "    learning_rate=0.5, random_state=42)\n",
        "ada_clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aDhLJhmhdfw2",
        "outputId": "7b365b79-8e1d-4ac6-ab4a-8dcd7433cf96"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=1),\n",
              "                   learning_rate=0.5, n_estimators=30, random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 115
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_decision_boundary(ada_clf, X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "Gsh_380VkHX_",
        "outputId": "20d453fc-efc5-4ea5-c391-af74e94502fd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Gradient Boosting"
      ],
      "metadata": {
        "id": "91HFYHBB-Bmv"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "First, let’s generate a noisy quadratic dataset and fit a `DecisionTreeRegressor`to it:"
      ],
      "metadata": {
        "id": "2uM9bAyb-D1m"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "np.random.seed(42)\n",
        "X = np.random.rand(100, 1) - 0.5\n",
        "y = 3 * X[:, 0] ** 2 + 0.05 * np.random.randn(100)  # y = 3x² + Gaussian noise\n",
        "\n",
        "tree_reg1 = DecisionTreeRegressor(max_depth=2, random_state=42)\n",
        "tree_reg1.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LgoG2Cg58-oC",
        "outputId": "a702998b-7ece-466f-a438-3d2ada3b4f24"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DecisionTreeRegressor(max_depth=2, random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "A simpler way to train GBRT ensembles is to use Scikit-Learn’s `GradientBoostingRegressor` class (there’s also a `GradientBoostingClassifier` class for classification). Much like the `RandomForestRegressor` class, it has hyperparameters to control the growth of Decision Trees (e.g., `max_depth`, `min_samples_leaf`), as well as hyperparameters to control the ensemble training, such as the number of trees (`n_estimator`s). "
      ],
      "metadata": {
        "id": "PXBmpTZ6-TUf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "gbrt = GradientBoostingRegressor(max_depth=2, n_estimators=3, learning_rate=1.0, random_state=42)\n",
        "gbrt.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QUJv2gs2-pmE",
        "outputId": "016aca12-5341-403d-ede4-03d9ddf357ad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GradientBoostingRegressor(learning_rate=1.0, max_depth=2, n_estimators=3,\n",
              "                          random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 122
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The `learning_rate` hyperparameter scales the contribution of each tree. If you set it to a low value, such as 0.05, you will need more trees in the ensemble to fit the training set, but the predictions will usually generalize better. This is a regularization technique called shrinkage.\n",
        "\n",
        "To find the optimal number of trees, you could perform cross-validation using `GridSearchCV` or `RandomizedSearchCV`, as usual, but there’s a simpler way: if you set the `n_iter_no_change` hyperparameter to an integer value, say 10, then the `GradientBoostingRegressor` will automatically stop adding more trees during training if it sees that the last 10 trees didn’t help. This is simply early stopping, but with a little bit of patience: it tolerates having no progress for a few iterations before it stops. Let’s train the ensemble using early stopping:"
      ],
      "metadata": {
        "id": "HONBXa5P--iv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "gbrt_best = GradientBoostingRegressor(\n",
        "    max_depth=2, learning_rate=0.05, n_estimators=500,\n",
        "    n_iter_no_change=10, random_state=42)\n",
        "gbrt_best.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hSZK_Cvb-zke",
        "outputId": "c4b43dca-b453-4619-c547-b0bb44bb4c50"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GradientBoostingRegressor(learning_rate=0.05, max_depth=2, n_estimators=500,\n",
              "                          n_iter_no_change=10, random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "If you set `n_iter_no_change` too low, training may stop too early and the model will underfit. But if you set it too high, it will overfit instead. We also set a fairly small learning rate and a high number of estimators, but the actual number of estimators in the trained ensemble is much lower, thanks to early stopping:"
      ],
      "metadata": {
        "id": "-JbdlvKh_hwO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "gbrt_best.n_estimators_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8zloEgk6_UJj",
        "outputId": "77810466-8ed3-4d26-848c-2faea98bc8b4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "92"
            ]
          },
          "metadata": {},
          "execution_count": 124
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "When `n_iter_no_change` is set, the `fit(`) method automatically splits the training set into a smaller training set and a validation set: this allows it to evaluate the model’s performance each time it adds a new tree. The size of the validation set is controlled by the `validation_fraction` hyperparameter, which is 10% by default. The tol hyperparameter determines the maximum performance improvement that still counts as negligible. It defaults to 0.0001."
      ],
      "metadata": {
        "id": "L-vULKsT_mEO"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The GradientBoostingRegressor class also supports a subsample hyperparameter, which specifies the fraction of training instances to be used for training each tree. For example, if `subsample=0.25`, then each tree is trained on 25% of the training instances, selected randomly. As you can probably guess by now, this technique trades a higher bias for a lower variance. It also speeds up training considerably. This is called **Stochastic Gradient Boosting**."
      ],
      "metadata": {
        "id": "E8JMjIRF_xdd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## XGBoost"
      ],
      "metadata": {
        "id": "KHUBGzQqZRhM"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Classification task"
      ],
      "metadata": {
        "id": "HIsGDU0-bB-a"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here are the essential steps to build an XGBoost classification model in scikit-learn using cross-validation."
      ],
      "metadata": {
        "id": "l41vqfEkbKcK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "iris = datasets.load_iris()\n",
        "df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],columns= iris['feature_names'] + ['target'])\n",
        "df.head()"
      ],
      "metadata": {
        "id": "ZVg6wzfA_Vyq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "6a21e676-4e2b-44e6-f754-6b1c55ed0d58"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
              "0                5.1               3.5                1.4               0.2   \n",
              "1                4.9               3.0                1.4               0.2   \n",
              "2                4.7               3.2                1.3               0.2   \n",
              "3                4.6               3.1                1.5               0.2   \n",
              "4                5.0               3.6                1.4               0.2   \n",
              "\n",
              "   target  \n",
              "0     0.0  \n",
              "1     0.0  \n",
              "2     0.0  \n",
              "3     0.0  \n",
              "4     0.0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6ddd473a-163a-487f-8e4b-e82cec39574d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>sepal length (cm)</th>\n",
              "      <th>sepal width (cm)</th>\n",
              "      <th>petal length (cm)</th>\n",
              "      <th>petal width (cm)</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5.1</td>\n",
              "      <td>3.5</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4.7</td>\n",
              "      <td>3.2</td>\n",
              "      <td>1.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4.6</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.0</td>\n",
              "      <td>3.6</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6ddd473a-163a-487f-8e4b-e82cec39574d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-6ddd473a-163a-487f-8e4b-e82cec39574d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-6ddd473a-163a-487f-8e4b-e82cec39574d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(iris['data'], iris['target'], random_state=42)"
      ],
      "metadata": {
        "id": "G3R0nCm-Z4ly"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The following template is for building an XGBoost classifier"
      ],
      "metadata": {
        "id": "Q6wn7paXZpj-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "xgb = XGBClassifier(booster='gbtree', objective='multi:softprob', \n",
        "                    learning_rate=0.1, n_estimators=100, random_state=42, n_jobs=-1)"
      ],
      "metadata": {
        "id": "rsjJb0POZ8Y7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. `booster='gbtree'`: The booster is the base learner. It's the machine learning model that is constructed during every round of boosting. You may have guessed that 'gbtree' stands for gradient boosted tree, the XGBoost default base learner. It's uncommon but possible to work with other base learners, \n",
        "\n",
        "2. `objective='multi:softprob'`: Standard options for the objective can be viewed in the XGBoost official documentation, https://xgboost.readthedocs.io/en/latest/parameter.html, under Learning Task Parameters. The multi:softprob objective is a standard alternative to binary:logistic when the dataset includes multiple classes. It computes the probabilities of classification and chooses the highest one. If not explicitly stated, XGBoost will often find the right objective for you.\n",
        "\n",
        "3. `max_depth=6`: The max_depth of a tree determines the number of branches each tree has. It's one of the most important hyperparameters in making balanced predictions. XGBoost uses a default of 6, unlike random forests, which don't provide a value unless explicitly programmed.\n",
        "\n",
        "3. `learning_rate=0.1`: Within XGBoost, this hyperparameter is often referred to as eta. This hyperparameter limits the variance by reducing the weight of each tree to the given percentage. \n",
        "\n",
        "4. `n_estimators=100`: Popular among ensemble methods, `n_estimators` is the number of boosted trees in the model. Increasing this number while decreasing `learning_rate` can lead to more robust results."
      ],
      "metadata": {
        "id": "rDCT_1p9aNBi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "xgb.fit(X_train, y_train)\n",
        "y_pred = xgb.predict(X_test)\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "print('Score: ' + str(score))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QYLOzVdMazA5",
        "outputId": "e270ed09-e3f4-4ade-d8df-cc3f1fd6433d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 1.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xgb.get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q8LM-RkUQLDd",
        "outputId": "cd49710a-45fa-4a2e-ce82-860b66c41fa0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'base_score': 0.5,\n",
              " 'booster': 'gbtree',\n",
              " 'callbacks': None,\n",
              " 'colsample_bylevel': 1,\n",
              " 'colsample_bynode': 1,\n",
              " 'colsample_bytree': 1,\n",
              " 'early_stopping_rounds': None,\n",
              " 'enable_categorical': False,\n",
              " 'eval_metric': None,\n",
              " 'gamma': 0,\n",
              " 'gpu_id': -1,\n",
              " 'grow_policy': 'depthwise',\n",
              " 'importance_type': None,\n",
              " 'interaction_constraints': '',\n",
              " 'learning_rate': 0.1,\n",
              " 'max_bin': 256,\n",
              " 'max_cat_to_onehot': 4,\n",
              " 'max_delta_step': 0,\n",
              " 'max_depth': 6,\n",
              " 'max_leaves': 0,\n",
              " 'min_child_weight': 1,\n",
              " 'missing': nan,\n",
              " 'monotone_constraints': '()',\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': -1,\n",
              " 'num_parallel_tree': 1,\n",
              " 'objective': 'multi:softprob',\n",
              " 'predictor': 'auto',\n",
              " 'random_state': 42,\n",
              " 'reg_alpha': 0,\n",
              " 'reg_lambda': 1,\n",
              " 'sampling_method': 'uniform',\n",
              " 'scale_pos_weight': None,\n",
              " 'subsample': 1,\n",
              " 'tree_method': 'exact',\n",
              " 'use_label_encoder': False,\n",
              " 'validate_parameters': 1,\n",
              " 'verbosity': None}"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Regression task"
      ],
      "metadata": {
        "id": "e4LM00_-bExu"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here are the essential steps to build an XGBoost regression model in scikit-learn using cross-validation."
      ],
      "metadata": {
        "id": "7uFOM-SqbIZ6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X,y = datasets.load_diabetes(return_X_y=True)\n",
        "\n",
        "xgb = XGBRegressor(booster='gbtree', objective='reg:squarederror', \n",
        "                    learning_rate=0.1, n_estimators=100, random_state=42, n_jobs=-1)\n",
        "\n",
        "scores = cross_val_score(xgb, X, y, scoring='neg_mean_squared_error', cv=5)\n",
        "\n",
        "# Take square root of the scores\n",
        "rmse = np.sqrt(-scores)\n",
        "\n",
        "# Display accuracy\n",
        "print('RMSE:', np.round(rmse, 3))\n",
        "\n",
        "# Display mean score\n",
        "print('RMSE mean: %0.3f' % (rmse.mean()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q5w9Jclca8Gs",
        "outputId": "7d67260e-3a04-4ba2-84f6-b95dc8227cf1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "RMSE: [63.033 59.689 64.538 63.699 64.661]\n",
            "RMSE mean: 63.124\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xgb.fit(X,y)\n",
        "xgb.get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CbihBstGRQom",
        "outputId": "ce8ceff2-0207-4fac-e3c4-050b039028ad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'base_score': 0.5,\n",
              " 'booster': 'gbtree',\n",
              " 'callbacks': None,\n",
              " 'colsample_bylevel': 1,\n",
              " 'colsample_bynode': 1,\n",
              " 'colsample_bytree': 1,\n",
              " 'early_stopping_rounds': None,\n",
              " 'enable_categorical': False,\n",
              " 'eval_metric': None,\n",
              " 'gamma': 0,\n",
              " 'gpu_id': -1,\n",
              " 'grow_policy': 'depthwise',\n",
              " 'importance_type': None,\n",
              " 'interaction_constraints': '',\n",
              " 'learning_rate': 0.1,\n",
              " 'max_bin': 256,\n",
              " 'max_cat_to_onehot': 4,\n",
              " 'max_delta_step': 0,\n",
              " 'max_depth': 6,\n",
              " 'max_leaves': 0,\n",
              " 'min_child_weight': 1,\n",
              " 'missing': nan,\n",
              " 'monotone_constraints': '()',\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': -1,\n",
              " 'num_parallel_tree': 1,\n",
              " 'objective': 'reg:squarederror',\n",
              " 'predictor': 'auto',\n",
              " 'random_state': 42,\n",
              " 'reg_alpha': 0,\n",
              " 'reg_lambda': 1,\n",
              " 'sampling_method': 'uniform',\n",
              " 'scale_pos_weight': 1,\n",
              " 'subsample': 1,\n",
              " 'tree_method': 'exact',\n",
              " 'validate_parameters': 1,\n",
              " 'verbosity': None}"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Without a baseline of comparison, we have no idea what that score means. Converting the target column, `y`, into a pandas DataFrame with the `.describe()` method will give the quartiles and the general statistics of the predictor column, as follows:"
      ],
      "metadata": {
        "id": "HAY1SjZxboMz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pd.DataFrame(y).describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "F35eRsafbXnV",
        "outputId": "93fd4504-2aba-439f-e9b1-d85ed4f768fb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                0\n",
              "count  442.000000\n",
              "mean   152.133484\n",
              "std     77.093005\n",
              "min     25.000000\n",
              "25%     87.000000\n",
              "50%    140.500000\n",
              "75%    211.500000\n",
              "max    346.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-224dc2ae-fe4b-49ec-8593-de470c0efdde\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>442.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>152.133484</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>77.093005</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>25.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>87.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>140.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>211.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>346.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-224dc2ae-fe4b-49ec-8593-de470c0efdde')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-224dc2ae-fe4b-49ec-8593-de470c0efdde button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-224dc2ae-fe4b-49ec-8593-de470c0efdde');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "A score of 63.124 is less than 1 standard deviation, a respectable result."
      ],
      "metadata": {
        "id": "gqIRc8v6bubr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Speed comparsion"
      ],
      "metadata": {
        "id": "FLDE-3-RfD4c"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's now compare `GradientBoostingClassifier` and `XGBoostClassifier` with the [exoplanet dataset](https://www.kaggle.com/datasets/keplersmachines/kepler-labelled-time-series-data) for its speed "
      ],
      "metadata": {
        "id": "vKnuqcvafF1c"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://github.com/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/raw/master/Chapter04/exoplanets.csv"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YLKF2oiTbtbn",
        "outputId": "f23977cc-bfed-44af-ffe2-48f5252e0b51"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2022-05-21 11:00:29--  https://github.com/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/raw/master/Chapter04/exoplanets.csv\n",
            "Resolving github.com (github.com)... 140.82.114.4\n",
            "Connecting to github.com (github.com)|140.82.114.4|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://media.githubusercontent.com/media/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/master/Chapter04/exoplanets.csv [following]\n",
            "--2022-05-21 11:00:29--  https://media.githubusercontent.com/media/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/master/Chapter04/exoplanets.csv\n",
            "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 262223348 (250M) [text/plain]\n",
            "Saving to: ‘exoplanets.csv’\n",
            "\n",
            "exoplanets.csv      100%[===================>] 250.08M   193MB/s    in 1.3s    \n",
            "\n",
            "2022-05-21 11:00:34 (193 MB/s) - ‘exoplanets.csv’ saved [262223348/262223348]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('exoplanets.csv')\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "PqwtKb68dZi7",
        "outputId": "c02e13ce-5922-46c1-bea3-86fe790b8a16"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   LABEL   FLUX.1   FLUX.2   FLUX.3   FLUX.4   FLUX.5   FLUX.6  FLUX.7  \\\n",
              "0      2    93.85    83.81    20.10   -26.98   -39.56  -124.71 -135.18   \n",
              "1      2   -38.88   -33.83   -58.54   -40.09   -79.31   -72.81  -86.55   \n",
              "2      2   532.64   535.92   513.73   496.92   456.45   466.00  464.50   \n",
              "3      2   326.52   347.39   302.35   298.13   317.74   312.70  322.33   \n",
              "4      2 -1107.21 -1112.59 -1118.95 -1095.10 -1057.55 -1034.48 -998.34   \n",
              "\n",
              "    FLUX.8  FLUX.9  ...  FLUX.3188  FLUX.3189  FLUX.3190  FLUX.3191  \\\n",
              "0   -96.27  -79.89  ...     -78.07    -102.15    -102.15      25.13   \n",
              "1   -85.33  -83.97  ...      -3.28     -32.21     -32.21     -24.89   \n",
              "2   486.39  436.56  ...     -71.69      13.31      13.31     -29.89   \n",
              "3   311.31  312.42  ...       5.71      -3.73      -3.73      30.05   \n",
              "4 -1022.71 -989.57  ...    -594.37    -401.66    -401.66    -357.24   \n",
              "\n",
              "   FLUX.3192  FLUX.3193  FLUX.3194  FLUX.3195  FLUX.3196  FLUX.3197  \n",
              "0      48.57      92.54      39.32      61.42       5.08     -39.54  \n",
              "1      -4.86       0.76     -11.70       6.46      16.00      19.93  \n",
              "2     -20.88       5.06     -11.80     -28.91     -70.02     -96.67  \n",
              "3      20.03     -12.67      -8.77     -17.31     -17.35      13.98  \n",
              "4    -443.76    -438.54    -399.71    -384.65    -411.79    -510.54  \n",
              "\n",
              "[5 rows x 3198 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c563b13f-8b8b-463b-a27f-cf57cd8fa339\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>LABEL</th>\n",
              "      <th>FLUX.1</th>\n",
              "      <th>FLUX.2</th>\n",
              "      <th>FLUX.3</th>\n",
              "      <th>FLUX.4</th>\n",
              "      <th>FLUX.5</th>\n",
              "      <th>FLUX.6</th>\n",
              "      <th>FLUX.7</th>\n",
              "      <th>FLUX.8</th>\n",
              "      <th>FLUX.9</th>\n",
              "      <th>...</th>\n",
              "      <th>FLUX.3188</th>\n",
              "      <th>FLUX.3189</th>\n",
              "      <th>FLUX.3190</th>\n",
              "      <th>FLUX.3191</th>\n",
              "      <th>FLUX.3192</th>\n",
              "      <th>FLUX.3193</th>\n",
              "      <th>FLUX.3194</th>\n",
              "      <th>FLUX.3195</th>\n",
              "      <th>FLUX.3196</th>\n",
              "      <th>FLUX.3197</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2</td>\n",
              "      <td>93.85</td>\n",
              "      <td>83.81</td>\n",
              "      <td>20.10</td>\n",
              "      <td>-26.98</td>\n",
              "      <td>-39.56</td>\n",
              "      <td>-124.71</td>\n",
              "      <td>-135.18</td>\n",
              "      <td>-96.27</td>\n",
              "      <td>-79.89</td>\n",
              "      <td>...</td>\n",
              "      <td>-78.07</td>\n",
              "      <td>-102.15</td>\n",
              "      <td>-102.15</td>\n",
              "      <td>25.13</td>\n",
              "      <td>48.57</td>\n",
              "      <td>92.54</td>\n",
              "      <td>39.32</td>\n",
              "      <td>61.42</td>\n",
              "      <td>5.08</td>\n",
              "      <td>-39.54</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>-38.88</td>\n",
              "      <td>-33.83</td>\n",
              "      <td>-58.54</td>\n",
              "      <td>-40.09</td>\n",
              "      <td>-79.31</td>\n",
              "      <td>-72.81</td>\n",
              "      <td>-86.55</td>\n",
              "      <td>-85.33</td>\n",
              "      <td>-83.97</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.28</td>\n",
              "      <td>-32.21</td>\n",
              "      <td>-32.21</td>\n",
              "      <td>-24.89</td>\n",
              "      <td>-4.86</td>\n",
              "      <td>0.76</td>\n",
              "      <td>-11.70</td>\n",
              "      <td>6.46</td>\n",
              "      <td>16.00</td>\n",
              "      <td>19.93</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>532.64</td>\n",
              "      <td>535.92</td>\n",
              "      <td>513.73</td>\n",
              "      <td>496.92</td>\n",
              "      <td>456.45</td>\n",
              "      <td>466.00</td>\n",
              "      <td>464.50</td>\n",
              "      <td>486.39</td>\n",
              "      <td>436.56</td>\n",
              "      <td>...</td>\n",
              "      <td>-71.69</td>\n",
              "      <td>13.31</td>\n",
              "      <td>13.31</td>\n",
              "      <td>-29.89</td>\n",
              "      <td>-20.88</td>\n",
              "      <td>5.06</td>\n",
              "      <td>-11.80</td>\n",
              "      <td>-28.91</td>\n",
              "      <td>-70.02</td>\n",
              "      <td>-96.67</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "      <td>326.52</td>\n",
              "      <td>347.39</td>\n",
              "      <td>302.35</td>\n",
              "      <td>298.13</td>\n",
              "      <td>317.74</td>\n",
              "      <td>312.70</td>\n",
              "      <td>322.33</td>\n",
              "      <td>311.31</td>\n",
              "      <td>312.42</td>\n",
              "      <td>...</td>\n",
              "      <td>5.71</td>\n",
              "      <td>-3.73</td>\n",
              "      <td>-3.73</td>\n",
              "      <td>30.05</td>\n",
              "      <td>20.03</td>\n",
              "      <td>-12.67</td>\n",
              "      <td>-8.77</td>\n",
              "      <td>-17.31</td>\n",
              "      <td>-17.35</td>\n",
              "      <td>13.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>-1107.21</td>\n",
              "      <td>-1112.59</td>\n",
              "      <td>-1118.95</td>\n",
              "      <td>-1095.10</td>\n",
              "      <td>-1057.55</td>\n",
              "      <td>-1034.48</td>\n",
              "      <td>-998.34</td>\n",
              "      <td>-1022.71</td>\n",
              "      <td>-989.57</td>\n",
              "      <td>...</td>\n",
              "      <td>-594.37</td>\n",
              "      <td>-401.66</td>\n",
              "      <td>-401.66</td>\n",
              "      <td>-357.24</td>\n",
              "      <td>-443.76</td>\n",
              "      <td>-438.54</td>\n",
              "      <td>-399.71</td>\n",
              "      <td>-384.65</td>\n",
              "      <td>-411.79</td>\n",
              "      <td>-510.54</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 3198 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c563b13f-8b8b-463b-a27f-cf57cd8fa339')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c563b13f-8b8b-463b-a27f-cf57cd8fa339 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c563b13f-8b8b-463b-a27f-cf57cd8fa339');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split data into X and y\n",
        "X = df.iloc[:,1:]\n",
        "y = df.iloc[:,0]\n",
        "\n",
        "# Split data into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=2)"
      ],
      "metadata": {
        "id": "NiW1d_VCdk7t"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "start = time.time()\n",
        "\n",
        "gbr = GradientBoostingClassifier(n_estimators=100, max_depth=2, random_state=42)\n",
        "gbr.fit(X_train, y_train)\n",
        "y_pred = gbr.predict(X_test)\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "print('Score: ' + str(score))\n",
        "\n",
        "end = time.time()\n",
        "elapsed = end - start\n",
        "\n",
        "print('Run Time: ' + str(elapsed) + ' seconds')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Oslw4JXer5J",
        "outputId": "44836778-7c2a-4e79-e1a9-c240a978c260"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 0.9874213836477987\n",
            "Run Time: 248.6392641067505 seconds\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "start = time.time()\n",
        "\n",
        "# Instantiate the XGBRegressor, xg_reg\n",
        "xg_reg = XGBClassifier(n_estimators=100, max_depth=2, random_state=42)\n",
        "\n",
        "# class column has to start from 0 (as required since version 1.3.2).\n",
        "le = LabelEncoder()\n",
        "y_train = le.fit_transform(y_train)\n",
        "y_test = le.fit_transform(y_test)\n",
        "# Fit xg_reg to training set\n",
        "xg_reg.fit(X_train, y_train)\n",
        "\n",
        "# Predict labels of test set, y_pred\n",
        "y_pred = xg_reg.predict(X_test)\n",
        "\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "\n",
        "print('Score: ' + str(score))\n",
        "\n",
        "end = time.time()\n",
        "elapsed = end - start\n",
        "\n",
        "print('Run Time: ' + str(elapsed) + ' seconds')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KWi-TIeee2BU",
        "outputId": "9c05b685-c9ee-4502-b4af-b405459c22b4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 0.9913522012578616\n",
            "Run Time: 52.40076756477356 seconds\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "When it comes to big data, an algorithm six as fast can save weeks or months of computational time and resources! This advantage is huge in the world of big data. In the world of boosting, XGBoost is the model of choice due to its unparalleled speed and impressive accuracy."
      ],
      "metadata": {
        "id": "jmX7g10ofdRD"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Hyperparameter"
      ],
      "metadata": {
        "id": "AT6OUUregRfk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "XGBoost has many hyperparameters. XGBoost base learner hyperparameters incorporate all decision tree hyperparameters as a starting point. There are gradient boosting hyperparameters, since XGBoost is an enhanced version of gradient boosting."
      ],
      "metadata": {
        "id": "_YB4388QgiQr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://github.com/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/raw/master/Chapter06/heart_disease.csv"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H-GTv5RtffvG",
        "outputId": "2c48abfd-5517-41c8-fe37-188b1b2b059d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2022-05-21 05:49:41--  https://github.com/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/raw/master/Chapter06/heart_disease.csv\n",
            "Resolving github.com (github.com)... 140.82.121.3\n",
            "Connecting to github.com (github.com)|140.82.121.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://media.githubusercontent.com/media/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/master/Chapter06/heart_disease.csv [following]\n",
            "--2022-05-21 05:49:41--  https://media.githubusercontent.com/media/PacktPublishing/Hands-On-Gradient-Boosting-with-XGBoost-and-Scikit-learn/master/Chapter06/heart_disease.csv\n",
            "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 11328 (11K) [text/plain]\n",
            "Saving to: ‘heart_disease.csv.1’\n",
            "\n",
            "heart_disease.csv.1 100%[===================>]  11.06K  --.-KB/s    in 0s      \n",
            "\n",
            "2022-05-21 05:49:41 (101 MB/s) - ‘heart_disease.csv.1’ saved [11328/11328]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('heart_disease.csv')\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "Q6qyxyV5guIL",
        "outputId": "59a4af8c-697f-4897-eab8-7ee096d3bae1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
              "0   63    1   3       145   233    1        0      150      0      2.3      0   \n",
              "1   37    1   2       130   250    0        1      187      0      3.5      0   \n",
              "2   41    0   1       130   204    0        0      172      0      1.4      2   \n",
              "3   56    1   1       120   236    0        1      178      0      0.8      2   \n",
              "4   57    0   0       120   354    0        1      163      1      0.6      2   \n",
              "\n",
              "   ca  thal  target  \n",
              "0   0     1       1  \n",
              "1   0     2       1  \n",
              "2   0     2       1  \n",
              "3   0     2       1  \n",
              "4   0     2       1  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c435d19a-2687-423c-822f-ec8f63e1e7e9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>age</th>\n",
              "      <th>sex</th>\n",
              "      <th>cp</th>\n",
              "      <th>trestbps</th>\n",
              "      <th>chol</th>\n",
              "      <th>fbs</th>\n",
              "      <th>restecg</th>\n",
              "      <th>thalach</th>\n",
              "      <th>exang</th>\n",
              "      <th>oldpeak</th>\n",
              "      <th>slope</th>\n",
              "      <th>ca</th>\n",
              "      <th>thal</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>63</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>145</td>\n",
              "      <td>233</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>150</td>\n",
              "      <td>0</td>\n",
              "      <td>2.3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>37</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>130</td>\n",
              "      <td>250</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>187</td>\n",
              "      <td>0</td>\n",
              "      <td>3.5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>41</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>130</td>\n",
              "      <td>204</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>172</td>\n",
              "      <td>0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>56</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>120</td>\n",
              "      <td>236</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>178</td>\n",
              "      <td>0</td>\n",
              "      <td>0.8</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>57</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>120</td>\n",
              "      <td>354</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>163</td>\n",
              "      <td>1</td>\n",
              "      <td>0.6</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c435d19a-2687-423c-822f-ec8f63e1e7e9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c435d19a-2687-423c-822f-ec8f63e1e7e9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c435d19a-2687-423c-822f-ec8f63e1e7e9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split data into X and y\n",
        "X = df.iloc[:, :-1]\n",
        "y = df.iloc[:, -1]"
      ],
      "metadata": {
        "id": "161dPMyKgx8q"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Before tuning hyperparameters, let's build a classifier so that we can obtain a baseline score as a starting point."
      ],
      "metadata": {
        "id": "7H3mCXj8hItV"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "When fine-tuning hyperparameters, GridSearchCV and RandomizedSearchCV are the standard options. However, `cross_val_score` and `GridSearchCV/RandomizedSearchCV` do not split data the same way. One solution is to use `StratifiedKFold` whenever cross-validation is used.\n",
        "\n",
        "A stratified fold includes the same percentage of target values in each fold. If a dataset contains 60% 1s and 40% 0s in the target column, each stratified test set contains 60% 1s and 40% 0s. When folds are random, it's possible that one test set contains a 70-30 split while another contains a 50-50 split of target values."
      ],
      "metadata": {
        "id": "_sr6Y00Zhsyr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "kfold = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)"
      ],
      "metadata": {
        "id": "iZzwJoOFiM4V"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# The 'binary:logistic' objective is standard for binary classification in determining the loss function\n",
        "model = XGBClassifier(booster='gbtree', objective='binary:logistic', random_state=42)\n",
        "# Obtain scores of cross-validation\n",
        "scores = cross_val_score(model, X, y, cv=kfold)\n",
        "\n",
        "# Display accuracy\n",
        "print('Accuracy:', np.round(scores, 2))\n",
        "\n",
        "# Display mean accuracy\n",
        "print('Accuracy mean: %0.2f' % (scores.mean()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-IHX7FVVg_DM",
        "outputId": "9079ca1c-719f-42cf-bb3a-f4bd92254fce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: [0.85 0.72 0.74 0.82 0.78]\n",
            "Accuracy mean: 0.78\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The point here is to use the same folds to obtain new scores when fine-tuning hyperparameters with GridSearchCV and RandomizedSearchCV so that the comparison of scores is fair."
      ],
      "metadata": {
        "id": "fl1AJZVFiacl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def grid_search(params, random=False): \n",
        "    \n",
        "    xgb = XGBClassifier(booster='gbtree', objective='binary:logistic', random_state=42)\n",
        "    \n",
        "    kfold = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
        "    \n",
        "    if random:\n",
        "        grid = RandomizedSearchCV(xgb, params, cv=kfold, n_iter=20, n_jobs=-1, random_state=42)\n",
        "    else:\n",
        "        # Instantiate GridSearchCV as grid_reg\n",
        "        grid = GridSearchCV(xgb, params, cv=kfold, n_jobs=-1)\n",
        "    \n",
        "    # Fit grid_reg on X_train and y_train\n",
        "    grid.fit(X, y)\n",
        "\n",
        "    # Extract best params\n",
        "    best_params = grid.best_params_\n",
        "\n",
        "    # Print best params\n",
        "    print(\"Best params:\", best_params)\n",
        "    \n",
        "    # Compute best score\n",
        "    best_score = grid.best_score_\n",
        "\n",
        "    # Print best score\n",
        "    print(\"Best score: {:.5f}\".format(best_score))"
      ],
      "metadata": {
        "id": "hirVfvYGimTz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The XGBoost hyperparameters presented here are not meant to be exhaustive, but they are meant to be comprehensive. For a complete list of hyperparameters, read the official documentation, XGBoost Parameters, at https://xgboost.readthedocs.io/en/latest/parameter.html."
      ],
      "metadata": {
        "id": "FL9tTjDViy0U"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### learning_rate"
      ],
      "metadata": {
        "id": "C5vqCuCyjgvc"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`learning_rate` shrinks the weights of trees for each round of boosting. By lowering `learning_rate`, more trees are required to produce better scores. Lowering `learning_rate` prevents overfitting because the size of the weights carried forward is smaller.\n",
        "\n",
        "A default value of 0.3 is used. Here is a starting range for `learning_rate` as placed inside our grid_search function:"
      ],
      "metadata": {
        "id": "9XT4L0KzjjVd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'learning_rate':[0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-Gte23anjEJL",
        "outputId": "3a126484-69f4-4d5a-d987-37a98f755acb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'learning_rate': 0.5}\n",
            "Best score: 0.80525\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "lowering `learning_rate` may be advantageous when `n_estimators` goes up."
      ],
      "metadata": {
        "id": "bOJ9FRG8kidc"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### max_depth"
      ],
      "metadata": {
        "id": "y7l7T2CNkSs0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`max_depth` determines the length of the tree, equivalent to the number of rounds of splitting. Limiting `max_depth` prevents overfitting because the individual trees can only grow as far as `max_depth` allows. XGBoost provides a default `max_depth` value of six:"
      ],
      "metadata": {
        "id": "45DQQ5BUkVCk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'max_depth':[2, 3, 5, 6, 8]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yAjw3wKmjxRH",
        "outputId": "faa0934a-b667-46ed-b5ae-7d5417fbc051"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'max_depth': 2}\n",
            "Best score: 0.79552\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Changing `max_depth` from 6 to 2 gave a better score. The lower value for `max_depth` means variance has been reduced."
      ],
      "metadata": {
        "id": "L4Yw25h0kd_m"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### gamma"
      ],
      "metadata": {
        "id": "Y8K3Shxwkm80"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Known as a Lagrange multiplier, `gamma` provides a threshold that nodes must surpass before making further splits according to the loss function. There is no upper limit to the value of `gamma`. The default is 0, and anything over 10 is considered very high. Increasing `gamma` results in a more conservative model:"
      ],
      "metadata": {
        "id": "I7oV3JtKkoK0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'gamma':[0, 0.01, 0.1, 0.5, 1, 2]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3q-MZjFLkaMz",
        "outputId": "18e0426b-b9da-4e98-b015-3188f8d35e50"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'gamma': 1}\n",
            "Best score: 0.79880\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Changing gamma from 0 to 1 has resulted in a slight improvement."
      ],
      "metadata": {
        "id": "LlAtNDfuk4Yr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### min_child_weight"
      ],
      "metadata": {
        "id": "PE8jqb3lk7kb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`min_child_weight` refers to the minimum sum of weights required for a node to split into a child. If the sum of the weights is less than the value of `min_child_weight`, no further splits are made. `min_child_weight` reduces overfitting by increasing its value:"
      ],
      "metadata": {
        "id": "H9m98Cg9k84E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'min_child_weight':[1, 2, 3, 4, 5]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FXgP8k9pkyMG",
        "outputId": "31488e96-01c4-4c94-9a9b-af13fc07c6f8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'min_child_weight': 5}\n",
            "Best score: 0.81202\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "A slight adjustment to `min_child_weight` form 1 to 5 gives the best results yet."
      ],
      "metadata": {
        "id": "_Uj4v6NclNTk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### subsample"
      ],
      "metadata": {
        "id": "lpUThvNwlYR0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The `subsample` hyperparameter limits the percentage of training instances (rows) for each boosting round. Decreasing `subsample` from 100% reduces overfitting:"
      ],
      "metadata": {
        "id": "0IbGAEDvlaEE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'subsample':[0.5, 0.7, 0.8, 0.9, 1]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BZmXj7R8lFqH",
        "outputId": "fa57581a-ec6d-42e4-8310-87b1f56a5ca8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'subsample': 0.5}\n",
            "Best score: 0.82525\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The score has improved by a slight amount once again, indicating a small presence of overfitting."
      ],
      "metadata": {
        "id": "oT8tJABxlkLt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### colsample_bytree"
      ],
      "metadata": {
        "id": "STS11E58lpS-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Similar to `subsample`, `colsample_bytree` randomly selects particular columns according to the given percentage. `colsample_bytree` is useful for limiting the influence of columns and reducing variance. Note that `colsample_bytree` takes a percentage as input, not the number of columns:"
      ],
      "metadata": {
        "id": "3k5abqx8lq_F"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'colsample_bytree':[0.5, 0.7, 0.8, 0.9, 1]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "02a7625clgR7",
        "outputId": "a9d61e46-da14-47ce-cf48-af670272c8aa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'colsample_bytree': 0.5}\n",
            "Best score: 0.79874\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You are encouraged to try `colsample_bylevel` and `colsample_bynode` on your own. `colsample_bylevel` randomly selects columns for each tree depth, and `colsample_bynode` randomly selects columns when evaluating each tree split."
      ],
      "metadata": {
        "id": "C-6xeLBMl9nt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### n_estimators"
      ],
      "metadata": {
        "id": "5ISk5sini444"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Recall that `n_estimators` provides the number of trees in the ensemble. In the case of XGBoost, `n_estimators` is the number of trees trained on the residuals. Initialize a grid search of `n_estimators` with the default of 100, then double the number of trees through 800 as follows:"
      ],
      "metadata": {
        "id": "QbeWxPWei76T"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search(params={'n_estimators':[100, 200, 400, 800]})"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "djxGbftfhKm8",
        "outputId": "a420b09b-6e7a-4552-cd6f-f9cf58adbe15"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best params: {'n_estimators': 200}\n",
            "Best score: 0.79219\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Since our dataset is small, increasing `n_estimators` did not produce better results."
      ],
      "metadata": {
        "id": "QYDz9YREjcLO"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Applying early stopping"
      ],
      "metadata": {
        "id": "s42ccMKbmDLF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`early_stopping_rounds` is not a hyperparameter, but a strategy for optimizing the `n_estimators` hyperparameter.\n",
        "\n",
        "Normally when choosing hyperparameters, a test score is given after all boosting rounds are complete. To use early stopping, we need a test score after each round. `eval_metric` and `eval_set` may be used as parameters for `.fit` to generate test scores for each training round. `eval_metric` provides the scoring method, commonly 'error' for classification, and 'rmse' for regression. `eval_set` provides the test to be evaluated, commonly X_test and y_test."
      ],
      "metadata": {
        "id": "_f41kkk0mPil"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The following steps display an evaluation metric for each round of training with the default `n_estimators=100:`"
      ],
      "metadata": {
        "id": "b_ND_pUUm9md"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)\n",
        "model = XGBClassifier(booster='gbtree', objective='binary:logistic', random_state=42)\n",
        "eval_set = [(X_test, y_test)]\n",
        "eval_metric='error'\n",
        "model.fit(X_train, y_train, eval_metric=eval_metric, eval_set=eval_set)\n",
        "# make predictions for test data\n",
        "y_pred = model.predict(X_test)\n",
        "# evaluate predictions\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(\"Accuracy: %.2f%%\" % (accuracy * 100.0))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yUgNVaUJl10d",
        "outputId": "d8d452c7-7fa7-4679-c98c-e04efbb02ada"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0]\tvalidation_0-error:0.23684\n",
            "[1]\tvalidation_0-error:0.22368\n",
            "[2]\tvalidation_0-error:0.22368\n",
            "[3]\tvalidation_0-error:0.21053\n",
            "[4]\tvalidation_0-error:0.22368\n",
            "[5]\tvalidation_0-error:0.18421\n",
            "[6]\tvalidation_0-error:0.21053\n",
            "[7]\tvalidation_0-error:0.22368\n",
            "[8]\tvalidation_0-error:0.19737\n",
            "[9]\tvalidation_0-error:0.19737\n",
            "[10]\tvalidation_0-error:0.18421\n",
            "[11]\tvalidation_0-error:0.18421\n",
            "[12]\tvalidation_0-error:0.19737\n",
            "[13]\tvalidation_0-error:0.17105\n",
            "[14]\tvalidation_0-error:0.18421\n",
            "[15]\tvalidation_0-error:0.18421\n",
            "[16]\tvalidation_0-error:0.18421\n",
            "[17]\tvalidation_0-error:0.18421\n",
            "[18]\tvalidation_0-error:0.18421\n",
            "[19]\tvalidation_0-error:0.19737\n",
            "[20]\tvalidation_0-error:0.18421\n",
            "[21]\tvalidation_0-error:0.18421\n",
            "[22]\tvalidation_0-error:0.18421\n",
            "[23]\tvalidation_0-error:0.18421\n",
            "[24]\tvalidation_0-error:0.18421\n",
            "[25]\tvalidation_0-error:0.18421\n",
            "[26]\tvalidation_0-error:0.18421\n",
            "[27]\tvalidation_0-error:0.18421\n",
            "[28]\tvalidation_0-error:0.18421\n",
            "[29]\tvalidation_0-error:0.18421\n",
            "[30]\tvalidation_0-error:0.18421\n",
            "[31]\tvalidation_0-error:0.17105\n",
            "[32]\tvalidation_0-error:0.18421\n",
            "[33]\tvalidation_0-error:0.18421\n",
            "[34]\tvalidation_0-error:0.18421\n",
            "[35]\tvalidation_0-error:0.17105\n",
            "[36]\tvalidation_0-error:0.17105\n",
            "[37]\tvalidation_0-error:0.17105\n",
            "[38]\tvalidation_0-error:0.18421\n",
            "[39]\tvalidation_0-error:0.17105\n",
            "[40]\tvalidation_0-error:0.18421\n",
            "[41]\tvalidation_0-error:0.18421\n",
            "[42]\tvalidation_0-error:0.18421\n",
            "[43]\tvalidation_0-error:0.18421\n",
            "[44]\tvalidation_0-error:0.18421\n",
            "[45]\tvalidation_0-error:0.18421\n",
            "[46]\tvalidation_0-error:0.18421\n",
            "[47]\tvalidation_0-error:0.18421\n",
            "[48]\tvalidation_0-error:0.18421\n",
            "[49]\tvalidation_0-error:0.18421\n",
            "[50]\tvalidation_0-error:0.18421\n",
            "[51]\tvalidation_0-error:0.18421\n",
            "[52]\tvalidation_0-error:0.18421\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/xgboost/sklearn.py:797: UserWarning: `eval_metric` in `fit` method is deprecated for better compatibility with scikit-learn, use `eval_metric` in constructor or`set_params` instead.\n",
            "  UserWarning,\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[53]\tvalidation_0-error:0.18421\n",
            "[54]\tvalidation_0-error:0.18421\n",
            "[55]\tvalidation_0-error:0.18421\n",
            "[56]\tvalidation_0-error:0.18421\n",
            "[57]\tvalidation_0-error:0.18421\n",
            "[58]\tvalidation_0-error:0.18421\n",
            "[59]\tvalidation_0-error:0.18421\n",
            "[60]\tvalidation_0-error:0.18421\n",
            "[61]\tvalidation_0-error:0.18421\n",
            "[62]\tvalidation_0-error:0.18421\n",
            "[63]\tvalidation_0-error:0.18421\n",
            "[64]\tvalidation_0-error:0.18421\n",
            "[65]\tvalidation_0-error:0.18421\n",
            "[66]\tvalidation_0-error:0.18421\n",
            "[67]\tvalidation_0-error:0.18421\n",
            "[68]\tvalidation_0-error:0.18421\n",
            "[69]\tvalidation_0-error:0.18421\n",
            "[70]\tvalidation_0-error:0.18421\n",
            "[71]\tvalidation_0-error:0.18421\n",
            "[72]\tvalidation_0-error:0.18421\n",
            "[73]\tvalidation_0-error:0.18421\n",
            "[74]\tvalidation_0-error:0.18421\n",
            "[75]\tvalidation_0-error:0.18421\n",
            "[76]\tvalidation_0-error:0.18421\n",
            "[77]\tvalidation_0-error:0.18421\n",
            "[78]\tvalidation_0-error:0.18421\n",
            "[79]\tvalidation_0-error:0.18421\n",
            "[80]\tvalidation_0-error:0.18421\n",
            "[81]\tvalidation_0-error:0.18421\n",
            "[82]\tvalidation_0-error:0.18421\n",
            "[83]\tvalidation_0-error:0.18421\n",
            "[84]\tvalidation_0-error:0.18421\n",
            "[85]\tvalidation_0-error:0.18421\n",
            "[86]\tvalidation_0-error:0.18421\n",
            "[87]\tvalidation_0-error:0.18421\n",
            "[88]\tvalidation_0-error:0.18421\n",
            "[89]\tvalidation_0-error:0.18421\n",
            "[90]\tvalidation_0-error:0.18421\n",
            "[91]\tvalidation_0-error:0.18421\n",
            "[92]\tvalidation_0-error:0.18421\n",
            "[93]\tvalidation_0-error:0.18421\n",
            "[94]\tvalidation_0-error:0.18421\n",
            "[95]\tvalidation_0-error:0.18421\n",
            "[96]\tvalidation_0-error:0.18421\n",
            "[97]\tvalidation_0-error:0.18421\n",
            "[98]\tvalidation_0-error:0.18421\n",
            "[99]\tvalidation_0-error:0.18421\n",
            "Accuracy: 81.58%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We know that `StratifiedKFold` cross-validation gives a mean accuracy of 78% when n_estimators=100. The disparity in scores comes from the difference in test sets."
      ],
      "metadata": {
        "id": "YUYIc8cvm3X1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### early_stopping_rounds"
      ],
      "metadata": {
        "id": "rHYUq8m-nCed"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`early_stopping_rounds` is an optional parameter to include with `eval_metric` and `eval_set` when fitting a model. Let's try `early_stopping_rounds=10`.\n",
        "The previous code is repeated with `early_stopping_rounds=10` added in:"
      ],
      "metadata": {
        "id": "rbV-EIxdnD3F"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = XGBClassifier(booster='gbtree', objective='binary:logistic', random_state=42)\n",
        "eval_set = [(X_test, y_test)]\n",
        "eval_metric=\"error\"\n",
        "model.fit(X_train, y_train, eval_metric=eval_metric, eval_set=eval_set, early_stopping_rounds=10, verbose=True)\n",
        "y_pred = model.predict(X_test)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(\"Accuracy: %.2f%%\" % (accuracy * 100.0))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S-8zyAq5nQQN",
        "outputId": "87c02e94-59b9-4db7-e953-0654af1f16c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0]\tvalidation_0-error:0.23684\n",
            "[1]\tvalidation_0-error:0.22368\n",
            "[2]\tvalidation_0-error:0.22368\n",
            "[3]\tvalidation_0-error:0.21053\n",
            "[4]\tvalidation_0-error:0.22368\n",
            "[5]\tvalidation_0-error:0.18421\n",
            "[6]\tvalidation_0-error:0.21053\n",
            "[7]\tvalidation_0-error:0.22368\n",
            "[8]\tvalidation_0-error:0.19737\n",
            "[9]\tvalidation_0-error:0.19737\n",
            "[10]\tvalidation_0-error:0.18421\n",
            "[11]\tvalidation_0-error:0.18421\n",
            "[12]\tvalidation_0-error:0.19737\n",
            "[13]\tvalidation_0-error:0.17105\n",
            "[14]\tvalidation_0-error:0.18421\n",
            "[15]\tvalidation_0-error:0.18421\n",
            "[16]\tvalidation_0-error:0.18421\n",
            "[17]\tvalidation_0-error:0.18421\n",
            "[18]\tvalidation_0-error:0.18421\n",
            "[19]\tvalidation_0-error:0.19737\n",
            "[20]\tvalidation_0-error:0.18421\n",
            "[21]\tvalidation_0-error:0.18421\n",
            "[22]\tvalidation_0-error:0.18421\n",
            "Accuracy: 82.89%\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/xgboost/sklearn.py:797: UserWarning: `eval_metric` in `fit` method is deprecated for better compatibility with scikit-learn, use `eval_metric` in constructor or`set_params` instead.\n",
            "  UserWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/xgboost/sklearn.py:797: UserWarning: `early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
            "  UserWarning,\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "A more thorough approach is to use larger values, say, `n_estimators = 5000` and `early_stopping_rounds=100`. By setting `early_stopping_rounds=100`, you are guaranteed to reach the default of 100 boosted trees presented by XGBoost.\n",
        "Here is the code that gives a maximum of 5,000 trees and that will stop after 100 consecutive rounds fail to find any improvement:"
      ],
      "metadata": {
        "id": "iFNCBPFVnoI-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = XGBClassifier(random_state=42, n_estimators=5000)\n",
        "eval_set = [(X_test, y_test)]\n",
        "eval_metric=\"error\"\n",
        "model.fit(X_train, y_train, eval_metric=eval_metric, eval_set=eval_set, early_stopping_rounds=100)\n",
        "y_pred = model.predict(X_test)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(\"Accuracy: %.2f%%\" % (accuracy * 100.0))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TM8sE9vJmqrf",
        "outputId": "be9a8754-3fa0-4e34-d8a1-1a9407333ec9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0]\tvalidation_0-error:0.23684\n",
            "[1]\tvalidation_0-error:0.22368\n",
            "[2]\tvalidation_0-error:0.22368\n",
            "[3]\tvalidation_0-error:0.21053\n",
            "[4]\tvalidation_0-error:0.22368\n",
            "[5]\tvalidation_0-error:0.18421\n",
            "[6]\tvalidation_0-error:0.21053\n",
            "[7]\tvalidation_0-error:0.22368\n",
            "[8]\tvalidation_0-error:0.19737\n",
            "[9]\tvalidation_0-error:0.19737\n",
            "[10]\tvalidation_0-error:0.18421\n",
            "[11]\tvalidation_0-error:0.18421\n",
            "[12]\tvalidation_0-error:0.19737\n",
            "[13]\tvalidation_0-error:0.17105\n",
            "[14]\tvalidation_0-error:0.18421\n",
            "[15]\tvalidation_0-error:0.18421\n",
            "[16]\tvalidation_0-error:0.18421\n",
            "[17]\tvalidation_0-error:0.18421\n",
            "[18]\tvalidation_0-error:0.18421\n",
            "[19]\tvalidation_0-error:0.19737\n",
            "[20]\tvalidation_0-error:0.18421\n",
            "[21]\tvalidation_0-error:0.18421\n",
            "[22]\tvalidation_0-error:0.18421\n",
            "[23]\tvalidation_0-error:0.18421\n",
            "[24]\tvalidation_0-error:0.18421\n",
            "[25]\tvalidation_0-error:0.18421\n",
            "[26]\tvalidation_0-error:0.18421\n",
            "[27]\tvalidation_0-error:0.18421\n",
            "[28]\tvalidation_0-error:0.18421\n",
            "[29]\tvalidation_0-error:0.18421\n",
            "[30]\tvalidation_0-error:0.18421\n",
            "[31]\tvalidation_0-error:0.17105\n",
            "[32]\tvalidation_0-error:0.18421\n",
            "[33]\tvalidation_0-error:0.18421\n",
            "[34]\tvalidation_0-error:0.18421\n",
            "[35]\tvalidation_0-error:0.17105\n",
            "[36]\tvalidation_0-error:0.17105\n",
            "[37]\tvalidation_0-error:0.17105\n",
            "[38]\tvalidation_0-error:0.18421\n",
            "[39]\tvalidation_0-error:0.17105\n",
            "[40]\tvalidation_0-error:0.18421\n",
            "[41]\tvalidation_0-error:0.18421\n",
            "[42]\tvalidation_0-error:0.18421\n",
            "[43]\tvalidation_0-error:0.18421\n",
            "[44]\tvalidation_0-error:0.18421\n",
            "[45]\tvalidation_0-error:0.18421\n",
            "[46]\tvalidation_0-error:0.18421\n",
            "[47]\tvalidation_0-error:0.18421\n",
            "[48]\tvalidation_0-error:0.18421\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/xgboost/sklearn.py:797: UserWarning: `eval_metric` in `fit` method is deprecated for better compatibility with scikit-learn, use `eval_metric` in constructor or`set_params` instead.\n",
            "  UserWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/xgboost/sklearn.py:797: UserWarning: `early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
            "  UserWarning,\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[49]\tvalidation_0-error:0.18421\n",
            "[50]\tvalidation_0-error:0.18421\n",
            "[51]\tvalidation_0-error:0.18421\n",
            "[52]\tvalidation_0-error:0.18421\n",
            "[53]\tvalidation_0-error:0.18421\n",
            "[54]\tvalidation_0-error:0.18421\n",
            "[55]\tvalidation_0-error:0.18421\n",
            "[56]\tvalidation_0-error:0.18421\n",
            "[57]\tvalidation_0-error:0.18421\n",
            "[58]\tvalidation_0-error:0.18421\n",
            "[59]\tvalidation_0-error:0.18421\n",
            "[60]\tvalidation_0-error:0.18421\n",
            "[61]\tvalidation_0-error:0.18421\n",
            "[62]\tvalidation_0-error:0.18421\n",
            "[63]\tvalidation_0-error:0.18421\n",
            "[64]\tvalidation_0-error:0.18421\n",
            "[65]\tvalidation_0-error:0.18421\n",
            "[66]\tvalidation_0-error:0.18421\n",
            "[67]\tvalidation_0-error:0.18421\n",
            "[68]\tvalidation_0-error:0.18421\n",
            "[69]\tvalidation_0-error:0.18421\n",
            "[70]\tvalidation_0-error:0.18421\n",
            "[71]\tvalidation_0-error:0.18421\n",
            "[72]\tvalidation_0-error:0.18421\n",
            "[73]\tvalidation_0-error:0.18421\n",
            "[74]\tvalidation_0-error:0.18421\n",
            "[75]\tvalidation_0-error:0.18421\n",
            "[76]\tvalidation_0-error:0.18421\n",
            "[77]\tvalidation_0-error:0.18421\n",
            "[78]\tvalidation_0-error:0.18421\n",
            "[79]\tvalidation_0-error:0.18421\n",
            "[80]\tvalidation_0-error:0.18421\n",
            "[81]\tvalidation_0-error:0.18421\n",
            "[82]\tvalidation_0-error:0.18421\n",
            "[83]\tvalidation_0-error:0.18421\n",
            "[84]\tvalidation_0-error:0.18421\n",
            "[85]\tvalidation_0-error:0.18421\n",
            "[86]\tvalidation_0-error:0.18421\n",
            "[87]\tvalidation_0-error:0.18421\n",
            "[88]\tvalidation_0-error:0.18421\n",
            "[89]\tvalidation_0-error:0.18421\n",
            "[90]\tvalidation_0-error:0.18421\n",
            "[91]\tvalidation_0-error:0.18421\n",
            "[92]\tvalidation_0-error:0.18421\n",
            "[93]\tvalidation_0-error:0.18421\n",
            "[94]\tvalidation_0-error:0.18421\n",
            "[95]\tvalidation_0-error:0.18421\n",
            "[96]\tvalidation_0-error:0.18421\n",
            "[97]\tvalidation_0-error:0.18421\n",
            "[98]\tvalidation_0-error:0.18421\n",
            "[99]\tvalidation_0-error:0.18421\n",
            "[100]\tvalidation_0-error:0.18421\n",
            "[101]\tvalidation_0-error:0.18421\n",
            "[102]\tvalidation_0-error:0.18421\n",
            "[103]\tvalidation_0-error:0.18421\n",
            "[104]\tvalidation_0-error:0.18421\n",
            "[105]\tvalidation_0-error:0.18421\n",
            "[106]\tvalidation_0-error:0.18421\n",
            "[107]\tvalidation_0-error:0.18421\n",
            "[108]\tvalidation_0-error:0.18421\n",
            "[109]\tvalidation_0-error:0.18421\n",
            "[110]\tvalidation_0-error:0.18421\n",
            "[111]\tvalidation_0-error:0.18421\n",
            "[112]\tvalidation_0-error:0.18421\n",
            "Accuracy: 82.89%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "After 100 rounds of boosting, the score provided by 13 trees is the best."
      ],
      "metadata": {
        "id": "bSdrD_UWn9z2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Automatically hyperparamter tuning"
      ],
      "metadata": {
        "id": "826da7dWNGeH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "You are encourage to try https://github.com/optuna/optuna-examples/blob/main/xgboost/xgboost_simple.py for hyperparameter tuning."
      ],
      "metadata": {
        "id": "O76mLAlyNNU-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### For categorical variable and missing value"
      ],
      "metadata": {
        "id": "SHHXkhtpJ0w9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "XGBoost has experiment support for categorical variable, you can check out here: https://xgboost.readthedocs.io/en/latest/tutorials/categorical.html. But it only works for a few tree methods, it is still recommend to encode your data https://www.kaggle.com/code/shahules/an-overview-of-encoding-techniques/notebook. Missing value, on the other hand can be handled by XGBoost as described at https://xgboost.readthedocs.io/en/stable/faq.html#how-to-deal-with-missing-values."
      ],
      "metadata": {
        "id": "BNREDHc_Ig5E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Select target\n",
        "data = pd.read_csv('melb_data.csv')\n",
        "y = data.Price\n",
        "\n",
        "# To keep things simple, we'll split the columns into numerical can categorical features\n",
        "melb_predictors = data.drop(['Price', 'Date', 'Address'], axis=1)\n",
        "cat_col = melb_predictors.select_dtypes(exclude=['int64','float64'])\n",
        "\n",
        "# Divide data into training and validation subsets\n",
        "X, X_v, y_train, y_valid = train_test_split(melb_predictors, y, train_size=0.8, test_size=0.2, random_state=0)\n",
        "X_train = X.select_dtypes(exclude=['object'])\n",
        "X_valid = X_v.select_dtypes(exclude=['object'])\n",
        "X_train_cat = X.select_dtypes(exclude=['int64','float64'])\n",
        "X_valid_cat = X_v.select_dtypes(exclude=['int64','float64'])"
      ],
      "metadata": {
        "id": "qFSRGgsFI4H_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for col in X_train_cat.columns:\n",
        "  X[col] = X[col].astype('category')\n",
        "  X_v[col] = X_v[col].astype('category')"
      ],
      "metadata": {
        "id": "5dJbtmI5-U7t"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "xgb = XGBRegressor(booster='gbtree', objective='reg:squarederror', \n",
        "                    random_state=42, n_jobs=-1) # You can either specify missing=-9999 or leave it as it is\n",
        "xgb.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PqKgSpglLclJ",
        "outputId": "cb544b30-328a-404c-a403-44aedea4f4d8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XGBRegressor(base_score=0.5, booster='gbtree', callbacks=None,\n",
              "             colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
              "             early_stopping_rounds=None, enable_categorical=False,\n",
              "             eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',\n",
              "             importance_type=None, interaction_constraints='',\n",
              "             learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
              "             max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
              "             missing=nan, monotone_constraints='()', n_estimators=100,\n",
              "             n_jobs=-1, num_parallel_tree=1, predictor='auto', random_state=42,\n",
              "             reg_alpha=0, reg_lambda=1, ...)"
            ]
          },
          "metadata": {},
          "execution_count": 176
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "preds = xgb.predict(X_valid)"
      ],
      "metadata": {
        "id": "9JpLF4ECLnha"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1yuRQhnjLtLi",
        "outputId": "875024d9-f93b-4b39-e0af-0dcd88e1253d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "173639.76325478646"
            ]
          },
          "metadata": {},
          "execution_count": 178
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "explainer = shap.Explainer(xgb)\n",
        "shap_values = explainer(X_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oVDw3PEvgj6b",
        "outputId": "00821643-0773-4399-958e-d155e52e3005"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "ntree_limit is deprecated, use `iteration_range` or model slicing instead.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### SHAP values"
      ],
      "metadata": {
        "id": "Z7FHMo3tj_Bj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "shap.summary_plot(shap_values, X_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 411
        },
        "id": "v2dATeDRgwLh",
        "outputId": "0b950859-186a-426a-98b7-23b796a1183e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x453.6 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Lightgbm"
      ],
      "metadata": {
        "id": "t58ejslZMJ_P"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Classification task"
      ],
      "metadata": {
        "id": "RiQQUpHDP0In"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "iris = datasets.load_iris()\n",
        "df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],columns= iris['feature_names'] + ['target'])\n",
        "X_train, X_test, y_train, y_test = train_test_split(iris['data'], iris['target'], random_state=42)"
      ],
      "metadata": {
        "id": "X50nza8ILxQb"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf = lgb.LGBMClassifier(boosting_type='gbdt', learning_rate=0.1, n_estimators=100, random_state=42, n_jobs=-1)\n",
        "clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ff20VXGMQIXR",
        "outputId": "1a131638-cefd-45df-9754-5e8fd88934f0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMClassifier(random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clf.fit(X_train, y_train)\n",
        "y_pred = clf.predict(X_test)\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "print('Score: ' + str(score))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zC6X-CFEQ2m3",
        "outputId": "d35489eb-ef3d-4ee6-a334-ddf2294e0419"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 1.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clf.get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KcTMWhn8QBil",
        "outputId": "3e98dceb-2ef7-4c64-b601-18a0a5115e4b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'boosting_type': 'gbdt',\n",
              " 'class_weight': None,\n",
              " 'colsample_bytree': 1.0,\n",
              " 'importance_type': 'split',\n",
              " 'learning_rate': 0.1,\n",
              " 'max_depth': -1,\n",
              " 'min_child_samples': 20,\n",
              " 'min_child_weight': 0.001,\n",
              " 'min_split_gain': 0.0,\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': -1,\n",
              " 'num_leaves': 31,\n",
              " 'objective': None,\n",
              " 'random_state': 42,\n",
              " 'reg_alpha': 0.0,\n",
              " 'reg_lambda': 0.0,\n",
              " 'silent': 'warn',\n",
              " 'subsample': 1.0,\n",
              " 'subsample_for_bin': 200000,\n",
              " 'subsample_freq': 0}"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Regression task"
      ],
      "metadata": {
        "id": "W_yDFE8oRmPh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X,y = datasets.load_diabetes(return_X_y=True)\n",
        "\n",
        "lgbr = lgb.LGBMRegressor(boosting_type='gbdt', learning_rate=0.1, n_estimators=100, random_state=42, n_jobs=-1)\n",
        "\n",
        "scores = cross_val_score(lgbr, X, y, scoring='neg_mean_squared_error', cv=5)\n",
        "\n",
        "# Take square root of the scores\n",
        "rmse = np.sqrt(-scores)\n",
        "\n",
        "# Display accuracy\n",
        "print('RMSE:', np.round(rmse, 3))\n",
        "\n",
        "# Display mean score\n",
        "print('RMSE mean: %0.3f' % (rmse.mean()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SQnUryTZQ-l_",
        "outputId": "8e67190f-8a7c-4ad0-adfb-d54dd41d6fd9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "RMSE: [56.081 59.172 63.191 61.833 60.542]\n",
            "RMSE mean: 60.164\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lgbr.fit(X,y)\n",
        "lgbr.get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DeDcf6bhRdOO",
        "outputId": "d9451285-9cee-438a-b857-58aec5d3ed3d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'boosting_type': 'gbdt',\n",
              " 'class_weight': None,\n",
              " 'colsample_bytree': 1.0,\n",
              " 'importance_type': 'split',\n",
              " 'learning_rate': 0.1,\n",
              " 'max_depth': -1,\n",
              " 'min_child_samples': 20,\n",
              " 'min_child_weight': 0.001,\n",
              " 'min_split_gain': 0.0,\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': -1,\n",
              " 'num_leaves': 31,\n",
              " 'objective': None,\n",
              " 'random_state': 42,\n",
              " 'reg_alpha': 0.0,\n",
              " 'reg_lambda': 0.0,\n",
              " 'silent': 'warn',\n",
              " 'subsample': 1.0,\n",
              " 'subsample_for_bin': 200000,\n",
              " 'subsample_freq': 0}"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Speed"
      ],
      "metadata": {
        "id": "gEFp_iDMR8Jw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('exoplanets.csv')\n",
        "# Split data into X and y\n",
        "X = df.iloc[:,1:]\n",
        "y = df.iloc[:,0]\n",
        "\n",
        "# Split data into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=2)"
      ],
      "metadata": {
        "id": "sFvSB-AoSM-g"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "start = time.time()\n",
        "\n",
        "# Instantiate the XGBRegressor, xg_reg\n",
        "lg_reg = lgb.LGBMClassifier(n_estimators=100, max_depth=2, random_state=42)\n",
        "\n",
        "# Fit xg_reg to training set\n",
        "lg_reg.fit(X_train, y_train)\n",
        "\n",
        "# Predict labels of test set, y_pred\n",
        "y_pred = lg_reg.predict(X_test)\n",
        "\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "\n",
        "print('Score: ' + str(score))\n",
        "\n",
        "end = time.time()\n",
        "elapsed = end - start\n",
        "\n",
        "print('Run Time: ' + str(elapsed) + ' seconds')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P2mf-IgRR3zt",
        "outputId": "108b41d9-5f96-48e3-d77c-5a413ee77d85"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 0.9913522012578616\n",
            "Run Time: 7.302560567855835 seconds\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Hyperparameter"
      ],
      "metadata": {
        "id": "rfvM90-HSayn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Following set of practices can be used to improve your model efficiency.\n",
        "\n",
        "* **num_leaves** : This is the main parameter to control the complexity of the tree model. Ideally, the value of `num_leaves` should be less than or equal to 2^(max_depth). Value more than this will result in overfitting.\n",
        "\n",
        "* **min_data_in_leaf** : Setting it to a large value can avoid growing too deep a tree, but may cause under-fitting. In practice, setting it to hundreds or thousands is enough for a large dataset.\n",
        "\n",
        "* **max_depth** : We also can use `max_depth` to limit the tree depth explicitly."
      ],
      "metadata": {
        "id": "vJIrKihuSd83"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. **For Faster Speed**\n",
        "\n",
        "* Use bagging by setting `bagging_fraction` and `bagging_freq`.\n",
        "* Use feature sub-sampling by setting `feature_fraction`.\n",
        "* Use small `max_bin`.\n",
        "* Use `save_binary` to speed up data loading in future learning."
      ],
      "metadata": {
        "id": "Xiv_JVHiS4ww"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "2. **For better accuracy**\n",
        "\n",
        "* Use large `max_bin` (may be slower).\n",
        "* Use small `learning_rate` with `large num_iterations`\n",
        "* Use large `num_leaves`(may cause over-fitting)\n",
        "* Try to use categorical feature directly."
      ],
      "metadata": {
        "id": "X4V_yKNBS_CI"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "3. **To deal with over-fitting**\n",
        "\n",
        "* Use `min_data_in_leaf` and `min_sum_hessian_in_leaf`\n",
        "* Try `lambda_l1`, `lambda_l2` and `min_gain_to_split` to regularization\n",
        "* Try `max_depth` to avoid growing deep tree\n",
        "* Try `dart`"
      ],
      "metadata": {
        "id": "SFw9dI43TLbA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Check https://lightgbm.readthedocs.io/en/latest/Parameters-Tuning.html for hyperparamter tuning."
      ],
      "metadata": {
        "id": "OnbuqeIvTXP5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### For categorical variable and missing value"
      ],
      "metadata": {
        "id": "V7LmHQzR7RtF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "LightGBM enables the missing value handle by default. See https://lightgbm.readthedocs.io/en/latest/Advanced-Topics.html#missing-value-handle. It also deal with categorical variables as described here https://lightgbm.readthedocs.io/en/latest/Advanced-Topics.html#categorical-feature-support"
      ],
      "metadata": {
        "id": "HTaO1Nx47YZd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Select target\n",
        "data = pd.read_csv('melb_data.csv')\n",
        "y = data.Price\n",
        "\n",
        "# To keep things simple, we'll split the columns into numerical can categorical features\n",
        "melb_predictors = data.drop(['Price', 'Date', 'Address'], axis=1)\n",
        "cat_col = melb_predictors.select_dtypes(exclude=['int64','float64'])\n",
        "\n",
        "# Divide data into training and validation subsets\n",
        "X, X_v, y_train, y_valid = train_test_split(melb_predictors, y, train_size=0.8, test_size=0.2, random_state=0)\n",
        "X_train = X.select_dtypes(exclude=['object'])\n",
        "X_valid = X_v.select_dtypes(exclude=['object'])\n",
        "X_train_cat = X.select_dtypes(exclude=['int64','float64'])\n",
        "X_valid_cat = X_v.select_dtypes(exclude=['int64','float64'])"
      ],
      "metadata": {
        "id": "69qkUi3qTWxS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for col in X_train_cat.columns:\n",
        "  X[col] = X[col].astype('category')\n",
        "  X_v[col] = X_v[col].astype('category')"
      ],
      "metadata": {
        "id": "N3DIZVNS9d4d"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lgbr = lgb.LGBMRegressor(boosting_type='gbdt', random_state=42, n_jobs=-1)\n",
        "lgbr.fit(X, y_train)"
      ],
      "metadata": {
        "id": "jiF3wMm7SHRP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8a07db4b-b48b-401f-dd98-3cfaf98f9584"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMRegressor(random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 158
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "preds = lgbr.predict(X_v)"
      ],
      "metadata": {
        "id": "brGm9FFR8eIf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wpS4eGVe9ujW",
        "outputId": "fd13a1a8-e48c-4811-f918-d5db5fd39463"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "160267.37406568974"
            ]
          },
          "metadata": {},
          "execution_count": 160
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### SHAP values"
      ],
      "metadata": {
        "id": "RQpYAqIKkEM6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "explainer = shap.Explainer(lgbr)\n",
        "shap_values = explainer.shap_values(X)"
      ],
      "metadata": {
        "id": "RrbIM2sUg5_B"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "shap.summary_plot(shap_values, X)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "Oeql_GBTiMwh",
        "outputId": "08e906f1-cf14-4d3e-8eb5-997e33285888"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x626.4 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## CatBoost"
      ],
      "metadata": {
        "id": "tMrt7a0u-wAG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this section, we would explore some base cases of using catboost, such as model training, cross-validation and predicting"
      ],
      "metadata": {
        "id": "UpYxeNoyM7S9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Classification task"
      ],
      "metadata": {
        "id": "-LL0QrhgM_i0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "iris = datasets.load_iris()\n",
        "df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],columns= iris['feature_names'] + ['target'])\n",
        "X_train, X_test, y_train, y_test = train_test_split(iris['data'], iris['target'], random_state=42)"
      ],
      "metadata": {
        "id": "a1UBXodT9yvW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf = CatBoostClassifier(boosting_type='Plain', learning_rate=0.1, n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent')\n",
        "clf.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PojeH09ENdbi",
        "outputId": "181716a5-6e96-4907-fd27-adb98f7e1b36"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<catboost.core.CatBoostClassifier at 0x7fc859a9b710>"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clf.fit(X_train, y_train)\n",
        "y_pred = clf.predict(X_test)\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "print('Score: ' + str(score))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yM5MyOi1P9Fd",
        "outputId": "6a64a378-6ea1-4e93-e784-2e18131690c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 1.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clf.get_all_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "B1jqoOOXOpVL",
        "outputId": "67ad9981-1813-4dbf-9410-2b64f0bdd01c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'auto_class_weights': 'None',\n",
              " 'bagging_temperature': 1,\n",
              " 'bayesian_matrix_reg': 0.10000000149011612,\n",
              " 'best_model_min_trees': 1,\n",
              " 'boost_from_average': False,\n",
              " 'boosting_type': 'Plain',\n",
              " 'bootstrap_type': 'Bayesian',\n",
              " 'border_count': 254,\n",
              " 'class_names': [0, 1, 2],\n",
              " 'classes_count': 0,\n",
              " 'depth': 6,\n",
              " 'eval_metric': 'MultiClass',\n",
              " 'feature_border_type': 'GreedyLogSum',\n",
              " 'force_unit_auto_pair_weights': False,\n",
              " 'grow_policy': 'SymmetricTree',\n",
              " 'iterations': 100,\n",
              " 'l2_leaf_reg': 3,\n",
              " 'leaf_estimation_backtracking': 'AnyImprovement',\n",
              " 'leaf_estimation_iterations': 1,\n",
              " 'leaf_estimation_method': 'Newton',\n",
              " 'learning_rate': 0.10000000149011612,\n",
              " 'loss_function': 'MultiClass',\n",
              " 'max_leaves': 64,\n",
              " 'min_data_in_leaf': 1,\n",
              " 'model_shrink_mode': 'Constant',\n",
              " 'model_shrink_rate': 0,\n",
              " 'model_size_reg': 0.5,\n",
              " 'nan_mode': 'Min',\n",
              " 'penalties_coefficient': 1,\n",
              " 'pool_metainfo_options': {'tags': {}},\n",
              " 'posterior_sampling': False,\n",
              " 'random_seed': 42,\n",
              " 'random_strength': 1,\n",
              " 'rsm': 1,\n",
              " 'sampling_frequency': 'PerTree',\n",
              " 'score_function': 'Cosine',\n",
              " 'sparse_features_conflict_fraction': 0,\n",
              " 'task_type': 'CPU',\n",
              " 'use_best_model': False}"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Regression"
      ],
      "metadata": {
        "id": "Gm_vj7HGQYn1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X,y = datasets.load_diabetes(return_X_y=True)\n",
        "\n",
        "catb = CatBoostRegressor(boosting_type='Plain', learning_rate=0.1, random_state=42, n_estimators=100, thread_count=-1, logging_level = 'Silent')\n",
        "\n",
        "scores = cross_val_score(catb, X, y, scoring='neg_mean_squared_error', cv=5)\n",
        "\n",
        "# Take square root of the scores\n",
        "rmse = np.sqrt(-scores)\n",
        "\n",
        "# Display accuracy\n",
        "print('RMSE:', np.round(rmse, 3))\n",
        "\n",
        "# Display mean score\n",
        "print('RMSE mean: %0.3f' % (rmse.mean()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n0pWxCpTPS1I",
        "outputId": "2af613d0-0a89-49e1-ef39-24156a9e18f9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "RMSE: [54.744 56.125 59.595 56.908 56.583]\n",
            "RMSE mean: 56.791\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "catb.fit(X,y)\n",
        "catb.get_all_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RtQ7lRh9Q9v1",
        "outputId": "29888ec1-eb04-48b3-840e-b875f7f2cc38"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'auto_class_weights': 'None',\n",
              " 'bayesian_matrix_reg': 0.10000000149011612,\n",
              " 'best_model_min_trees': 1,\n",
              " 'boost_from_average': True,\n",
              " 'boosting_type': 'Plain',\n",
              " 'bootstrap_type': 'MVS',\n",
              " 'border_count': 254,\n",
              " 'classes_count': 0,\n",
              " 'depth': 6,\n",
              " 'eval_metric': 'RMSE',\n",
              " 'feature_border_type': 'GreedyLogSum',\n",
              " 'force_unit_auto_pair_weights': False,\n",
              " 'grow_policy': 'SymmetricTree',\n",
              " 'iterations': 100,\n",
              " 'l2_leaf_reg': 3,\n",
              " 'leaf_estimation_backtracking': 'AnyImprovement',\n",
              " 'leaf_estimation_iterations': 1,\n",
              " 'leaf_estimation_method': 'Newton',\n",
              " 'learning_rate': 0.10000000149011612,\n",
              " 'loss_function': 'RMSE',\n",
              " 'max_leaves': 64,\n",
              " 'min_data_in_leaf': 1,\n",
              " 'model_shrink_mode': 'Constant',\n",
              " 'model_shrink_rate': 0,\n",
              " 'model_size_reg': 0.5,\n",
              " 'nan_mode': 'Min',\n",
              " 'penalties_coefficient': 1,\n",
              " 'pool_metainfo_options': {'tags': {}},\n",
              " 'posterior_sampling': False,\n",
              " 'random_seed': 0,\n",
              " 'random_strength': 1,\n",
              " 'rsm': 1,\n",
              " 'sampling_frequency': 'PerTree',\n",
              " 'score_function': 'Cosine',\n",
              " 'sparse_features_conflict_fraction': 0,\n",
              " 'subsample': 0.800000011920929,\n",
              " 'task_type': 'CPU',\n",
              " 'use_best_model': False}"
            ]
          },
          "metadata": {},
          "execution_count": 47
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Speed"
      ],
      "metadata": {
        "id": "tIPXVGF0RlOn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('exoplanets.csv')\n",
        "# Split data into X and y\n",
        "X = df.iloc[:,1:]\n",
        "y = df.iloc[:,0]\n",
        "\n",
        "# Split data into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=2)"
      ],
      "metadata": {
        "id": "wInbb9ASREGF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "start = time.time()\n",
        "\n",
        "# Instantiate the XGBRegressor, xg_reg\n",
        "ca_reg = CatBoostClassifier(n_estimators=100, max_depth=2, random_state=42, logging_level = 'Silent')\n",
        "\n",
        "# Fit xg_reg to training set\n",
        "ca_reg.fit(X_train, y_train)\n",
        "\n",
        "# Predict labels of test set, y_pred\n",
        "y_pred = ca_reg.predict(X_test)\n",
        "\n",
        "score = accuracy_score(y_pred, y_test)\n",
        "\n",
        "print('Score: ' + str(score))\n",
        "\n",
        "end = time.time()\n",
        "elapsed = end - start\n",
        "\n",
        "print('Run Time: ' + str(elapsed) + ' seconds')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sUqahaIQRrvQ",
        "outputId": "01bc5fa3-967b-429e-dee9-d217ab87aaaa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 0.9913522012578616\n",
            "Run Time: 15.904590129852295 seconds\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Hyperparameter"
      ],
      "metadata": {
        "id": "ELzRE-rLSEgm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can check more details at https://catboost.ai/en/docs/references/training-parameters/. There are also tutorials about hyperparameter tuning https://github.com/catboost/tutorials/blob/master/hyperparameters_tuning/hyperparameters_tuning_using_optuna_and_hyperopt.ipynb"
      ],
      "metadata": {
        "id": "QPfUwGMySJW-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### For categorical variable and missing value"
      ],
      "metadata": {
        "id": "Eh9WywLIS2L2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "One of the differences between CatBoost and other gradient boosting libraries is its advanced processing of the categorical features (in fact \"Cat\" in the package name stands for \"CATegorical\").\n",
        "\n",
        "CatBoost deals with the categorical data quite well out-of-the-box. However, it also has a huge number of training parameters, which provide fine control over the categorical features preprocessing."
      ],
      "metadata": {
        "id": "2ms2jQBMS7Pm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The amount of parameters related to categorical features processing in CatBoost is overwhelming. Here is a hopefully the full list:\n",
        "\n",
        "* `one_hot_max_size` (int) - use one-hot encoding for all categorical features with a number of different values less than or equal to the given parameter value. No complex encoding is performed for such features. \n",
        "\n",
        "* `model_size_reg` (float from 0 to inf) - The model size regularization coefficient. The larger the value, the smaller the model size. This regularization is needed only for models with categorical features (other models are small). Models with categorical features might weight tens of gigabytes or more if categorical features have a lot of values. If the value of the regularizer differs from zero, then the usage of categorical features or feature combinations with a lot of values has a penalty, so fewer of them are used in the resulting model. Default value is 0.5\n",
        "\n",
        "* `max_ctr_complexity` - The maximum number of features that can be combined. Each resulting combination consists of one or more categorical features and can optionally contain binary features in the following form: “numeric feature > value”. For regression task on CPU the default value is 4.\n",
        "\n",
        "* `has_time (bool)` - if true, the 1-st step of categorical features processing, permutation, is not performed. Useful when the objects in your dataset are ordered by time. For our dataset, we don't need it. Default value is False\n",
        "\n",
        "* `simple_ctr` - Quantization settings for simple categorical features.\n",
        "combinations_ctr - Quantization settings for combinations of categorical features.\n",
        "\n",
        "* `per_feature_ctr` - Per-feature quantization settings for categorical features.\n",
        "\n",
        "* `counter_calc_method` determines whether to use validation dataset (provided through parameter eval_set of fit method) to estimate categories frequencies with Counter. By default, it is Full and the objects from validation dataset are used; Pass SkipTest value to ignore the objects from the validation set\n",
        "ctr_target_border_count - The maximum number of borders to use in target quantization for categorical features that need it. Default for regression task is 1.\n",
        "\n",
        "* `ctr_leaf_count_limit` - The maximum number of leaves with categorical features. Default value is None i.e. no limit.\n",
        "\n",
        "* `store_all_simple_ctr`- If the previous parameter ctr_leaf_count_limit at some point gradient boosting tree can no longer make splits by categorical features. With Default value False the limitation applies both to original categorical features and the features, that CatBoost creates by combining different features. If this parameter is set to True only the number of splits made on combination features is limited.\n",
        "\n",
        "The three parameters `simple_ctr`, `combinations_ctr`, and `per_feature_ctr` are complex parameters that control the second and the third steps of categorical features processing."
      ],
      "metadata": {
        "id": "o6zMME3hZvbw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Select target\n",
        "data = pd.read_csv('melb_data.csv')\n",
        "y = data.Price\n",
        "\n",
        "# To keep things simple, we'll split the columns into numerical can categorical features\n",
        "melb_predictors = data.drop(['Price', 'Date', 'Address'], axis=1)\n",
        "cat_col = melb_predictors.select_dtypes(exclude=['int64','float64'])\n",
        "\n",
        "# Divide data into training and validation subsets\n",
        "X, X_v, y_train, y_valid = train_test_split(melb_predictors, y, train_size=0.8, test_size=0.2, random_state=0)\n",
        "X_train = X.select_dtypes(exclude=['object'])\n",
        "X_valid = X_v.select_dtypes(exclude=['object'])\n",
        "X_train_cat = X.select_dtypes(exclude=['int64','float64'])\n",
        "X_valid_cat = X_v.select_dtypes(exclude=['int64','float64'])"
      ],
      "metadata": {
        "id": "RjqW39jGSEAG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "categorical_features_names = list(X_train_cat.columns)\n",
        "\n",
        "for col in categorical_features_names:\n",
        "  X[col] = X.loc[:,col].fillna(value='nan')\n",
        "  X_v[col] = X_v.loc[:,col].fillna(value='nan')\n",
        "  X[col] = X[col].astype('category')\n",
        "  X_v[col] = X_v[col].astype('category')"
      ],
      "metadata": {
        "id": "WJHEOWfJUTf4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names)\n",
        "catbr.fit(X, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wlDd6-DsR2_4",
        "outputId": "c3ed784d-3c64-45e0-f238-ce87e6dd38eb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<catboost.core.CatBoostRegressor at 0x7fc843eb8c10>"
            ]
          },
          "metadata": {},
          "execution_count": 170
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "preds = catbr.predict(X_v)"
      ],
      "metadata": {
        "id": "U0d4xfjRVhy6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-x8Pex8GVjZT",
        "outputId": "0d1494ea-bb33-4049-e9d4-a3a2ff071259"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "164721.46472522244"
            ]
          },
          "metadata": {},
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The first thing we try is to make CatBoost use one-hot encoding for all our categorical features. The documentation says, that for the features for which one-hot encoding is used no other encodings are computed."
      ],
      "metadata": {
        "id": "jBmQWDoLYI6g"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, one_hot_max_size=500)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4Zh03QOFXndQ",
        "outputId": "09002728-77a8-4cbf-a533-a113efa028b4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "163165.1377008556"
            ]
          },
          "metadata": {},
          "execution_count": 112
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let us try to set model size regularization coefficient to 0 - thus we allow our model to use as many categorical features and its combinations as it wants."
      ],
      "metadata": {
        "id": "N9S8KIKCa_ev"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, model_size_reg=0)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zF2EsTaUZct7",
        "outputId": "546c2025-155a-443b-c088-6b673e173312"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "165007.46639148306"
            ]
          },
          "metadata": {},
          "execution_count": 113
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, model_size_reg=1)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Evt8PjPXav1V",
        "outputId": "8110d5ee-b0a3-4ae8-aa13-050e6c13bc36"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "160827.54657606702"
            ]
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Note that any combination of several categorical features could be considered as a new one. Although it is not mentioned in the documentation, this parameter value has to be smaller than 15."
      ],
      "metadata": {
        "id": "gMYfIC-XbIkr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, max_ctr_complexity=6)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hn4vYk5Ja4vv",
        "outputId": "d9780de7-cc42-4787-d583-b32e0d44b403"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "163529.3517553931"
            ]
          },
          "metadata": {},
          "execution_count": 115
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, max_ctr_complexity=0)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gDpLpL_ybXjn",
        "outputId": "b7a59525-0e3b-4b91-a912-411187943ba7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "164721.46472522244"
            ]
          },
          "metadata": {},
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Counter method is very similar to the traditional Frequency Encoding"
      ],
      "metadata": {
        "id": "ygXiBf7Hcdpg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, simple_ctr='Counter', combinations_ctr='Counter')\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Wq3m4Bo4bv9F",
        "outputId": "8c667b12-5b9b-4f0a-8ea0-7bdb330842ce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "164993.47954656056"
            ]
          },
          "metadata": {},
          "execution_count": 118
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we proceed to the settings of the encodings methods that require target quantization. The first choice is `Borders` vs. Buckets"
      ],
      "metadata": {
        "id": "5O4sYVCFcztY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, simple_ctr='Borders', combinations_ctr='Borders')\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HiUGNJBzcj4T",
        "outputId": "02ce18b2-9d1d-4e8b-aa41-f9a4c3db0133"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "166920.94141120723"
            ]
          },
          "metadata": {},
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, simple_ctr='Buckets', combinations_ctr='Buckets')\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fKZCjPIcc8pe",
        "outputId": "888d596e-0853-422c-a75d-86273c49ae25"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "168168.9386958123"
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "It is quite common to use several encodings for a categorical feature. For instance, CatBoost creates 4 different encodings for each categorical feature by default. By default, CatBoost uses several encoding techniques to encode each categorical feature.\n",
        "\n",
        "* First it uses `Borders` method with one target border `TargetBorderCount=1` (in our example for each categorical feature we just want to see if it makes the car more expensive). The obtained float encodings are further discretized into `CtrBorderCount=15` different values. Three values of Prior parameter are used to create 3 three different encodings: `Prior=0/1:Prior=0.5/1:Prior=1/1`\n",
        "\n",
        "* Also for each categorical feature, we create an encoding with Counter method. The number of categorical encoding value borders `CtrBorderCount` is also equal to 15, and only one value of `Prior=0/1` is used."
      ],
      "metadata": {
        "id": "Rt8jFH_IdNL3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names, ctr_target_border_count=10)\n",
        "catbr.fit(X, y_train)\n",
        "preds = catbr.predict(X_v)\n",
        "mean_absolute_error(y_valid, preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1T-Q0uVTc_7G",
        "outputId": "a066252c-eb26-43fc-f125-6bb7b2bcd943"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "166281.68070428775"
            ]
          },
          "metadata": {},
          "execution_count": 171
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "catbr = CatBoostRegressor(n_estimators=100, random_state=42, thread_count=-1, logging_level = 'Silent', cat_features=categorical_features_names)\n",
        "catbr.fit(X, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SHRj34ZEdxzw",
        "outputId": "a1c40924-bb0e-4fc1-c580-a8d26afa60ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<catboost.core.CatBoostRegressor at 0x7fc843e49ed0>"
            ]
          },
          "metadata": {},
          "execution_count": 172
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### SHAP values"
      ],
      "metadata": {
        "id": "JbvYB_SrkHqJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "explainer = shap.Explainer(catbr)\n",
        "shap_values = explainer(X)"
      ],
      "metadata": {
        "id": "NNOn6R-dfQ5G"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "shap.summary_plot(shap_values, X)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "HymD7ehvfYav",
        "outputId": "4308601c-da6f-444a-e77e-b445ee2ba06e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x626.4 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "hZX-5gX4f8o2"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.10"
    },
    "nav_menu": {},
    "toc": {
      "navigate_menu": true,
      "number_sections": true,
      "sideBar": true,
      "threshold": 6,
      "toc_cell": false,
      "toc_section_display": "block",
      "toc_window_display": false
    },
    "colab": {
      "name": "14_ensemble.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}