Modelling_Normal_RE.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Modelling_Normal_RE.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "vvLQsSsVAsUA",
        "uLh2j2XvAxeO",
        "0y9PvYUNX-t2",
        "msPtDUHT0sQN",
        "GZqVnGrga8Tu",
        "xA9bSzUARRZ_",
        "LeV2q75ULzvm",
        "DXPMkhFs0nZp",
        "_EASzpf21O7L",
        "ARz9jBqc1O7c",
        "XGpFgTtG5vny",
        "LECmbx42-V2m"
      ],
      "toc_visible": true,
      "mount_file_id": "1rq84Um8lCPYGC_GfNTm02hG7VTom3oXh",
      "authorship_tag": "ABX9TyO1MPxWaIxklITxKx8ok5Lp",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ashishthomaschempolil/Santander-Customer-Satisfaction/blob/new_1/Modelling_Normal_RE.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QVQr_7USbczr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "sns.set(palette='muted',style='whitegrid')\n",
        "import sklearn\n",
        "from sklearn import metrics,ensemble,model_selection,linear_model,tree,calibration\n",
        "import xgboost as xgb\n",
        "import lightgbm as lgb\n",
        "import scipy\n",
        "import random\n",
        "from tqdm import tqdm\n",
        "import os\n",
        "np.random.seed(13154)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hRB4PxAYa5xu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 221
        },
        "outputId": "697de7be-df62-4270-faca-522d8f8bdbbb"
      },
      "source": [
        "file = [k for k in os.listdir('/content/drive/My Drive/Santander_Customer_Satisfaction') if (('train' in k)|('test' in k))&('X' not in k)] #getting all the necessary files for modelling\n",
        "file"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['train_Normal.pkl',\n",
              " 'test_Normal.pkl',\n",
              " 'train_Normal with Response encoding.pkl',\n",
              " 'test_Normal with Response encoding.pkl',\n",
              " 'train_Normal with One Hot Encoding.pkl',\n",
              " 'test_Normal with One Hot Encoding.pkl',\n",
              " 'train_Log Transformed.pkl',\n",
              " 'test_Log Transformed.pkl',\n",
              " 'train_Log Transformed with Response encoding.pkl',\n",
              " 'test_Log Transformed with Response encoding.pkl',\n",
              " 'train_Log Transformed with One Hot Encoding.pkl',\n",
              " 'test_Log Transformed with One Hot Encoding.pkl']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mV0hmG66S7ZC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "197fc266-0d86-40ee-f4c5-132a1bb14e9c"
      },
      "source": [
        "file_path = '/content/drive/My Drive/Santander_Customer_Satisfaction/'\n",
        "train_file = file_path+'train_Normal with Response encoding.pkl' #Normal with RE\n",
        "test_file = file_path+'test_Normal with Response encoding.pkl'\n",
        "dataset= \"Normal with Response Encoding\"\n",
        "train = pd.read_pickle(train_file)\n",
        "test = pd.read_pickle(test_file)\n",
        "X_train = train.drop(['ID','TARGET'],axis=1)\n",
        "y_train = train['TARGET'].values\n",
        "X_test = test.drop('ID',axis=1)\n",
        "del train,test\n",
        "X_train,X_val,y_train,y_val = model_selection.train_test_split(X_train,y_train,stratify=y_train,test_size=0.15)\n",
        "X_train.shape,X_val.shape,X_test.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((64617, 374), (11403, 374), (75818, 374))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v4yLi9O_CtFR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "global i\n",
        "i=0\n",
        "def plot_auc(y_true,y_pred,label,dataset=dataset):\n",
        "  \"\"\"\n",
        "  Plots auc curve when y_true and y_pred are given\n",
        "  dataset is string which tells which dataset were used.\n",
        "  label: tells which model was used\n",
        "  If label is a list plots all the auc curve for all labels (here y_pred will be a list of lists )\n",
        "  \"\"\"\n",
        "  \n",
        "  \n",
        "  if (type(label)!=list) & (type(label)!=np.array):\n",
        "    print(\"\\t\\t %s on %s dataset \\t\\t \\n\"%(label,dataset))\n",
        "    fpr,tpr,threshold = metrics.roc_curve(y_true,y_pred) #getting true positive rate, false positive rate\n",
        "    auc = metrics.roc_auc_score(y_true,y_pred) #getting auc score\n",
        "    log_loss = metrics.log_loss(y_true,y_pred) #getting logloss\n",
        "    label_1 = label+' AUC=%.3f'%(auc)\n",
        "\n",
        "    #plotting tpr vs fpr\n",
        "    sns.lineplot(fpr,tpr,label=label_1)\n",
        "    x=np.arange(0,1.1,0.1) #auc=0.5 line \n",
        "    sns.lineplot(x,x,label=\"AUC=0.5\")\n",
        "    plt.title(\"ROC on %s dataset\"%(dataset))\n",
        "    plt.xlabel('False Positive Rate')\n",
        "    plt.ylabel(\"True Positive Rate\")\n",
        "    plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.) #legend box outside graph\n",
        "    plt.show()\n",
        "    print(\"The log loss and auc was found to be %.3f and %.3f for %s on %s dataset.\"%(log_loss,auc,label,dataset))\n",
        "    #creating a dataframe of the result\n",
        "    result_dict = {\"Model\":label,'Dataset':dataset,'log_loss':log_loss,'auc':auc}\n",
        "    \n",
        "    return pd.DataFrame(result_dict,index=[i])\n",
        "\n",
        "  else:\n",
        "    plt.figure(figsize=(12,8))\n",
        "    for k,y in enumerate(y_pred):\n",
        "      fpr,tpr,threshold = metrics.roc_curve(y_true,y)\n",
        "      auc = metrics.roc_auc_score(y_true,y)\n",
        "      label_ = label[k]+' AUC=%.3f'%(auc)\n",
        "      #plotting tpr vs fpr\n",
        "      sns.lineplot(fpr,tpr,label=label_)\n",
        "\n",
        "    x=np.arange(0,1.1,0.1)\n",
        "    sns.lineplot(x,x,label=\"AUC=0.5\")\n",
        "    plt.title(\"Combined ROC\")\n",
        "    plt.xlabel('False Positive Rate')\n",
        "    plt.ylabel(\"True Positive Rate\")\n",
        "    plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.) #legend box outside graph\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "spCK7X7beIiy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def find_best_params(model,params,cv=10,n_jobs=-1,X_train=X_train):\n",
        "  \"\"\"\n",
        "  Funcion which implements Random Seacrh CV and returns best model\n",
        "  \"\"\"\n",
        "  random_cv = model_selection.RandomizedSearchCV(model,param_distributions=params,scoring='roc_auc',n_jobs=n_jobs,cv=cv,verbose=2)\n",
        "  random_cv.fit(X_train,y_train)\n",
        "  print(\"The best auc score was %.3f\"%(random_cv.best_score_))\n",
        "  print(\"The best params were: %s\"%(random_cv.best_params_))\n",
        "  return random_cv.best_estimator_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9Mho8KPmabxm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def plot_feature_importance(df,feat_imp,top=20,size=(6,6)):\n",
        "    \"\"\"\n",
        "    getting most important top features if df and feature_importance array is passed\n",
        "    \"\"\"\n",
        "    top_indices = np.argsort(feat_imp)[::-1][:top] #getting the indices with top feature importace\n",
        "    most_important_feat = df.columns[top_indices] #getiing the columns names\n",
        "    plt.figure(figsize=size)\n",
        "    sns.barplot(x=feat_imp[top_indices],y=most_important_feat) #plotting feature importance \n",
        "    plt.title('Feature Importance')\n",
        "    plt.xlabel('Importance')\n",
        "    plt.ylabel(\"Feature names\")\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VaQtOvkvfNeO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def get_top_features(df,feat_imp,top=50):\n",
        "  \"\"\"\n",
        "  gets the top features.\n",
        "  if top is of integer dtype it will return top features,\n",
        "  if top is between 0 and 1, it will return features that have at least feature importance value= top\n",
        "  \"\"\"\n",
        "  if (top>0.)&(top<=1.): #getting the features that have feature importance value greater than top\n",
        "    feature_to_consider = [1 if i>=top else 0 for i in feat_imp]\n",
        "    most_important_feat = [i  for i,j in zip(df.columns,feature_to_consider) if (j==1)]#getiing the columns names\n",
        "  else:\n",
        "    top_indices = np.argsort(feat_imp)[::-1][:top] #getting the indices with top feature importace\n",
        "    most_important_feat = df.columns[top_indices] #getiing the columns names\n",
        "  return most_important_feat    "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vvLQsSsVAsUA",
        "colab_type": "text"
      },
      "source": [
        "# Modeling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uLh2j2XvAxeO",
        "colab_type": "text"
      },
      "source": [
        "# Logistic Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "nsbNtvxAYB2O",
        "colab": {}
      },
      "source": [
        "model = linear_model.LogisticRegression(class_weight='balanced')\n",
        "params = {'penalty':['l2','l1'], 'C':[10.**i for i in np.arange(-3,3,1)], \n",
        "          'fit_intercept':[True,False] ,\n",
        "          }\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "mXTZa5oPYB2R",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "f47fdf92-bb28-475f-be7f-4a30abfa4cbd"
      },
      "source": [
        "%%time\n",
        "find_best_params(model,params)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 10 folds for each of 10 candidates, totalling 100 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:  1.2min\n",
            "[Parallel(n_jobs=-1)]: Done 100 out of 100 | elapsed:  4.0min finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "The best auc score was 0.811\n",
            "The best params were: {'penalty': 'l2', 'fit_intercept': True, 'C': 0.001}\n",
            "CPU times: user 12.7 s, sys: 1.24 s, total: 13.9 s\n",
            "Wall time: 4min 8s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=0.001, class_weight='balanced', dual=False,\n",
              "                   fit_intercept=True, intercept_scaling=1, l1_ratio=None,\n",
              "                   max_iter=100, multi_class='auto', n_jobs=None, penalty='l2',\n",
              "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
              "                   warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "eYp1oCU5YB2U",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "0c8c8f51-906b-4f00-fd0a-3541e0c2fc82"
      },
      "source": [
        "model = linear_model.LogisticRegression(C=0.001, class_weight='balanced', dual=False,\n",
        "                   fit_intercept=True, intercept_scaling=1, l1_ratio=None,\n",
        "                   max_iter=100, multi_class='auto', n_jobs=-1, penalty='l2',\n",
        "                  solver='lbfgs', tol=0.0001, verbose=0,\n",
        "                   warm_start=False)\n",
        "\n",
        "model.fit(X_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=0.001, class_weight='balanced', dual=False,\n",
              "                   fit_intercept=True, intercept_scaling=1, l1_ratio=None,\n",
              "                   max_iter=100, multi_class='auto', n_jobs=-1, penalty='l2',\n",
              "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
              "                   warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "DlDUHTI-YB2X",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "38086c28-c524-491a-c8e7-445e76491bae"
      },
      "source": [
        "labels=[] #storing all labels\n",
        "y_preds = [] #storing all y_pred\n",
        "y_pred = model.predict_proba(X_val)[:,1]\n",
        "label=\"Logisic Regression\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "result = plot_auc(y_val,y_pred,label,dataset)\n",
        "result_df = result\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t Logisic Regression on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.524 and 0.824 for Logisic Regression on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hh6_qnWsBOVa",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 80
        },
        "outputId": "da7b40c8-0585-4c62-e1c1-25b4726ad501"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                Model                        Dataset  log_loss       auc\n",
              "0  Logisic Regression  Normal with Response Encoding  0.523996  0.823515"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0y9PvYUNX-t2",
        "colab_type": "text"
      },
      "source": [
        "# Decision Trees"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0kVyDscneGv3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = tree.DecisionTreeClassifier(class_weight='balanced')\n",
        "params = { \n",
        "          'max_depth':[10,100,500,1000,-1], \n",
        "         'min_samples_split':[2,5,10,100,500], \n",
        "          'min_samples_leaf':[1,3,5,7,10],\n",
        "          'max_leaf_nodes':[100,500,1000,5000,-1] \n",
        "          }\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OWmU9ZdnjHEk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "5e8261c9-6509-45b5-a750-331252f17b20"
      },
      "source": [
        "%%time\n",
        "# find_best_params(model,params)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 3 µs, sys: 0 ns, total: 3 µs\n",
            "Wall time: 7.39 µs\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "11n9ZcTNirnF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "27037a5d-595d-41fd-d2e3-ae577f86ebae"
      },
      "source": [
        "model = tree.DecisionTreeClassifier(ccp_alpha=0.0, class_weight='balanced', criterion='gini',\n",
        "                       max_depth=10, max_features=None, max_leaf_nodes=5000,\n",
        "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
        "                       min_samples_leaf=5, min_samples_split=5,\n",
        "                       min_weight_fraction_leaf=0.0, presort='deprecated',\n",
        "                      splitter='best')\n",
        "\n",
        "model.fit(X_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DecisionTreeClassifier(ccp_alpha=0.0, class_weight='balanced', criterion='gini',\n",
              "                       max_depth=10, max_features=None, max_leaf_nodes=5000,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=5, min_samples_split=5,\n",
              "                       min_weight_fraction_leaf=0.0, presort='deprecated',\n",
              "                       random_state=None, splitter='best')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IqO7PiHUhFIR",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "22fce3f8-6a34-4b28-8ed8-89271e891de6"
      },
      "source": [
        "cc_model = calibration.CalibratedClassifierCV(model,cv='prefit')\n",
        "cc_model.fit(X_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CalibratedClassifierCV(base_estimator=DecisionTreeClassifier(ccp_alpha=0.0,\n",
              "                                                             class_weight='balanced',\n",
              "                                                             criterion='gini',\n",
              "                                                             max_depth=10,\n",
              "                                                             max_features=None,\n",
              "                                                             max_leaf_nodes=5000,\n",
              "                                                             min_impurity_decrease=0.0,\n",
              "                                                             min_impurity_split=None,\n",
              "                                                             min_samples_leaf=5,\n",
              "                                                             min_samples_split=5,\n",
              "                                                             min_weight_fraction_leaf=0.0,\n",
              "                                                             presort='deprecated',\n",
              "                                                             random_state=None,\n",
              "                                                             splitter='best'),\n",
              "                       cv='prefit', method='sigmoid')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CIxYiOfoih44",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "05684a6c-bfad-4a0e-88a5-811474b0116f"
      },
      "source": [
        "i+=1\n",
        "y_pred = cc_model.predict_proba(X_val)[:,1]\n",
        "label=\"Decision Tree Classifier\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "result = plot_auc(y_val,y_pred,label)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t Decision Tree Classifier on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.149 and 0.770 for Decision Tree Classifier on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qusVRO8HdpHW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "19c44146-6058-44bb-a28c-d1d90b0d97cf"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                      Model                        Dataset  log_loss       auc\n",
              "0        Logisic Regression  Normal with Response Encoding  0.523996  0.823515\n",
              "1  Decision Tree Classifier  Normal with Response Encoding  0.149110  0.769599"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "msPtDUHT0sQN"
      },
      "source": [
        "# Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "OwHc1zRG0sQP",
        "colab": {}
      },
      "source": [
        "model = ensemble.RandomForestClassifier(class_weight='balanced')\n",
        "params = { 'n_estimators':[1000,2000],\n",
        "          'max_depth':[500,1000], \n",
        "         'min_samples_split':[100,500],\n",
        "          'min_samples_leaf':[3,5],\n",
        "          'max_leaf_nodes':[100,500] \n",
        "          }\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "zDOPEuJy0sQS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "b114df07-a3c0-4c69-ee3e-8b76f5a346e2"
      },
      "source": [
        "%%time\n",
        "find_best_params(model,params,cv=3)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 10 candidates, totalling 30 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  30 out of  30 | elapsed: 75.3min finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "The best auc score was 0.820\n",
            "The best params were: {'n_estimators': 1000, 'min_samples_split': 500, 'min_samples_leaf': 3, 'max_leaf_nodes': 500, 'max_depth': 1000}\n",
            "CPU times: user 3min 21s, sys: 476 ms, total: 3min 22s\n",
            "Wall time: 1h 18min 40s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
              "                       criterion='gini', max_depth=1000, max_features='auto',\n",
              "                       max_leaf_nodes=500, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=3, min_samples_split=500,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=1000,\n",
              "                       n_jobs=None, oob_score=False, random_state=None,\n",
              "                       verbose=0, warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "RBeRKL7R0sQW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        },
        "outputId": "7365b368-04a5-44cf-94f7-2cee9982325a"
      },
      "source": [
        "model = ensemble.RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
        "                       criterion='gini', max_depth=1000, max_features='auto',\n",
        "                       max_leaf_nodes=500, max_samples=None,\n",
        "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
        "                       min_samples_leaf=3, min_samples_split=500,\n",
        "                       min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
        "                       n_jobs=-1, oob_score=False,\n",
        "                       verbose=0, warm_start=False)\n",
        "\n",
        "%time model.fit(X_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 11min 21s, sys: 2.3 s, total: 11min 24s\n",
            "Wall time: 5min 51s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
              "                       criterion='gini', max_depth=1000, max_features='auto',\n",
              "                       max_leaf_nodes=500, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=3, min_samples_split=500,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
              "                       n_jobs=-1, oob_score=False, random_state=None, verbose=0,\n",
              "                       warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "oJLF56Ts0sQY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "outputId": "8c7bd12a-4757-4bd0-e5bf-6c9730197fe7"
      },
      "source": [
        "cc_model = calibration.CalibratedClassifierCV(model,cv='prefit')\n",
        "cc_model.fit(X_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CalibratedClassifierCV(base_estimator=RandomForestClassifier(bootstrap=True,\n",
              "                                                             ccp_alpha=0.0,\n",
              "                                                             class_weight='balanced',\n",
              "                                                             criterion='gini',\n",
              "                                                             max_depth=1000,\n",
              "                                                             max_features='auto',\n",
              "                                                             max_leaf_nodes=500,\n",
              "                                                             max_samples=None,\n",
              "                                                             min_impurity_decrease=0.0,\n",
              "                                                             min_impurity_split=None,\n",
              "                                                             min_samples_leaf=3,\n",
              "                                                             min_samples_split=500,\n",
              "                                                             min_weight_fraction_leaf=0.0,\n",
              "                                                             n_estimators=2000,\n",
              "                                                             n_jobs=-1,\n",
              "                                                             oob_score=False,\n",
              "                                                             random_state=None,\n",
              "                                                             verbose=0,\n",
              "                                                             warm_start=False),\n",
              "                       cv='prefit', method='sigmoid')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "4Jct69qq0sQa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "51f5a522-ed96-420b-a0fb-50f10eff9b1f"
      },
      "source": [
        "i+=1\n",
        "y_pred = cc_model.predict_proba(X_val)[:,1]\n",
        "label=\"Random Forest Classifier\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "result = plot_auc(y_val,y_pred,label,dataset)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t Random Forest Classifier on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.137 and 0.836 for Random Forest Classifier on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "0igsmQs_0sQd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 142
        },
        "outputId": "062f59fc-6fa5-49fe-dbb4-a0f85f1371f7"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                      Model                        Dataset  log_loss       auc\n",
              "0        Logisic Regression  Normal with Response Encoding  0.523996  0.823515\n",
              "1  Decision Tree Classifier  Normal with Response Encoding  0.149110  0.769599\n",
              "2  Random Forest Classifier  Normal with Response Encoding  0.136829  0.836315"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GZqVnGrga8Tu",
        "colab_type": "text"
      },
      "source": [
        "# Selecting top features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tFB132azbBQS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 736
        },
        "outputId": "fee45316-56db-40f4-fea8-346ce448d79d"
      },
      "source": [
        "plot_feature_importance(X_train,model.feature_importances_,top=50,size=(7,12))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAq8AAALPCAYAAABMlSnHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeZxcVZ3//1dIIIAkEBFECRgE/EDokKVZouDCogSGZZwBZElMWCQIMwLK4k8RERRwQFEBB1kkYZdtAsMWth86KCgUJKQB34okAkIElAAR0500/f3jnDKXtqu3VHf18n4+HnlQdc+953xudQOfnDr3fIa0tLRgZmZmZtYfrFbrAMzMzMzMOsvJq5mZmZn1G05ezczMzKzfcPJqZmZmZv2Gk1czMzMz6zecvJqZmZlZv+Hk1czMzMz6jWG1DsDMzKwjEbEIeD/QXDj8EUkvrWKfR0q6b1Vi68J4pwNbSJraG+N1EMsM0r3vXOtYzLrKyauZmfUX+/RWotkZETFM0opax9FVEeH/91u/5l9gMzPrtyJiXeD7wF7AO8AVwDclNUfE5sClwHigBZgLHCtpSURcBWwK/G9ENANnAL8BrpY0utD/IvLsbJ45rQOWAfsCX46IGyuN34nYW4BjgROAjYAfALOAq/I4dwNTJTVFxKeAq4EfA18GlgJfl3RN4XO4ANgTeDvf91mS3smzrF/I9/d54EHgX4DVI2IpsELSehHxL8C3gc2BN4DLJZ2e+x8DLARmAGcCawPnS/pObh8KnAIcAWwI/A74V0kvRMRWObZ64FXgG5Ju6OjzMavEa17NzKw/mwWsALYAJgKfAY7MbUOAs4EPAlsDmwCnA0iaBjxPms1dR9J/dXK8/YCbgPWAazoYvzP2ICV1k4GTgUuAqTnWOuDgwrkbAe8DNgamA5dEROS2C4B1gQ8DnyQlqYcVrt0ReI609GIqcDTwcL739fI5f8vXrUdKbr8YEf/aKt6dgQB2A06LiK3z8S/nWPcCRgKHA29HxHuAe4FrSUntQcCPI2JsFz4js3fxzKuZmfUXcyKi/DX9g8BMUrK0nqS/A3+LiPOBo4CfSHoWeDaf/2pEfB/45irG8LCkOQARMbK98TvZ339JehN4KiIagHskPZf7v4uUEM8unP8NSY3AzyPiDuDAiDiLlBROkPQW8FZEfA+YBlyer3tJ0gX59YqVOe9Kkh4svH0yIq4jJcJzCse/le91fkTMJ81qP0NK2E+WpHze/HwPnwMWSboiH38iIm4GDgC+1cnPyOxdnLyamVl/8a/FNa8RsQOwOvByIRlbDXght78f+CHwcWBEbnt9FWN4ofD6Q+2N30l/Lrz+exvvNyq8f13S3wrv/0iaVX5fjuOPrdo2rhB3myJiR+Ac0ozvGsBw4MZWpy0uvH4bWCe/3gT4QxvdfgjYMSKWFI4NIy2NMOsWJ69mZtZfvQA0Au+r8ODUWaS1ruMk/TV/BX5hob2l1fl/I63lBP6xjnODVucUr+lo/GobFRHvKSSwmwINwGvAclKi+HSh7U+Fa1vfa+v3kL7avxDYU9KyiPgBKTHujBdIa2Ub2jj+c0mf7mQ/Zh1y8mpmZv2SpJcj4h7gexHxDdJDTJsBoyX9nDTb+gbwRkRsDJzUqos/k9aIlv0OWDM/uHQP8DXS7GN3x+8J34qIr5HWsO7NyofTbgC+ExGfB95LWoN6Xjv9/BkYHRFrSGrKx0YAf82J6w7AIaTPoTMuA86MiKdJSzXGkZLn24FzImIacH0+dwKwVNIznezb7F38wJaZmfVnnyd9xf00aUnATcAHctu3gEmkBPYO4JZW154NnBoRSyLiRElvAMeQErE/kWZiX1yF8attcR7jJdLDYkdL+m1u+09SvM8BD5FmUX/aTl8PAE8BiyPitXzsGOCMiHgLOA3oyo4A38/n3wO8SVpru1Zeg/sZ0prcl/I9fJd2/lJg1pEhLS1tfXNgZmZmfUV5q6ziNl5mg5VnXs3MzMys33DyamZmZmb9hpcNmJmZmVm/4ZlXMzMzM+s3vFWWmdVUqVQaDmwPvAx0WA/ezMwGvKGkXTsera+vb2zd6OTVzGpte+D/ah2EmZn1OR8nbf32Lk5ezazWXgb4yEc+whprrFHrWMzMbBW0NDczZOjQVeqjqamJ3/3ud5D//9Cak1czq7VmgKU33MmwpuW1jsXMzFbBBl+cWs3u2lxK5ge2zMzMzKzf8Myr2SAUEdcAu5AWxI+QtLTQ1gIsAN7Jh6ZJWtBOX1OBk4GxwPGSLuyxwM3MbNBz8mo2iETEMEkrSHXHTwD+XOHUjxUT2g7MI9Ut/2oVQjQzM2uXk1ezfiQiTgXWl3RCfr8+IGA6cCqwJunf6+9Iuj6f8yApwZwM/BXYS9IDuW2VY5LUkPt6p6NzzczMVpXXvJr1L1cCB0VE+S+ehwC3Ab8CdpY0EdgdOC8iRhWu+3Bu36uT4zwYEfMi4uyIGF6t4M3MzFaVk1ezfkTS88BTQDkJnQHMAjYAboqIBmAu8F6gOK16bV4u0BmbStoO+ARpHes3Vj1yMzOz6nDyatb/zAKmR8Q4YF3SBv//DTwIjJM0AXiRtISgrLPrV5H0Qv7nm8BlwE5VidrMzKwKnLya9T+3kGZFvwLMktQCrAcsktQSEZ8GtuhOxxExKiLWyq+HAfuT1suamZn1CU5ezfoZSW8DtwLTSGtgIT3pf15EzAMOBJ5sr4+IuCUiXix3GRFz8+utgF9HxPzcx3I6WDYQEQfnvg4AzoyIFyNibDduzczMrENDWlpaah2DmQ1ipVJpDLCwrq6O4cP9bJiZWX/WsqKZIcNWrTxsY2MjDQ0NAJvV19cvat3umVczsyoplUq1DqFqfC9900C5l4FyH+B7aW1VE9dOjeGZVzPrSETsBZzVRtPXJN25Kn175tXMrG9qWbGCIcN6vyRARzOvLlJgZh3KCeoqJakd+es11zGsqaknhzAzsy7Y4Oijah1Cm7xswMzMzMz6Dc+8mg1CEXENsAvwAWCEpKWFthZgAVAu9zpN0oJ2+lobuAKoB1YAJ0q6vadiNzOzwc0zr2aDSKGs7OXAhHZO/ZikCflPxcQ1OxF4U9IWwD7AZRGxThXCNTMz+yeeeTXrRyLiVGB9SSfk9+sDAqYDp5Kqag0DviPp+nzOg6RCA5OBvwJ7SXogt1UjrM/l8ZH0+4h4DNgTuLEanZuZmRV55tWsf7kSOKgwg3oIcBvwK2BnSROB3UkFC0YVrvtwbt+rk+M8GBHzIuLsiOhoC4BNgT8W3j8PbNLJcczMzLrEyatZPyLpeeApoJyEzgBmARsAN0VEAzAXeC9QnFa9VtKKTg6zqaTtSCVox9JBhS0zM7Pe5OTVrP+ZBUyPiHHAusD/Af8NPAiMkzQBeJG0hKBsKZ0k6YX8zzeBy4CdOrjkeeBDhfebAi90djwzM7OucPJq1v/cQpoV/QowS1ILsB6wSFJLRHwa2KI7HUfEqIhYK78eBuxPWi/bnhuBmfmaLYHtgbu7M76ZmVlHnLya9TOS3gZuBaaR1sACfJW0znUecCDwZHt9RMQtEfFiucuImJtfbwX8OiLm5z6W0/GygXOB9SLiWeB24ChJb3XxtszMzDrF5WHNrKZcHtbMrG/qq+VhPfNqZlYlpVKp1iFUje+lbxoo9zJQ7gMG9r3UInHtjL4ZlZn1KRGxF3BWG01fk3Rnb8fTV9XX19c6hKrxvfRNre+lVjNjZrXk33gz61BOUHs0SX3t6ssY1rSsJ4cwG3De/8Wv1DoEs17nZQNmZmZm1m945tWsD4iI04F1JJ3YRtsMYG9J+/d2XIUYPgD8L+m/GUOB35J2FXg9t+9D2nVgGFACDsu7IpiZmVWVZ17NrF15v9fXgE9ImiBpHKkIwjdy+zrApcA+krYA3gL+KQk3MzOrBs+8mvWAiFgbmA1sQ9orVcCXgOuAkaTqV3dIOrmNa9cALgB2JSWNTxTahgLfBabkQ3cDp0hqrhDHZcACST/M7+uA24DNgYOB44A18uknSro/n7cIuD7HsEDSEfk+yjGsA7yRr9sTeEzS7/P7i/O9n9HhB2VmZtZFnnk16xl7ACMljZU0nlSBaglpdrIemABsFxFT2rh2JrAZMBbYDdih0HZUvnZS/jMxH6tkFjC98P4wVlblmgtMljQROIiUcBaNlLRDTlwByEUQXgW2ZGVyuinwx8J1zwObtBOTmZlZtzl5NesZ84GtI+KiiDgAaCStFT03V68qAXWkRLS1XYDZkpbndaNXF9p2JyWfTZKagCvysTZJeggYERHj8tf/B7MySd0cmBsRTwE/AzaKiI0Kl19JK5ImAO8HngGO7vBTMDMzqzInr2Y9QNJzpCUD95KSy/nAl4FRwI6StgXmkJYP9LTZwAzS1/vPSCrPkl4H/FjSNqRZ3BWt4lnaVmeSluc+p+VDzwMfKpyyKfBCtYI3MzMrcvJq1gMiYjTQLGkOcAKwAfBh4GVJyyJiY2C/Cpc/AEyLiGERsRZwSKHtPmB6RKweEauTlgTc20E4V5JmXI8kzdSWrQcszK8PByrWZo2ITfKDWUTEasC/Awty893A9hGxZX5/NHBDBzGZmZl1i5NXs54xDng4LxH4DXA2cDqwU0Q0AJcD91e49hLSbOYzpET20VZtT5Ie4noiv760vUAkPQ88DXwKuKXQdDwwJyIeJyXWf2mnmwB+GRFP5jE/QHoADUlvkdbd3h4RzwLrAue1F5OZmVl3DWlpaal1DGY2iJVKpTHAwrq6OoYPrzj5a2Zt6K/lYUul0oAp2+t7qb7GxkYaGhoANquvr1/Uut0zr2ZmVVIqlWodQtX4Xvqm1vfSHxNXs1Xl33qzASAiLgYmtzq8QtJ2tYhnsOoLMxbV0t/vpWXFcoYMW73WYZhZD3DyajYASOr321a9ctW5DGtyRVmrjg8cc1atQzCzHuJlA2ZmZmbWb3jm1cxWWS4Z+yNS2doW4BxJl9U2KjMzG4g882pm1XAosAWpbOxHgdMjYkxNIzIzswHJM69mA1xEtABnkIoirAV8TdLNue2jwLnAiHz6SZLuiYjzgE8CawCvAYcXKnO15XPApZLeAV6NiDnAAblvMzOzqvHMq9ng0CxpArAvcElEbBgR7wX+BzhZ0nhSidhyQYRzJG2fj18HfLeD/jcFisnt88AmVb0DMzMzPPNqNlhcDiBJuaLWZKAZeFrSr3JbM/B6Pn/PiDgWWAf/d8LMzPoQz7ya2btExIeA84GDJdUBhwNrdnDZ88CHCu83BV7omQjNzGwwc/JqNjgcBhARWwITgUeAh4Gxed0rETE0IkYBI4EmYHFErAZ0Zg/ZG4EvRMRqEbEB8K/ATdW/DTMzG+ycvJoNDsMi4gngdmCmpFck/RX4N+D7EfEkUALqJS0gJaNPA78GFnai/6uA54DfkxLjMyR15jozM7Mu8Vo2s8HhPEmntz6Y17t+tI3jxwHHFQ59s73O83rZL65KgBtOO4nhw4evShdm/+DysGYDl2dezcyqpFQq1TqEqunv9+LE1Wzg8syr2QAnaUg1+omICcCsNpoudDWtpL6+vtYhVE1fvZeWFU0MGbZGrcMwsxpy8mpmnSJpHjChp/p/efbxDG18s6e6twFik/+8ptYhmFmNedmAmZmZmfUbnnm1PikiTgfWkXRiG20zgL0l7d/bcVVSjDcijgbWknR+DeMZAzwLNBQO7ybpL93srwUYIWlpRBwPXCvpldw2hVSBqwVYHZgDnCqpZRVuwczMrE1OXs2qTNLFtRw/Isr/Xi/JJWGr7XjgPuCV/P4hYJKk5ohYHfglaYut23pgbDMzG+ScvFqviIi1gdnANsByQMCXgOtIm+KvCdwh6eQ2rl0DuADYFXgNeKLQNpQ06zclH7obOCVv3dRWHGOAx4BL8zVrAYeSNuLfEfg7sJ+kxfn8U4B/J/278ifgC5IWR8S6pJKrdcBiUjWpP+drTmflLGxX47sMWCDph/l9HSkJ3Bw4mLR9VflplRMl3Z/PWwRcnz+jBcCZbfXfnuLsalvv87GvAx8EboqIZcAhkp4udLNmju+dro5vZmbWGV7zar1lD2CkpLGSxgMzgSXAPpLqSQ8CbZe/gm5tJrAZMBbYDdih0HZUvnZS/jMxH2vP+sBDkiaSEtD7gYskbUvaqP8/ACJiKilpnCxpEnAn8L3cx2nAm5K2AvYHPllhrK7GNwuYXnh/GDArfwU/N8cyETiI9JeBopGSdpB0RPl9RDwWEaWIOCkiVnnXAUnfAV4C9pc0oZy4RsR2udDBK8ADwB2rOpaZmVlbnLxab5kPbB0RF0XEAUAjMBQ4NyLmk5LGOtp+mn0XYLak5ZLeBq4utO1OSu6aJDUBV+Rj7VkqqZxcPQ68mJ+kJ8exRX69b+7r8YiYBxwLjCnEdDmApNeAWyqM1aX4JD0EjIiIcfnr/4NZmaRuDsyNiKeAnwEbRcRGhcuvLLx+GRgtaTtgT9Ls8RH0EEmP5eR/E6Ae+HhPjWVmZoObk1frFZKeIy0ZuJeUvM0HvgyMAnbMic8c0tfOPa2x8LoZWNbqfXk5zRDg23mGcYKkOkk79UJ8s4EZpKTzGUl/zMevA34saRvSLO4K3v15/ePrfUmN5Qeq8j+vATqKvZn834SI6NbPISfydwEHdOd6MzOzjjh5tV4REaOBZklzgBOADYAPAy9LWhYRGwP7Vbj8AWBaRAyLiLWAQwpt9wHTI2L1/LDQdFKCXA23AcdExKh8D8MjYnwhpsPy8fWBz1boozvxXUmacT2SNFNbth6wML8+HKhYSzUiNszjldcb7wvMq3R+9iywfX59SDvnvQmsWxjrIxFRTnrfQ0q6F3QwlpmZWbf4gS3rLeOAcyIC0nKBs0kPGN0YEQ3Ai6S1p225BNgWeIb0wNajwPsLbVuw8iGuuaSHsVaZpKsi4n3Az3PcqwE/Js0anwn8NCJ+S3pg6xftxN6l+CQ9HxFPA58iJbFlxwNzIuJ10oNf7W17tTNwRkQ0k7avuh24sL1xSTPhP4mIN4Ab2jnvR8AVEfE2Kcn9F2BGRKwg/Wz/B3DFLTMz6xFDWlq8FaOZ1U6pVBoDLKyrq2P48IqTyWZA98rDlkqlPlvutqsGyr0MlPsA30tPaGxspKGhAWCz+vr6Ra3bvWzAzKxKSqVSrUOomr56L11NXM1s4PGyARuQIuJiYHKrwyvy0/c1V6v4IuJI8lZgrcwo7LhgZmbWZzl5tQFJ0tG1jqE9tYpP0mV4PWqP6Qtft1VLX7mXd1Y0sZpnW82swMmrmfUJf7jqC6zW+Eatw7A+Zqtjb611CGbWxzh5tV5RLJnaRtsMYG9J+/d2XJW0KvF6NLCWpPNrGM8Y0lZWDYXDu0mquONALhm7t6SGSudUKbaPAD8BPkDae/ZR4BhJf+/Jcc3MbHBy8mrWAUkX13L8XGkLYImktiqQ1VoT8GVJT+T9Xq8DTiRtJ2ZmZlZVTl6tW/LG97NJVbOWAwK+REpcRpIqP90h6eQ2rl0DuADYlbRv6xOFtqHAd4Ep+dDdwCmSmivEMQZ4jLR36hRgLeBQ4GhgR+DvwH6SFufzTyGVSh0G/An4gqTFEbEuqdxrHWnf1heAP+drTmflLGxX47sMWCDph/l9Han4weakPVyPA8oL+k6UdH8+bxFpH9xdSRv+dzcRPDAiLiXNip4n6cLc/3nAJ/PYrwGHlyt5RcTewOmk/WHfAaZLejIidgTOIf18AU6TdIekRcAiAEnvRMRvgK27Ga+ZmVm7vFWWddcewEhJYyWNB2YCS4B9JNUDE4DtImJKG9fOBDYDxgK7ATsU2o7K107KfybmY+1ZH3hI0kRSAno/cFEuOVsiP10fEVNJSeNkSZOAO4Hv5T5OA96UtBWwPymxa0tX45tFqqpVdhgwS1ILqWDB5Bz3QaS/DBSNlLSDpCPK7yPisYgoRcRJETGknXHL1pb0UVLBg3MiYp18/BxJ2+ef3XWkhLy8BOAy4ODcNhlYGBHrARcDh+Sf796kggbrFQfLFdAOJyXoZmZmVefk1bprPrB1RFwUEQcAjaTqSudGxHxS0lhHSvRa2wWYLWm5pLeBqwttu5OSuyZJTaTyqLt3EMtSSXfk148DLxa2fSqRKlxBKpG6O/B4RMwDjgXGFGK6HEDSa8AtFcbqUnySHgJGRMS4/PX/waxMUjcH5kbEU8DPgI0iYqPC5VcWXr8MjM5bae1Jmj0+go5dn+NYBLwOjM7H94yIR3J1sxNZ+XP6NHCnpN/n6xolvQV8jPQXjrvyZ3cX0MLKz7a8vOF64AFJTl7NzKxHeNmAdYuk5yJiG9LM6Z7AWaRkaxSwo6RlEXEJaflAT2ssvG4GlrV6X/49HwJ8W9JPeyGmotnADOBB4Jny1/OkGc+vSJqT14q+zbs/r6XlF5IagVfy61ci4hpgJzre9uqfPouI+BBwPrC9pIUR8THg2g76GQI8KekTbTXm5RTXkBLkL3XQl5mZWbd55tW6JSJGA82S5gAnABsAHwZezonrxsB+FS5/AJgWEcPy18yHFNruA6ZHxOoRsTrpK/d7qxT2bcAxETEq38PwiBhfiOmwfHx94LMV+uhOfFeSZlyPJM3Ulq0HLMyvDwcq1kaNiA3zeOX1xvsC3S0qMJL0kNXinDQX95y9B9grIrbMYw2PiBHAr4AtI2KXQkzbR8SQ3McsUnJ8RF4SYWZm1iOcvFp3jQMezksEfgOcTXrIZ6f8VXR57WlbLgGeB54hJY2Ptmp7kvQQ1xP59aXVCFjSVaTZwZ9HxJOkJQU75eYzgVER8VvgZuAX7cTepfgkPQ88TVp3WlyOcDwwJyIeJyX+Fbe9AnYGnsif92N57AvbG7edeBYAN+aYfs3KBJq8XOALwM/yWA8DYyS9TkqYvxkR8yPiGdLPewhp5n0q6XeiFBHzIuKi7sRmZmbWkSEtLZ4kMbPaKZVKY4CFdXV1DB9ecfLZBqlqVNgqlUp9pmLYqhoo9zJQ7gN8Lz2hsbGRhoYGgM3q6+sXtW73zKuZWZWUSqVah1A1feVeXBrWzFrzA1vWL0TExaRtm4pW5Kfva65W8UXEkeStwFqZUdhxwczMbMBw8mr9gqSjOz6rdmoVn6TL6HjHAeslfeHrtmqpxb1UY4mAmQ18Tl7NrE9ouOZIhjQtqXUYVkOTjv7fWodgZv2Ak1czW2UR8RnSXr/jgAsknVjjkMzMbIBy8mpm1fAcaR/b/emdwhRmZjZIOXk1G+AiogU4g1Q0Yi3ga5Juzm0fBc4FRuTTT5J0T0ScB3wSWAN4DTi8UBnsn0h6Nvf3rz12I2ZmZnirLLPBolnSBFKhgUtyxa73Av8DnCxpPDCJlQUjzpG0fT5+HfDdmkRtZmbWimdezQaHywEkKVf0mkwq5/q0pF/ltmbg9Xz+nhFxLLAO/u+EmZn1IZ55NbN3iYgPAecDB0uqAw7H61jNzKyPcPJqNjgcBhARWwITgUeAh4Gxed0rETE0IkYBI4EmYHFErAb06T12zcxscHHyajY4DIuIJ4DbgZmSXpH0V+DfgO9HxJNACaiXtAC4EXga+DWwsKPOI2LniHgR+DIwMyJejIg9eupmzMxs8PJaNrPB4TxJp7c+mNe7frSN48cBxxUOfbO9ziU9BIxelQDrDr2M4cOHr0oX1s+5wpaZdYZnXs3MqqRUKtU6hKqpxb04cTWzzvDMq9kAJ2lINfqJiAnArDaaLpR0WTXGMDMz64iTVzPrFEnzgAm1jqMvq6+vr3UIVdOb9+LlAmbWFU5ezaxPKF13BC1NS2odhtXATkfdXusQzKwfcfJq1sdFxOnAOpJObKNtBrC3pP17O65CDGOAZ4GGwuHdJP2lNhGZmdlA5uTVzLotIsr/DVmSy8+amZn1KCevZr0sItYGZgPbAMsBAV8CriMVCFgTuEPSyW1cuwZwAbAr8BrwRKFtKPBdYEo+dDdwSi772lYclwELJP0wv68DbgM2Bw4mbZVVXoh4oqT783mLgOtzDAuAM7v+KZiZmXWPt8oy6317ACMljZU0HpgJLAH2kVRPeihqu4iY0sa1M4HNgLHAbsAOhbaj8rWT8p+J+Vgls4DphfeHAbMktQBzgcmSJgIHkZLtopGSdpB0RPl9RDwWEaWIOCkiqrLDgZmZWWtOXs1633xg64i4KCIOABqBocC5ETGfVOmqjraf7N8FmC1puaS3gasLbbuTks8mSU3AFflYm3JhgRERMS5//X8wK5PUzYG5EfEU8DNgo4jYqHD5lYXXLwOjJW0H7An8O3AEZmZmPcDJq1kvk/QcacnAvaTkcj6prOooYEdJ2wJzSMsHetpsYAYp6XxG0h/z8euAH0vahjSLu6JVPEvLLyQ1Snolv34FuAbYqedDNzOzwcjJq1kvi4jRQLOkOcAJwAbAh4GXJS2LiI2B/Spc/gAwLSKGRcRawCGFtvuA6RGxekSsTloScG8H4VxJmnE9kjRTW7YesDC/PhyoWLc1IjbM45XX8+4LzOtgXDMzs27xA1tmvW8ccE5EQFoucDbpAagbI6IBeBG4v8K1lwDbAs+QHth6FHh/oW0LVj7ENRe4tL1AJD0fEU8DnyIlsWXHA3Mi4nXSg1/tbXu1M3BGRDQDqwO3Axe2N66ZmVl3DWlpaal1DGY2iJVKpTHAwrq6OoYPrzjBawNYT1fYKpVKA6b62UC5l4FyH+B76QmNjY00NDQAbFZfX7+odbuXDZiZVUmpVKp1CFXTm/fi0rBm1hVeNmA2wEXExcDkVodX5N0BzMzM+hUnr2YDnKSjax3DYNEXvm6rlt66l+YVTQz1zKuZdYGTVzPrUESsBvwSWDsfehk4WtKi3H44aeeEZtK2WidI+r+ujPHLnx3GO01Lqhaz9Q+7H3lnrUMws37Ga17NrF0RMUzSO8AUSeNzVbC7gO/n9vWBHwC7S5oAnAH8pGYBm5nZgOaZV7MBLiJOBdaXdEJ+vz4g0j6wp5KKDwwDviPp+nzOg6S9WicDfwX2kvRGoduRwDv59ZD8ZwTwZ9IesS/27F2Zmdlg5ZlXs4HvSuCgXAIWUmGD24BfATtLmkiq9HVeRIwqXPfh3L5X+UBE3BkRi4HPAV8CkPQaMBN4PCKeB84CjoeC1D4AACAASURBVOnhezIzs0HKyavZACfpeeApoJyEzgBmkSp73ZQLI8wF3gtE4dJrJa1o1ddewAdJ5WNPBYiIkcB/ANtL2pRU6vZ/ImJID92SmZkNYk5ezQaHWaTSseOAdYH/A/4beBAYl9eqvkhaQlC2tK2O8vrXy4Fp+dBngCWSlNtvADYH3lf1uzAzs0HPyavZ4HAL8AngK8AsSS2ktamLJLVExKdJpWXbFBEbREQxGT0AWJBfLwQmRcSG+dxdgDdJ5WvNzMyqyg9smQ0Ckt6OiFuBw4DN8uGvAj+OiG8BjwJPttPFRsCsiFid9HDWQmBq7rsUEf8F/DwimoBGYP+cIJuZmVWVk1ezQULSkcCRhff3AltWOPdTrd4vACruWi/p++Sts8zMzHqSk1cz6xN2+twVDB8+vNZhWC9zhS0z6yqveTUzq5JSqVTrEKqmt+7FiauZdZWTVzMzMzPrN5y8mplVSX19xWXB/U5n76V5RVMPR2Jm9m5e82pmfcIDN8ygefmSWodhXbT34XfVOgQzG2Q882pmnVIoL2tmZlYz/p+RWR8WES3A14HPAusDJ0m6ObdNAc4GhgKvAjMlPVuhnw2BewqHNgVmSzohInYEzgFG5rbTJN0REWOAx0jVuXYFLomI+4CfkErLrgC+JunuiFgbmA1sAywHJOnA6nwKZmZmK3nm1azve1PS9qRyrD+CfySjVwGHStoWuBa4plIHkl6RNCGXgT0MWAJcGBHrARcDh0iqB/YGfpKPQ0qYH5U0SdLFeYxr85hTgasjYgNgD2CkpLGSxgMzq/0hmJmZgZNXs/7g+vzPR4APRsSawI7AfElP57YrgAkRMaK9jiJiE+AmUtL7B+BjpIpbd0XEPOAuoIWVpWKXATfka0cAE/JY5LHnAZOB+cDWEXFRRBxAqrJlZmZWdV42YNb3LQOQ1BwR0M1/byNiJHA7cIqkh/PhIcCTkj7RxvljgL91psyrpOciYhtgN2BP4KyIGCdpWXdiNTMzq8Qzr2b90yPA+IjYKr+fDjwh6a22Ts4PW90EXCXppkLTr4AtI2KXwrnbR8SQ1n3kvuflsYiIrYHxwCMRMRpoljQHOIG0Jva9q3iPZmZm/8Qzr2b9kKRXI2IacG1OTF8lrUGtZCfSrOiGEVE+7xpJ50bEvsC5EfEDYA3gOWCfCv0cSloTewLpga1pOZY9gXPyzPBQ4GxJL63ibZqZmf0TJ69mfZikIZXeS7obuLuT/fyclFS21fYo8Kk2mhYB72t17rOkJLh1H3eR1suamZn1KCevZtYn7HrgLIYPH17rMKyLmlc0MXTYGrUOw8wGESevZgNIROwFnNVG09ck3dnb8Qw2pVJpwJSI7ey9OHE1s97m5NVsAMkJqpNUMzMbsLzbgJlZlQyUWVfo+F6aVzT1UiRmZu/mmVcz6xPuunE6K5YvqXUY1kn7H9apZwXNzKrOM69m1qaIOD0izqvQNiMibmqrzczMrCc5ebVeFxFtbtlk/8yflZmZ2bt52UAfEBHXAAEMB54FDgduBC6QdGs+Z2/gK5J2iYixpPry7yFVPNoC+Lak2yv0/yngB8CvgY+SatcfJOmZ3H4KMC2f/ijwn5KWVuir9dPsY4EDJN0aEdOBY0i/V28AX5SkiJhB2kD/LWBLYGpEbAScTdp79FVgpqRnI+1yPwtYO7fNktTm7F+OpwX4OvBZYH3gJEk357YpbY3RTl8P5vv/KPBB4AZJX81tWwA/IVWOWkF6ev/u9mLoS59VRKwNzAa2AZYDknRg7vs6YCSwJnCHpJPbuH4N4AJgV+A14IlC21Dgu8CUfOhuUgna5kqftZmZWXd55rVvOE7SdpLGAU8Bp5CSkumFcw4jJawAV5ES2zpSUrp9J8bYBrhY0rbADcCpALky0jTgY8A4UhL0jUqdSLpT0gRJE4CLgMeAuRHxceBA4BOS6oFzgZ8WLp0MnJhjfinfw6E5nmuBa/J5xwC3SRqfz728E/f2pqTt8338KN/Xhu2M0Z5NgU8AE4EjI2LLfPwa4Nrc11Tg6ojYoL0Y+thntQcwUtJYSeOBmfn4EmCfHMcEYLuc9Lc2E9iMlIDvBuxQaDsqXzsp/5mYj5mZmVWdk9e+4fMRUYqIBcAhpETgFuDjEbF+RKwPfBK4OSJGAnWkJAZJjwFPdmIMSSrPlj0CbJ5f7w5cL+lNSS3AJflYuyJiD+DLwL6SlpHKiY4Hfh0R84BzgE0Klzwk6Q/59Y7AfElP5/dXABMiYgTwC1LSeGZE7EpKrjpyfeG+PhgRa3YwRntulPSOpDeAZ4DN8zUTch/kPueRksz2YgD6zGc1H9g6Ii6KiAOAxnx8KKk07HygRPrdmtDG9bsAsyUtl/Q2cHWhbXfSrG+TpKYcY4e/Q2ZmZt3h5LXG8izcF4Epeeb1VGDNnCDcSkpmDwFulfS3wqUtXRxqWeF1M6uwZCQixgMXA/tJei0fHgL8tDzTmGcDNy1c1uYyhNbyV/4fB/4AfJU069iRZfna8tfUq7IcprufU5sx9JXPStJzpNn3e0mJ5fycYH8ZGAXsmGd255CWD5iZmfVJTl5rbz3Smse/RMRw0nrXslnAjPynPOv3JmlpwcEAETGJ9HV/d90HfC4iRkTEEOBIUoLTpojYGLgZmCrpd4Wm/yXNII/O5w2NiEobRT4CjI+IrfL76cATkt7Ka0sXS5oFfIt3fz3dFRXH6GpH+Zp5uQ8iYmvSzOkj7V3Xlz6rPFazpDnACaS1u+8l/f69LGlZjne/Cl08AEyLiGERsRbpL1Rl9wHTI2L1iFg9x1jxd8jMzGxV+IGt2rubtIbyd6QHYX5BTkIkPZSXCSDpocI1nwd+GhH/H7Ag/3mjO4NLuisitgUezoceA77dziVHkhKfi9LzQgCcIOn/j4ivA7flB3jWID10VmpjzFcjYhpwbUQMIz2ENDU3HwgcGhFNpNnl47p5X+2N0R2HAj+JiBNID2xNk/RqB9f0pc9qHHBOjmMocLaklyLiR8CNEdEAvAjcX+H6S4BtSUspXiM92Pb+QtsWrHyIay5waTuxmJmZdduQlpaufvtstRYR6wB/k9SSdx54EAhJr9c2MrOuK5VKY4CFdXV1DB8+vNbhWCc1r2hi6LA1ah1Gp5RKpQFT/Wyg3MtAuQ/wvfSExsZGGhoaADarr69f1Lrdywb6p48B8yLiSdKDQl9w4mpWe6XSP02e91sd3Ut/SVzNbODxsoF+SNI9wD2tj0fEbaStnoqel7RvV8eIiNOAf2uj6TOSXulqf6uimrG0sfdq2dck3dmd+PqSvvRzMzMz6wlOXgeQ7iSp7fR1BnBGtfpbFdWMJSeo/T5JraQv/dwGo77wdVu1VLqX/rRcwMwGJievZtYnzLnp8yxf3pltfa2Wps6YW+sQzGyQc/JqfVJEnA6sI+nENtpmAHtL2r+346qkGG9EHA2sJen8GsYzhlRquKFweDdJf+lmfy3ACElLI+J4UrWxV3JbuYLYRODOvvRzMTOzgcfJq1mVSbq4luPnLbUAluTStNV2PGlv1/Ia2ldIxQ4mAJ/ugfHMzMz+wcmr9YqIWBuYTarytBwQ8CXgOmAkqarTHZJObuPaNYALgF1Je4w+UWgbCnwXmJIP3Q2cUqh01bqvMaS9bC/N16xF2sP1aFIp1r+TqmEtzuefAvw76d+VP5F2dlgcEesCl5PKqS4GXgD+nK85nZWzsF2N7zJggaQf5vd1wG2kcr4Hk/ZyLS84PFHS/fm8RaSdJ3Yl7ft7Zlv9t6c4u9rW+3zs68AHgZsiYhlwSC5d+1Iu3mBmZtajvFWW9ZY9gJGSxkoaD8wElgD7SKonzdptFxFT2rh2JrAZMBbYjXdXkjoqXzsp/5mYj7VnfeAhSRNJCej9wEW5PGoJ+A+AiJhKShonS5pEetDre7mP04A3JW0F7A98ssJYXY1vFrmSV3YYMEtSC2nz/8k57oNIfxkoGilpB0lHlN9HxGMRUYqIk3IFtVUi6TvAS8D+ubTt06vap5mZWVc4ebXeMh/YOiIuiogDgEZSpadzI2I+KWmsIyV6re0CzJa0XNLbwNWFtt1JyV2TpCZSGd3dO4hlqaQ78uvHgRclzcvvS6RqUQD75r4ej4h5wLHAmEJMlwNIeg24pcJYXYovV1IbERHj8tf/B7MySd0cmBsRTwE/AzaKiI0Kl19ZeP0yMFrSdsCepNnjIzAzM+vnnLxar5D0HGnJwL2k5G0+aZ3kKGDHPOs5h7R8oKc1Fl43A8tavS8vpxkCfDvPME6QVCdpp16IbzYwg5R0PiPpj/n4dcCPJW1DmsVdwbs/r398vS+psfxAVf7nNUBHsTeT/5sQEb3xczAzM+syJ6/WKyJiNNAsaQ5wArAB8GHgZUnLImJjYL8Klz8ATIuIYRGxFnBIoe0+YHpErB4Rq5O+cr+3SmHfBhwTEaPyPQyPiPGFmA7Lx9cHPluhj+7EdyVpxvVI0kxt2XrAwvz6cKBiLdWI2DCPV15vvC8wr9L52bPA9vn1Ie2c9yawbgd9mZmZ9Qg/sGW9ZRxwTkRAWi5wNukBoxsjogF4kbT2tC2XANsCz5Ae2HoUeH+hbQtWPsQ1l/Qw1iqTdFVEvA/4eY57NeDHpFnjM4GfRsRvSQ9s/aKd2LsUn6TnI+Jp4FOkJLbseGBORLxOevCrvW2vdgbOiIhmYHXgduDC9sYlzYT/JCLeAG5o57wfAVdExNukJPdt4CFgbWDNiHgR+KakyzsYz8zMrMuGtLS01DoGMxvESqXSGGBhXV0dw4dXnEy2PqI/VtgqlUoDpvrZQLmXgXIf4HvpCY2NjTQ0NABsVl9fv6h1u5cNmJlVSalUqnUIVVPpXvpb4mpmA4+XDdiAFBEXA5NbHV6Rn76vuVrFFxFHkrcCa2VGYccFMzOzPsvJqw1Iko6udQztqVV8ki4DLqvF2INBX/i6rTNWNDcxbKhnUM2sf3LyamZ9wg23fJ6m5UtqHcagcMTn59Y6BDOzbnPyar2iWDK1jbYZwN6S9u/tuCppVeL1aGAtSefXMJ4xpK2sGgqHd5NUcceBXDJ2b0kNlc6pUmyrAb8k7TYAqUDC0ZIW9eS4ZmY2ODl5NeuApItrOX6utAWwRFJbFchqStI7ETFF0hsAEXEc8H3g32obmZmZDUROXq1b8sb3s0lVs5YDAr5EqgI1klT56Q5JJ7dx7RrABcCupH1bnyi0DQW+C0zJh+4GTpHUXCGOMcBjpL1TpwBrAYcCRwM7An8H9pO0OJ9/CqlU6jDgT8AXJC2OiHVJ5V7rSPu2vgD8OV9zOitnYbsa32XAAkk/zO/rSMUPNift4XocUF58eKKk+/N5i0j74O4KLCDtK9sdB0bEpcAHgPMkXZj7Pw/4ZB77NeDwciWviNgbOJ20P+w7wHRJT0bEjsA5pJ8vwGnlMrvlxDUbma8zMzOrOm+VZd21BzBS0lhJ44GZwBJgH0n1wARgu4iY0sa1M4HNgLHAbsAOhbaj8rWT8p+J+Vh71gcekjSRlIDeD1yUS86WyE/XR8RUUtI4WdIk4E7ge7mP04A3JW0F7E9K7NrS1fhmkapqlR0GzJLUQipYMDnHfRDpLwNFIyXtIOmI8vuIeCwiShFxUkQMaWfcsrUlfZRU8OCciFgnHz9H0vb5Z3cdKSEnIj5CeqDr4Nw2GVgYEesBFwOH5J/v3qSCBuuVB4qIOyNiMfA50l9kzMzMqs7Jq3XXfGDriLgoIg4AGkmVs86NiPmkpLGOlOi1tgswW9JySW8DVxfadicld02SmkjlUXfvIJal5RlA4HHgxcK2TyVShStIJVJ3Bx6PiHnAscCYQkyXA0h6Dbilwlhdik/SQ8CIiBiXv/4/mJVJ6ubA3Ih4CvgZsFFEbFS4/MrC65eB0XkrrT1Js8dH0LHrcxyLgNeB0fn4nhHxSK5udiIrf06fBu6U9Pt8XaOkt4CPkf7CcVf+7O4CWlj52SJpL+CDpGT41E7EZmZm1mVeNmDdIum5iNiGNHO6J3AWKdkaBewoaVlEXEJaPtDTGguvm4Flrd6Xf8+HAN+W9NNeiKloNjADeBB4pvz1PCnJ+4qkOfmhp7d59+e1tPxCUiPwSn79SkRcA+xEx9te/dNnEREfAs4Htpe0MCI+BlzbQT9DgCclfaK9k/L618uB3wPHdNCnmZlZl3nm1bolIkYDzZLmACcAGwAfBl7OievGwH4VLn8AmBYRwyJiLeCQQtt9wPSIWD0iVid95X5vlcK+DTgmIkblexgeEeMLMR2Wj68PfLZCH92J70rSjOuRpJnasvWAhfn14UDF2qgRsWEer7zeeF+gu0UFRgJNwOKcNBf3nL0H2CsitsxjDY+IEcCvgC0jYpdCTNtHxJCI2CAi3lfo4wDSOl0zM7Oqc/Jq3TUOeDgvEfgNcDbpIZ+d8lfR5bWnbbkEeB54hpQ0Ptqq7UnSQ1xP5NeXViNgSVcB1wA/j4gnSUsKdsrNZwKjIuK3wM3AL9qJvUvxSXoeeJq07rS4HOF4YE5EPE5K/CtuewXsDDyRP+/H8tgXtjduO/EsAG7MMf2alQk0ebnAF4Cf5bEeBsZIep2UMH8zIuZHxDOkn/cQYCPS8ocnI2IBaenB1O7EZmZm1pEhLS0ttY7BzAaxUqk0BlhYV1fH8OEVJ5+tijpTYatUKvWbimEd8b30PQPlPsD30hMaGxtpaGgA2Ky+vn5R63bPvJqZVUmpVKp1CJ3i0rBm1p/5gS3rFyLiYtK2TUUr8tP3NVer+CLiSPJWYK3MKOy4YGZmNmA4ebV+QdLRHZ9VO7WKT9JldLzjgPWSvvB1W0c6s2TAzKwvc/JqZn3CVXM+T+PyJbUOY8A7durcWodgZrZKnLxan1QsydpG2wxgb0n793ZclbQqIXs0sJak82sYzxjgWaChcHg3Se3taNBefy3ACElLI+J44FpJr+S2KaQKXS2kkrJzgFNzFTEzM7OqcvJqFUXEUEnNtY6jv5F0cS3Hz5W8AJZIaqvC2ao6nrTf7Sv5/UPAJEnNeS/aX5K24LqtB8Y2M7NBzslrF+SqRkHaTP5Z0sbyNwIXSLo1n7M3qWrSLhExlrQp/XtIG8pvQarwdHuF/j8F/ID0P/6PkmayDpL0TG4/BZiWT38U+E9JS9voiojYi1T1qmwscICkWyNiOqn60TDgDeCLkpRnNKcCbwFbAlNzudKzSaVfXwVmSno2IgKYBayd22ZJOq9CLGuTqkxtAywHJOnA3Pd1pE3z1wTukHRyG9evAVwA7Aq8RtrjtNw2lDTrNyUfuhs4pVLSnWckHyPtzToFWAs4lLRR/47A34H9JC3O559CKsU6DPgT8AVJiyNiXdJetnXAYuAF4M/5mtNZOQvb1fguAxZI+mF+X0dKAjcnFTo4DigvWDxR0v35vEWkUrC7kgoEnNlW/+0pzq629T4f+zqpBOxNEbEMOETS04Vu1szxvdPV8c3MzDrDW2V1zXGStpM0DngKOIWUwE0vnHMYK6soXUVKbOtISen2nRhjG+BiSdsCN5BrxEfEnqTE9WOkAgFDgW9U6kTSnZIm5Jm3i0gJ29yI+DhwIPAJSfXAuUCxXOpkUlJUB7yU7+HQHM+1pE3+ISW/t0kan8+9vJ172gMYKWmspPHAzHx8CbBPjmMCsF3+Crq1mcBmpAR8N2CHQttR+dpJ+c/EfKw96wMPSZrIymIKF+V7LJGf3o+IqaSkcbKkScCdwPdyH6cBb0raCtgf+GSFsboa3yz++fdpVv4Kfm6OZSJwEOkvBEUjJe0g6Yjy+4h4LCJKEXFSRAxpZ9xOkfQd0u/F/vn362mAiNguF354hVR44o5VHcvMzKwtTl675vM5EVhAKmk6gVQx6eMRsX4uK/pJ4OaIGEmalbsWQNJjpGpMHZGk8sziI6TkCWB34HpJb+ZE5pJ8rF0RsQfwZWBfScuAfYDxwK8jYh5wDrBJ4ZKHJP0hv94RmF+YWbsCmJDLhf4CODIizoyIXUmJaCXzga0j4qKIOABozMeHAufmSk4l0ufV1tfcuwCzJS2X9DZwdaFtd1Jy1ySpKcfY0eeyVFI5uXoceLGwrVSJNEMOqaLU7sDj+bM6FhhTiOlyAEmv8e7KWUVdik/SQ8CIiBiXv/4/mJVJ6uakv4A8BfwM2CjPXpddWXj9MjA6b9W1J2n2+Ah6iKTHcvK/CVAPfLynxjIzs8HNyWsn5RnLLwJT8szrqcCaOZm6lZTMHgLcKulvhUu7+tDKssLrZlZhaUdEjAcuJn0N/lo+PAT4aXlWNs+cblq4rM1lCK1JupmUoPwB+CpphrbSuc+RZpTvJSVu8yNiTVJSPQrYMSc+c0hfO/e0xsLrZip/5kNIyzzKn1WdpJ3oebOBGaSk8xlJf8zHrwN+LGkb0izuCt79ef3jZyepsfxAVf7nNawshVtJM/m/Cfnn02X59+wu4IDuXG9mZtYRJ6+dtx5pfehfImI4ab1r2SxSsjGDvGRA0pukpQUHA0TEJNLX/d11H/C5iBiRv/49kpQMtikiNgZuBqZK+l2h6X9JM8ij83lDI6LS5pSPAOMjYqv8fjrwhKS3ImILYLGkWcC3ePdX+a1jGQ00S5oDnABsALyX9Jm+LGlZjne/Cl08AEyLiGER8f/Yu/P4uqpy/+OfWiCMFSx4ERCLoF+BQFtCGUUGFVp/DBdFFCi01DKqF1EULyoiIqBwL4jUi4wBZFCGC5UZROBWKdJDWxqGryCtlYsV8DIPgdb8/ljrkN1wzslJmtOkzfN+vfLKnvfaO9OTZ6+9nlVI/ySU3QVMkLRiflloAjXuSw9NBY6WtFa+jqb8D0G5TYfm5cOBfascozftu4z0fTOZzi4okO7X3Dw9idT3uiJJ78/nK/c53pvU77qWJ+ns2nJgje1eBt5bONdHJZWD3tVIQfecbs4VQggh9EoEr/W7jZRl/BNwL+lxM/DOo95hpD6H0wr7HAJ8LXczOI70B/2l3pzc9q2kx+X30xkYnFJjl8mkIHGKpFn5Y1fb9wHfAabmx/VtVAkabT9H6md7Ze7POD5/QOo3O0fSTNLLVMfUaMsWwP35fH8ETrP9DHAOsKOkNjr7nlZyPjAfeIwUND7YZd3DpJe4ZubpC2q0pW62LydlLO/N11+iM3v5Q2AtSY+T/km4r0bbe9Q+2/OBR4FdWLw7wteAGyQ9BHwYqDXs1ceBmfmez8jnPrfWeUmZ8F9IKpG+d6o5B7gkf09tRvr+mZPP9QDpXkThhBBCCA0xpKMjhmJsFEmrA6/Z7sh/5O8BZPuF/m1ZCANHqVQaAcxtbm6mqalqMjn0kXorbJVKpWWiYlg94loGnuXlOiCupRHa29tpa2sD2KilpWVe1/WReW2sHYBZOWt3NWmYpQhcQ1hOlUql/m5Ct6I0bAhhWRfjvDaQ7TuAO7oulzQV2LDL4vm29+7pOSSdCHy2wqrdyy/sLC0DrC3nkYb9KlqY377vd/3VPkmTyUOBdTGxMOJCCCGEMGBF8NoPehOk1jjWycDJfXW8JTHA2nJkf7ehlv5qn+0Lif6oDTMQHrcV1dtFIIQQliURvIYQBoQLph7CmwtrDRcceuq4A27v7yaEEEKfi+A1LBXFkqkV1k0E9rS939JuVzVdSrweCaxi+6x+bM8I0lBWbYXFn7RddcSBXDJ2T9tt1bbpo7a9B/g9qVQwpAIJR9qe18jzhhBCGJwieB1EJA21vai/27GssX1ef54/V9oCeDGX+x1QbP9T0ljbLwFIOgb4Tyr3fw4hhBCWyKAOXiVdAYg02PuTpIHfrwF+ZvvGvM2ewDds75qHu7oEWI004PsmpApMN1U5/i7A2aSxL7cnVdv6ou3H8vrjSeOoQhq79Ku2K1a4kvQZ4NTCos2Az9u+UdIE4GjS1/Ml4CjbzhnN8cArwEeA8bmc6Gmk0qzPAUfYflKSSMUWVs3rWm2fWaUtq5KqQG0OvE0qabt/PvZVpDFvVwZutv2tCvuvRBobdjfgedIYpOV1Q4EfA2PzotuA46sF3TkjOYM0dupYYBXgIOBIUnnbN0gVxhbk7Y8nlUpdAfhf0ggQCyS9lzTWbDOwAPgr8Pe8z0l0ZmF72r4LgTm2f5rnm0nFDzYmFSI4Bih3SjzO9m/zdvNII1TsRhrX94eVjl+H/SVdAHwAONP2ufn4Z5JKGa9E+hpMKlfyyt/zJwErAv8EJth+WNK2pHLCw/KxTyyX2S0HrtmwvF8IIYTQ5wb7UFnH2N46l3t9BDieFMBNKGxzKJ1Vji4nBbbNpKB0DN3bHDgvlz/9NamsLJLGkQLXHUiD+A8FvlftILZvKZcpBaaQArbbc9na/YFP2G4BzgAuLuy6HSkoagaeyddwUG7PlaRB+CEFv1NzudhmUiBXzR6kggyb2R4JHJGXvwjsldsxCtha0tgK+x8BbEQKwD/J4tW5Ds/7bpU/RudltQwHptkeTWexgyn5Gkvkt+sljScFjdvZ3gq4BfiPfIwTgZdtfwzYjxTYVdLT9rXy7u+nVtsdwO25LaOBL5L+ISgaZnsb218qz0uaIakk6Zu50lp3VrW9Pangwel57GGA022PyV+/q0gBOZI+Snqh64C8bjtgrqQ1SaWGD8xf3z1JBQ3WLJ9I0i2SFgBfAP6tjraFEEIIPTbYg9dDciAwh1QOcxSpotFOkobnsp87A9dJGkbKyl0JYHsGqVpSd2y7nFmcTgqeAD4FXG375RzInJ+X1SRpD1IlpL1tvwnsBYwEHpA0i5QZ+2Bhl2m2/5yntwVm2340z18CjJK0Bqkq0mRJP5S0GykQrWY2sKmkKZI+D7Tn5UOBM3KlpRLpflV6zL0rcKntt22/TqocVvYpUnD3lu23chu7uy+vljOApMpnTxeGfSqRMuSQSqR+Cngo36svAyMKbboIwPbzLF7ZqqhH7csV19aQtEV+/H8AnUHqxqR/QB4BfgWsm7PXZZcV/olKsQAAIABJREFUpv8GbJCH0hpHyh5/ie5dndsxD3gB2CAvHydpeq5udhydX6dPA7fYfiLv1277FdI/WRsBt+Z7dyvpSUL53mL7M8B6pGD4u3W0LYQQQuixQdttIGcsjwJ2sP2cpAOBw22/LulGOmu732j7tRy8QvqD3RNvFqYXsQT3XNJIUvZrjxxgAQwBLrZ9YpXdKnZD6Mr2dZLuB3YHvk3qQjG+yrZPSdqclDUdB5wqaQtSUL0WsK3tNyWdT+o+0GjthelFVL/nQ0jdPIqZ6aXhUmAiqcLaY+XH86Qg7xu2b8gvPb3O4vfrna+d7Xbg2Tz9bO7ysiPdD3v1rnsh6UPAWcAY23Ml7UD+p6yGIcDDtj9Ra6Pc//Ui4AlSNj+EEELoU4M587omqX/oPyQ1kYK1slZSsDGR3GXA9sukrgUHAEjaivS4v7fuAr4gaY38+HcycGe1jSWtD1wHjLf9p8Kq35AyyBvk7YZKqjbY5HRgpKSP5fkJwEzbr0jaBFhguxX4AYs/yu/alg2ARbZvAI4F1gHeR7qnf8uB6/qkmveV3A0cLGkFSavQ+Y8CpPsyQdKKklbMbax6X3poKnC0pLXydTTlfwjKbTo0Lx8O7FvlGL1p32Wk75vJdHZBgXS/5ubpSaS+1xVJen8+X7nP8d6kfte9MQx4C1iQg+bimLN3AJ+R9JF8rqacmf8D8BFJuxbaNEbSEEnrSFq7cIzPk/rphhBCCH1uMAevtwF/Bv4E3Et63Ay886h3GKnP4bTCPocAX8vdDI4j/YEuvqhSN9u3kh6X30/nH/pTauwymRQkTpE0K3/savs+4DvA1Py4vo0qQaPt50j9bK9UKlk7ns7s6v7AHEkzSS9THVOjLVsA9+fz/RE4zfYzwDnAjvlRdLnvaSXnA/OBx0hB44Nd1j1MeolrZp6+oEZb6mb7clIf33vz9ZdI2UtIL0StJelx0j8J99Voe4/aZ3s+8Cip32mxO8LXgBskPQR8GKg67BXwcWBmvucz8rnPrXXeGu2ZQ3ox8VHSy4RzC+ueAA4DfpXPdT8wwqms8d7A9yXNlvQY6aWuIcC6pO4PD+efjU9TJWsfQgghLKkhHR09fQo+eOWXXV6z3aE08sA9gPIf9hBCL5RKpRHA3ObmZpqaqiafQy8sSYWtUqk04CqG9VZcy8CzvFwHxLU0Qnt7O21tbQAbtbS0zOu6fjBnXntjB2BWztpdTRpmKQLXEAKQfvEPJFEaNoSwPBq0L2z1hu07SH0CFyNpKrBhl8Xzbe/d03NIOpHKg7vvbvvZnh5vSQywtpxHGrapaGF++77f9Vf7JE0mDwXWxcTCiAshhBDCciOC1z7QmyC1xrFOBk7uq+MtiQHWliO736r/9Ff7bF9I9yMOhKVkIDxuK3p70VusGNnXEMJyJoLXEMKAcPbNh/DGwlrDC4eeOmn/2/u7CSGE0Oeiz2sIoSpJ35fUkcva1tpu91z9qz2Xng0hhBAaIoLXEMI7chWw8vRWpH68f6m+xzueIg3ndkaDmhZCCCEA0W0ghIaQ1EEaf3dfYDjwzVzFbAQww/baebt35svTpHFjxwKrAAeRighsC7wB7GN7QZVzjgc+Z3vfPL8CaTzdHYHVgZ8Dq5GqeJ1v++y8XSuwEBCwBqlkcBMwhVRc4Z7urtf2k/lY/1rnLQohhBB6JTKvITTOy7bHkApDnFPnPsOBabZH01noYYrtLUlFFSqNLFB2PbBTodrVOOBx23OBecCnbG9Fqp52uKRNC/uOAsbaHpXnTwZ+aXtene0OIYQQlooIXkNonKvz5+nAepJWrmOfV23fnKcfAp4uDHlVAjaptqPt14Eb6Cy3O5FU6hhgVeCiXAHr98B6wMjC7tfafg1A0vbA1qRMbQghhDCgRPAaQuO8CWB7UZ5fgfR4vvhz1zWgbS9MLyofozDfXVefVmCCpOHAzsC1efmpwAJgtO2RpLK+xXO/WpjeGdgUmCtpHrABqfzr7t2cO4QQQmi4CF5DWLoWACtKKmdQD6y1cU/ZngYMA04DbsjZWIA1gb/aXphHDtipxjFOt72e7RG2RwBPA3vkIh0hhBBCv4rgNYSlyPZC4BjgTkl/JGVT+9qlwGF0dhkAOAU4LJc2Pgm4ry9PKOnjkp4Gvg4cIelpSXv05TlCCCEEgCEdHR393YYQwiBWKpVGAHObm5tpamrq7+YsV5akwlapVBpwFcN6K65l4FlergPiWhqhvb2dtrY2gI1aWlrmdV0fmdcQQugjpVKpv5uwmCgNG0JYHsU4ryEsYyRNpvKQWRMLIxM04ryjWLwrQtm5ti9s1HlDCCGEogheQ1jG5EBxqQeLOTAe1e2Gg1h/P25bkm4CIYSwrIjgNYQwIJx82wReW/hifzdjmXb2527r7yaEEELDRZ/XEEJVknaRNKPKuhGSnl/abQohhDC4RfAawgAnKZ6QhBBCCFn8UQyhAkkdwHeAfYHhwDdtX5fXjSUVARgKPAccYfvJKsd5P1Ac3H9D4FLbx0raFjidVFQA4ETbN0saAcwgvRy1G3C+pLuAXwDrkKp0nWD7NkmrksZ13Rx4G7Dt/Wtc1/eBA0iVuzqAXW2/KOkKQEAT8CQwyfYLFfb/MnAs8DJwc5d1hwDfzMf9c74vz1ZrSwghhNAbkXkNobqXbY8BDgbOgXeC0cuBg2xvCVwJXFHtALaftT3K9ijgUOBF4FxJawLnAQfabgH2BH6Rl0MKmB+0vZXt8/I5rsznHA/8UtI6wB7AMNub5bKvR1Rri6T3kQLP0bk9n6CzLOwxtre2vQXwCHB8hf23JAX0O9reKrexvK6ZFIjvntvYBvysWltCCCGE3orgNYTqrs6fpwPrSVoZ2BaYbfvRvO4SYJSkNWodSNIHgWtJQe+fgR2AjYBbJc0CbiVlLMtlY98Efp33XYP0lv8lAPncs4DtgNnAppKmSPo80F6jGS+RsqqXSToMWD1X/AI4RFJJ0hxSydpKowrsAtxs++95/vzCul2BW2z/Lc//AvhUrXsSQggh9EYEryFU9yaA7XIJ1151s5E0DLgJON72/XnxEODhclY2f3zQdvnlqNdsd1v+zvZTpC4Dd5KCxdk5yK607SJSwHsusAFQkrSlpJ2Ao4CxOfP6XaDiMUIIIYT+FsFrCD0zHRgp6WN5fgIw0/YrlTbOL1tdC1xu+9rCqj8AH5G0a2HbMZKGdD1GPvasfC4kbQqMBKZL2gBYZPsGUpeAdYD3VWnLGsA6tu+1/X3So/1mYE1SVvYfkpqASVWu/R7gM7nrBMCXCut+l9etm+cPIwXUIYQQQp+KF7ZC6AHbz0k6GLgyB6bPkfqgVrMj8Eng/ZLK211h+wxJewNnSDobWAl4CtirynEOIvWJPZb0wtbBuS3jgNMlQXqB7DTbz1Q5xnuB6yStQvrH9SHgemBRvoY/Ac8D9wHbVLj2hyWdCvxe0svALYV1bZK+DdyZX3Z7ihr9b0MIIYTeGtLR0e2TyRBCaJhSqTQCmNvc3ExTU1N/N2eZ1pcVtkqlUr9XDOsrcS0Dz/JyHRDX0gjt7e20tbUBbNTS0jKv6/roNhBCCH2kVCr16/mjNGwIYTCIbgMh9AFJnwFOrbDqBNu3VFg+KNoSQggh9LUIXkPoAzkoHBCB4UBqy2DTyMdtby16i5UisxpCCBG8hhAGhi/fOYGXF73Y380YsH69z2393YQQQhgQIngNYZCSdAOpUMI/SZW2vmp7Vl73UVLZ2eHAP4BDbD9R41ijgCnAaFKxgv0a3PwQQgiDVLywFcIgk4f4Aphge6Tt0cCZwMWFzc4Dptj+KCko/UU3h30W+DpprNkQQgihYSLzGkIv5fFMvwPsS8pQftP2dZJGADNsr523e2e+PA1cAIwFViGN4XokqfTsG8A+thdUOed44HO2983zKwDzSePJrg78HFiNVCHrfNtn5+1aSePDClgDGGX7pcKh30vKwJKLEGwFfDqvuwo4V9I6tp+r1K48tuwzuYBCCCGE0DCReQ1hybxsewxwMHBOnfsMB6bljOdFwG9JWc4tgRLwlRr7Xg/sJGntPD8OeNz2XGAe8CnbW5GKDBzeJZgcRSoBO6q8QNKFkuYDPyJX8AI+CPxvuSxu/vxMXh5CCCH0qwheQ1gyV+fP04H1JK1cxz6v2r45Tz8EPF3ua0oKXjeptqPt14EbgAPzoolAa55eFbhI0hzg98B6pDKyZdfafq3L8Sbb3hA4ATijjraHEEII/SqC1xCWzJvwTnYSUlechSz+s9U1oG0vTC8qH6Mw3113nlZggqThwM7AtXn5qcACYLTtkcAfu5z71WoHtH05sGs+5l+B9SUNBcif18vLQwghhH4VwWsIfW8BsKKkcgb1wFob95TtacAw4DTghpyNBVgT+KvthZKagZ2qHUPS6pI+WJjfC/g/4P9sPwvMAg7Iqw8AZlbr7xpCCCEsTfHCVgh9LAePxwB3SnoOuLm7fXrhUuCHLB6gngJcLulLwJ+A+2rsvxpwjaTVSNne/wP2st2R1x8JXCrpROAF4JBajckvok0jdV1YWdLTwPdtX9TTCwshhBBqieA1hF6yPaTavO2LWXzoqR/k5fOAtQvb3QNsXZhvpbMPa61zn0IKVovLZgLNVbaf2GX+78B2NY7/OGn0g7rk69qg3u1DCCGE3orgNYQwIEz59KU0NTX1dzMGrCgPG0IISQSvIQxAkiZTecisiYWRCZaqPP7rHRVWXW/75KXdnoGoVCrR0tLSkGNH4BpCCEkEryEMQLYvBC7s73YU5Re5RnW74SDWqMAV4K1Fb7PS0BUbdvwQQlhWRPAaQhgQJt55PC8ueqW/mzFg3bpPvPsWQggQwWsIywRJE4GzSVW0AOaWS8T24li7AGfa3lrSmsDhtn9SWP8d4AukUQiGAKfZ/lWN4zUBN5JfPCuXxQ0hhBAaIYLXEAY4SeWf07ts79fHh18T+Bbwk8Kyc23/KJ97PeBxSXfYfqHKMRYBZwLPA3f1cftCCCGExUTwGpaYpA7gO8C+wHDgm7avy2N/zihn4orz5WngAmAssApwEGl80W2BN4B9bC+ocd55wC+BTwLrA98G3k8qCvA+YJLt+/K2n8ltXBl4CzjW9nRJIg1NtSowFGi1faaklYAfkSpYNQEPA0fZflXS4cCxpEpZ7wH2z0NLVWrjE8B+tmfn+a8ALbYPlXRmPv5KpMBvku2/FO5NK7AbcD6LV+HqVjG7Wmm+YAqwpqRZwOu2d7D9UmH96kAHNQqa2F4I3JXbHUIIITRUVNgKfeVl22OAg4Fz6txnODDN9mjgIuC3wBTbWwIlKr9t31WT7e2Bz5EC4bdtbwOcQCqXiqSNge8B42y3AJOBX+f9jwam2h5puzm3A1I28iXb2+RSq88A/57XnQHsZnsUMAaYX6N9lwITCvOHApfk6dNtj8nHvwr4cZd786DtrWyfl5ftLGmWpPsk/b867k09vgy8aHuU7R3KCyUdKelxYCapW8E/+uh8IYQQwhKJzGvoK1fnz9OB9SStXMc+r9ouV596CHi6MAxUCfh0Hcco98V8iJQ9Lc+XgHJ51j2AjYH7UqIVgBUk/QupCtVPJK0K/C5/AOwNDJNUfkzfBMzO03eTqk/9BrjZ9lM12ncZ8ICkbwGbkh7T/09eN07Sl0nZza4/i2/SGWAD3AT8yvYbkkYDt0ra1fZjNc7dazlgPk/SFsAVku6KADaEEMJAEMFr6CtvAthelAPEFYCFLJ7d7xrQthemF7H4o/FF1Pf92fW85WMU9x8C3Ga7UonT6yTdD+xO6nYwCRif9zna9t0V9vksKeO6G/A7SUfavrVS42zPl/QIMA7YhdQtoUPSh4CzgDG250raAbiysOtrhVKt2H6+MD1T0u+BbYBqwWt3974utudIeia3/breHCOEEELoS9FtIDTSAmBFSeUM6IH91I47gLGSNi8vkDQmf94EWJDLsv6AFBACTAW+LmmVvN0akjbNL0992PYfbZ+ejz26m/O3kroqHEDqRgAwjNT3doGk95D6+lYlaf3C9IdIpV0frrHLU8CHJa0laUg+dyUvA6sWXgpD0maF6Y1I1/dorfaFEEIIS0sEr6Fh8os8xwB3SvojKRvaH+14gpRNvUjSbEmPAUfk1fsDcyTNBH6W2wtwOqmbwIOSHgamkR77DwVaJc2RNBv4APCLbppwPSlz+ajt+blNc4BrSEHhA8Dcbo7xZUmP5HNOBU6wPbPGNT8D/Aep+8QfgL9V2e7/gCtI9+APefFJ+VyzgGuBf+uue4KkB4H7gbUkPS1pQBVYCCGEsPwY0tHR0f1WIYTQIKVSaQQw98y/XR5FCmpY2kUKGlnqdmmLaxl4lpfrgLiWRmhvb6etrQ1go5aWlnld10ef1xDCgND66R/T1NTU380YsKI8bAghJBG8hgFN0mQqD5k1sTAyQb+TNIN3/zxNt12zL2sfnPc8Uv/XooUVxnPti3NNBTbssni+7b37+lzLqkZmLSJwDSGEJILXMKDZvhAY8P0nGxEs1nnehgbHXc4VQWo3+jpwjWxrCCG8WwSvIYQB4dA7TufFRa/3dzMGlFv+9cfdbxRCCINMjDYQQgghhBCWGZF5DUuFpJOA1W0fV2HdRGBP2/t1Xddfiu2VdCSwiu2z+rE9I4AngbbC4k/WqnolaR7pvrZV26aP2vZR0nBhHyAVR3iQVODhjUaeN4QQwuAUwesgImmo7X4Za3VZlkul9ptCAYEXbY/qz7ZU8Rbw9Vz56z3AVcBxwA/7t1khhBCWR4M6eJV0BSBS3fonSaVBrwF+ZvvGvM2ewDds75orD10CrAbMAjYBTrF9U5Xj7wKcTRqEfnugA/hiecB3SccDB+fNHwS+avvVKsf6DHBqYdFmwOdt3yhpAnA06ev5EnCUbeeM5njgFeAjwHhJ6wKnkQbbfw44wvaTSrVVW4FV87pW22dWacuqpEpRmwNvA7a9fz72VaTqUSsDN9v+VoX9VyIVBNgNeB6YWVg3FPgxMDYvug04vlrQnTOSM4AL8j6rAAeRKlZtC7wB7GN7Qd7+eOBz+V79L3CY7QWS3gtcBDSTKoP9Ffh73uckOrOwPW3fhcAc2z/N882kIgMbk6peHQOslDc/zvZv83bzgKvzPZpD7wPB/SVdQMqKnmn73Hz8M4Gd87mfBybZ/ktetydwErAi8E9ggu2HJW1LKt4wLB/7RNs3254HzAOw/c9ckGLTXrY3hBBCqGmw93k9xvbWtrcAHgGOJwVwEwrbHEoKWAEuJwW2zaSgdEwd59gcOM/2lsCvge8CSBpHClx3ALYgBYzfq3YQ27fYHpUzb1NIAdvtknYiVYn6hO0W4Azg4sKu25GCombgmXwNB+X2XEmqrgQp+J1qe2TettaI6HsAw2xvZnskndWqXgT2yu0YBWwtaWyF/Y8ANiIF4J+ksyQrwOF5363yx+i8rJbhwDTbo3O7fwtMyddYIg+1JWk8KWjczvZWwC2kKlQAJwIv2/4YsB8psKukp+1r5d3fT622O4Dbc1tGA1+ks3Rs2TDb29j+Unle0gxJJUnfzGVfu7Oq7e1JFb5Ol7R6Xn667TH563cVKSAvdwG4EDggr9sOmCtpTeA84MD89d0T+EVe/o5cTncSKUAPIYQQ+tygzrwCh0g6iJR9Wg34E3AycJak4XmbnfN2w0hZuSsBbM/IZUO740IZz+nAXnn6U8DVtl8GkHQ+8NPuDiZpD+DrwE6235S0FzASeCAlTxkCrFXYZZrtP+fpbYHZtst16i8Bfi5pDeA+4Cc5q/q7/FHNbGBTSVOAe4Cb8/KhwBmSdsjtWJcU6N3WZf9dgUttvw28LemXwMcL96XV9lv5ei8B9gX+q0Z7XrVdbsNDwNOFMWBLwKfz9N7A1sBD+V6VM9XlNn0VwPbzkq6vcq4etc/2NElrSNoCeIyUbd0+r94YuErS+qQM9rqS1i1niYHLCof6G7CB7WclvZ8UHL5A98OIXZ3bMU/SC8AGwOPAOElfBlZn8d8DnwZuySV1sd0OtOfM/0bArfneQXqSsAnpH6ly94argbttR/AaQgihIQZt8JozlkcBO9h+TtKBwOG2X5d0I3Bg3vRG26/l4BXSH+yeeLMwvYgluOeSRpKyX3vYfj4vHgJcbPvEKrtV7IbQle3rJN0P7A58m5Q9G19l26ckbU7Kmo4DTs3B2ddJgfO2ObA+n9R9oNHaC9OLqH7Ph5C6eRQz00vDpcBEUqD/WPnxPCnj+Q3bN+S+oq+z+P1652uXg8hn8/SzucvLjnQfvL7rXkj6EHAWMMb23PzPxpXdHGcI8LDtT1RambtTXEEKqP+tm2OFEEIIvTaYuw2sScq6/UNSEylYK2slBRsTyV0Gcob0EVLmDElbkR7399ZdwBdyVm4IMBm4s9rGOTt3HTDe9p8Kq35DygxvkLcbKqnaSOnTgZGSPpbnJwAzbb8iaRNgge1W4Acs/ii/a1s2ABbZvgE4FlgHeB/pnv4tB67rA/tUOcTdwMGSVsiPmQ8srLsLmCBpRUkr5jZWvS89NBU4WtJa+Tqa8j8E5TYdmpcPJ2VTK+lN+y4jfd9MprMLCqT7NTdPTyL1va5I0vvz+cp9jvcm9bvujWGkl6wW5KC5WOjgDuAzkj6Sz9WUM/N/AD4iaddCm8ZIGpKP0UoKjr+Uu0SEEEIIDTGYg9fbgD+TugrcS3rcDKRHvaQ/8MPydNkhwNckzSG9TT2HzsfOPWL7VuCXwP35OACn1NhlMilInCJpVv7Y1fZ9wHeAqZJmk4ZSqhg02n6O1M/2ytzlYTyd2dX9gTmSZpJepjqmRlu2AO7P5/sjcJrtZ4BzgB0ltdHZ97SS84H5pMfod5NeViuue5j0EtfMPH1BjbbUzfblpOzgvfn6S6TsJaQXotaS9Djpn4T7arS9R+2zPR94lNTvtNgd4WvADZIeAj4MVB32itStYma+5zPyuc+tdd4a7ZlDejHxUdLLhHML654ADgN+lc91PzDC9gukgPn7kmZLeoz0UtcQUvZ9POn7opS/N6f0pm0hhBBCd4Z0dESSpF75ZZfXbHcojTxwD6D8hz2E0AulUmkEMLe5uZmmpqrJ50GpP8vDlkqlPi9321/iWgae5eU6IK6lEdrb22lrawPYqKWlZV7X9YM589obOwCzctbuatIwSxG4hhCA9Iu/L/VX4BpCCAPZoH1hqzds30HqE7gYSVOBDbssnm97756eQ9KJwGcrrNrd9rM9Pd6SGGBtOY80bFPRQttbL812VNNf7ZM0mTwUWBcTCyMuhKWkrzMWby1ayEpD49d0CCEUxW/FPtCbILXGsU4mDdfV7wZYW47sfqv+01/ts30h3Y84sEw49PZzeXFRVJQtumXf7/Z3E0IIYcCJbgMhhBBCCGGZEZnXEEJVkr5PGlVgC9ttNbYbShptYixpLOTTc1Y4hBBC6FOReQ0hvCNXySpPb0Xqx/uX6nu84yBSta2PkCqInSRpRCPaGEIIYXCLzGsIDSCpgzT+7r7AcOCbuYrZCGCG7bXzdu/Ml6dJ48aOBVYhBYVHkkr7vgHsUygf2/Wc44HP2d43z69AGk93R1IZ2J+TyiCvDJxv++y8XSuwEBCwBjAqF+6YQiqucE8dl/wF4ALb/wSek3QD8HngjDr2DSGEEOoWmdcQGudl22NIhSHOqXOf4cA026PpLPQwxfaWpKIKlUYWKLse2EnS2nl+HPC47bnAPOBTtrciVU87XNKmhX1HAWNtj8rzJwO/tD2vznZvyOIZ2vnAB+vcN4QQQqhbBK8hNM7V+fN0YD1JK9exz6u2b87TDwFPF4a8KpEezVdk+3XgBjrL7U4klW0FWBW4KFeH+z2wHjCysPu1tl8DkLQ9sDUpUxtCCCEMKBG8htA4bwLYXpTnVyA9ni/+3HUNaNsL04vKxyjMd9fVpxWYIGk4sDNwbV5+KrAAGG17JKmsb/HcrxamdwY2BeZKmgdsANwuafca550PfKgwvyHw127aGkIIIfRYBK8hLF0LgBUllTOoB9bauKdsTwOGAacBN+RsLMCawF9tL5TUDOxU4xin217P9gjbI4CngT1ykY5qrgEOk/QeSesA/0pn4BxCCCH0mQheQ1iKbC8EjgHulPRHUja1r10KHEZnlwGAU0jB5cOkoa/u6+NzXg48BTxB6iZxcu5rG0IIIfSpIR0dHf3dhhDCIFYqlUYAc5ubm2lqaurv5gwo/VketlQq9Xm52/4S1zLwLC/XAXEtjdDe3k5bWxvARi0tLfO6ro/Mawgh9JFSqdSnx+uvwDWEEAay+M0YwjJG0mQqD5k1sTAyQSPOO4rFuyKUnRvVtJK+ylj0Z8Y1hBAGuvjtGMIyJgeKSz1YzIHxqG437KVJt13Ii4ve7H7DQeDmz36jv5sQQggDVnQbCCGEEEIIy4zIvIYwQEiaBWxv+40e7jcP2NN2WyPaVTjPWODHQAewIqkgwndtd+T13yMVRgBotf3DRrYnhBDC4BSZ1xAGCNujehq4Li2SVgCmAVvlErKjgE8De+X1nwA+DzTnj8/nZSGEEEKfisxrCAOEpA5gDduv5mzqZaQA8QPAmbbPzdvtRGfp1nuBId0c9wlgP9uz8/xXgBbbh0o6k1RRayXgeWCS7b9IGgHMIL2gtRtwvu3zCoddOe/zzzz/BeCycvAt6bK8rK/Hkw0hhDDIReY1hIFrVdvbA7sAp0taXVITcDXwVdtbkILDDbs5zqXAhML8ocAlefp022NyydirSN0CyoYDD9reqhy4Sto6Fzp4FrgbuDlvuyHwl8K+84EP9uhqQwghhDr0KniVtKuknfu6MSGExVwNYHse8AKwASDgddv35HW/Bl7q5jiXAQdIWkHSFqRSsf+T142TNF1SG3Aci48m8Cbw6+KBbM+wvSUpMG2hRpnZEEIIoRHqCl4l3Stpxzx9POmP6pWSTmhk40IY5IrjRi2iejfX0IqOAAAgAElEQVSfmmXybM8HHgHGkV6oarXdIelDwFnAAbabgUmk7gBlr5VfxqpwzOeBW0n9XCFlWj9U2GRD4K+12hVCCCH0Rr2Z12ZSvXJINdN3BbYDjmxEo0IIVRlYJfd7RdJ+pExqd1qBycABpG4EAMOAt4AFkt5DNz/Pkj6at0PSaqRgeE5efQ1wiKRVJK0CHEKXrG0IIYTQF+oNXt8DdEjaGBhi+1HbfwXWalzTQghd2W4nBaA/z31PdyFlPbtzfd720ZyJxfYcUtD5KPAAMLebY+wDzJE0O29/H7lYQu7GcD0pw/sIcL3te3twaSGEEEJd6h1tYBpwLumt5/8GyIHs8w1qVwiDju0hhekRXdaNKEz/D7BFYXWlUrFdj/068N4Ky48Bjiks+n5ePg9Yu8u2ZwBn1DjHScBJ3bWlmovHTqapqam3uy9XojxsCCFUV2/mdSLwIvAwnX+cPgb8tO+bFEIIy6ZSqdQnx4nANYQQqqvrN6TtfwAndFl2c5XNQwj9QNIM3v0zPd129E1fSlpaWpZo/8i4hhBC9+r6LZnHljyR1NduuO33Stod+Gh54PQQQv+yvXV/t2FJTLrtMl5c1N7fzehXN3+22x4gIYQw6NXbbeAs0ogDB9E5LM8jwFGNaFQIIYQQQgiV1Pt8al9gE9uvSfongO3/lbR+45oWQgCQdBKwuu3jKqybCOxpe7+l3a5CG0YATwJthcWfzN2NQgghhD5Vb/D6VtdtJa0DxB+nMChIGmp7UX+3Y6CRVP698KLtUTU3DiGEEPpAvcHrNcClko4FkPQB4Gxy+cow+Ei6glSqtImUdZtE+j75me0b8zZ7At+wvaukzYBLgNWAWcAmwCm2b6py/F1I32MPANuTuqt80fZjef3xwMF58weBr9p+tcqxPgOcWli0GfB52zdKmgAcTfpZeAk4yrZzRnM88ArwEWC8pHWB04ChwHPAEbaflCRSEYBV87pW22dWacuqpCIBmwNvA7a9fz72VaTCASsDN9v+VoX9VwJ+BuxGGqpuZmHdUODHwNi86Dbg+GpBt6QLgTm2f5rnm4GpwMak/u3HACvlzY+z/du83TzSz/5upCIFP6x0/BBCCKER6u3zegJpAPM5pGo+TwDPACc3qF1h4DvG9ta2tyD1fz6eFMBNKGxzKClgBbicFNg2k4LSMXWcY3PgPNtbkqo1fRdA0jhS4LoDabzTocD3qh3E9i22R+XM4BRgBnB7rlK1P/AJ2y2kMUwvLuy6HSloayZ9v18OHJTbcyVwRd7uaGCq7ZF524tqXNMewDDbm9keCRyRl78I7JXbMQrYWtLYCvsfAWxECsA/CWxTWHd43ner/DE6L6umlXd/vVpzSdjbge1sjwa+SGdVrrJhtrex/aXyvKQZkkqSvilpCCGEEEID1DtU1lvAscCxubvA89VqnodB4xBJB5Eyc6sBfyL9M3OWpOF5m53zdsNIL/xdCWB7Rq4O1R3bLmcWpwN75elPAVfbfhlA0vnUMeawpD2ArwM72X5T0l7ASOCBlDxlCItXjZtm+895eltgtu1H8/wlpCpXa5AqTf0kZ1V/lz+qmQ1sKmkKcA9QHnJuKHCGpB1yO9YlBaK3ddl/V+BS228Db0v6JfDxwn1pzT+vSLqE1F/9vyo1xPY0SWtI2gJ4jJRt3T6v3hi4KvdrfxtYV9K6thfk9ZcVDvU3YAPbz0p6Pyl7+wK5+lYIIYTQl+rNvCJpVUlbkh6hbi9ph/yHNgwyOWN5FDA2Z16/C6ycqzjdCByYP260/Vph157+w/NmYXoR9XdzqdTmkcB5wD62y5XhhgAXl7OyOXO6YWG3it0QurJ9HbAT8Gfg26QMbbVtnyJllO8kBZuzJa1MCqrXArbNmd0bSN0HGu1SUhGSccBjtv+Sl18F/Nz25qQs7sIu7Xnn3thut/1snn6WlJHesfFNDyGEMBjVFbxKOgRYANwN/KrwEX1eB6c1Sf1D/5HHAJ5UWNdKCoYmkrsM5AzpI6TMHpK2YvHypj11F/CFnDUcAkwmBYMV5ezhdcB4238qrPoNKTO8Qd5uqKRqo8xPB0ZK+lienwDMtP2KpE2ABbZbgR+w+KP8rm3ZAFhk+wbS04x1gPeR7unfckZ4fWCfKoe4GzhY0gqSViH9k1B2FzBB0oqSVsxtrHpfsstIX5fJdHbxILdnbp6eROrbXO2a3p/PV+7TuzepX3MIIYTQ5+rNZP0E+Jzt7v4QhsHhNtLLTH8ivTR0Hzlgy4+ih5WnC/scAlws6d9JfafnkALgHrN9a34KcH9eNAM4pcYuk0lB4pTcPQDgWNu/k/QdYGp+2Wkl0ktn76rxafs5SQcDV+Y37J8j3QNI/WYPkvQWKbt8TI22bAGcntsxFDjN9jOSzgGukdQGPA38tsr+5wNbkh7zP096We1fCus2ofMlrtuBC2q0BdvzJT0K7EL+5yL7GnCDpBdIX+9aI4t8HDhZ0iJgReAmIIqXhBBCaIghHR3dP8mVNB/YOPezC6HHJK0OvGa7I488cA8g2y/0b8tCfyuVSiOAuc3NzTQ1VU3wDgoDqTxsqVRa4nK3A0Vcy8CzvFwHxLU0Qnt7O21tbQAbtbS0zOu6vt4+r98D/lPS2n3YtjC47ADMyi9qXQ0cFoFrWN6USu9K2vfIQAlcQwhhIKv3N2X5TfKjC49dhwAdtoc2omFh+WL7DuCOrsslTQU27LJ4vu29e3oOSScCn62wavfyC0VLywBry3mkYb+KFtreemm2YzCoJ2MxkLKrIYSwLKr3N+jlpBc7fgW80bjmhMGmN0FqjWOdzAAZe3iAteXI/m5DPSbddhUvLnyrv5vRcDd/rtbQuyGEELpTb/A6HDgxxnYNIYQQQgj9qd7g9RJSRaPLutswhEaSNAvY3vaAfAKQy8r+ocuQXI0835nA54ARwBa22wrrVgbOIo0n+yZwv+1epf0ktQIzbJ+bS/eulLuClNdfQSqg8AFgjWqlekMIIYQlVW/wug3wlTys0N+LK2x/os9bFZYrkobaXtQXx8olXgeyiaQhrBoavEpawfZCUjGDnwL/U2Gzn5CC1o/mUR7+pcI2vbELsDqL92G+iDRu7d8r7RBCCCH0lXqD1wvoZrzIUL+cpRJp4PcnSYPAXwP8zPaNeZs9gW/Y3jUPLXUJqQzrLNJYnqfYvqnK8XcBzgYeIJX77AC+aPuxvP54UiYd0jihX62WKZP0GeDUwqLNgM/bvlHSBOBo0vfRS8BRtp2zj+OBV0gV2cZLWhc4jTS26XPAEbafVHoDsBVYNa9rtX1mjXvXQc7sSZoH/BL4JLA+qbrV+0kD978PmGT7PkkjSGPBXgp8mvSy4dG2KwV85fMMA/6TNKbqyqSSr1+3vUjSPfm+bQ+sB/za9rclHQpsDZwj6RTgONt3VTj2eNK4yfvm+RWA+aSqVKsDPyd9rVcGzrd9dt6ulVTpSsAawKjyWLqFFynL51idNLbuBuXuPrZrBpbF7Gql+bxsC+BI4D2SymV6T7d9d6V2hBBCCH2trqGybF9a7aPRDVxOHWN761xa9RHgeFIAN6GwzaF0Vjy6nBTYNpOC0jF1nGNz4LxcavTXpBKuSBpHClx3IA2YP5Q0FFpFtm8pl08FppCCwNtzidj9gU/YbgHOAC4u7LodKXhrBp7J13BQbs+VpBKikILfqbk0azMpg9cTTba3Jz06vwB42/Y2wAksHnQPB2bn838VuCpXB6vmP4F787FGkYLiYiWxDYFPAKOByZI+YvsS0v35t3zP3hW4ZtcDOxWGnhsHPG57LjAP+JTtrUhPPA6XtGlh31GksrzdZaA3JhUW+L6kGZLukfTxbvbplu05pDK7l+VrPH1JjxlCCCH0RN3jteRHjtsAa5MyVwDYvrjqTqGaQyQdRKrotBqdQ5GdJWl43mbnvN0woJkU8GF7Rh4rtTu2Xa60NB3YK0+Xs2UvA0g6n/TYuSZJewBfB3bKJUz3AkYCD+Rs2xBgrcIu02z/OU9vSwocH83zlwA/l7QGqTrXT3JZ0d/lj574Vf78ECl7W54vkTLUZW+RsrTYvkfSG6QMZrV7uTewjaRv5PlVSZWvyq6x/U/gJUmPkYLFJ+ppsO3XJd1AyhCfQ+pq0Fo4z39JGgn8k5TZHUmqqAVwre3X6jjNUODDpBK235S0LfAbSZuUv/YhhBDCsqiu4FXSv5L+8D9Byug9QgqoprF4ti10I2csjwJ2yCVHDwQOzwHNjXTWqr/R9mvlUqukR/898WZhehE9+EelQptHkrJte9h+Pi8eAlxs+8Qqu9X1wo7t6yTdD+xOeuw/ic6yq/V4Mx9nUQ6iy9e9RNdMur5/tf1UrfMuwblagZ/mLiQ709mN41RgATDR9kJJd5C6D5TV+yLUfFIXg6sAbD8g6Xngo6TscCULWfxpzMpVtgshhBD6Tb0Vtk4BDrU9mlTiczRwOBVqwIdurUnqH/qP/Ni6+Ci6lZSFm0juMpCzZI+Q685L2or0uL+37gK+IGkNSUOAycCd1TaWtD5wHTC+yxv0vyFlhjfI2w2VVG2E9unASEkfy/MTSBnBVyRtAiyw3Qr8gJTdb4SVyP8Y5H8gVgEer7H9VODbkobmfdaWtFEd53kZeG93G+W+qsNI/YBvsP16XrUm8NccuDYDO9VxzkrHf56Uxf50bv9HSV0fnqyx25PkLimSPkAaPaCSuq4xhBBCaIR6g9cNbV/TZdmlpBdCQs/cBvyZ1FXgXtLjbmCxgGZY+UWc7BDga5LmAMcBc0gBcI/ZvpWURb8/HwfSPyfVTAbWAaZImpU/drV9H/AdYKqk2UAbsE+Vcz5Hyixembs8jKczu7o/MEfSTOBnwDG9ua46/AMYlc//c+AA27VGxP8aKaM6O9/320gvhXXnfODEfJ8+1c22lwKH0dllANLX4rDczpNI3SqqknSOpKeBDYC7JD1SWH0kcEJu/9XAwbZfrHG4C4ANJD0K/Bfphb9K/hsYk6/x27kd1+d2AFjS7bXaHUIIIfTWkI6O7p9GS3oS2NH233OQcTRpOKDptofX3jssqfzm+Gt5uKPNgHsA2X6hf1u2bCiPNmB77e62DUtfqVQaAcxtbm6mqanWO3TLh2WlPGypVKqr3O2yIK5l4FlergPiWhqhvb2dtrY2gI1aWlrmdV1fb+b1AqD8pvJZpMeRs0kZrNB4OwCzcjbuauCwCFxDGHhKpe57Ui0LgWsIIQxkdf0Wtf3jwvRleZzL1crjhobGypWM7ui6XNJU0pBNRfNt793Tc0g6EfhshVW72362p8dbEn3dFtvzSKNkdD3PKBZ/ZF92ru0Le3qeCsefDHylwqqJtmct6fF7Q9L7qfC9BFxv++Sl3Z7lTXcZi2Ul6xpCCANZj3+LSnoPecggSe/JwwWFftCbILXGsU4mDdfV75ZWW3IA2bCKXTkAXuIguC/l4H9AVin70m3X8uLCt/u7GQ110+cm9ncTQghhmVfvUFlbkQaoL1cbgjSUUAdpPMkQQgghhBAart7M66WkoZEmAa93s20IoY/lkrtnkypwAcwtl5ftxbF2Ac60vbWkNUnjDP+ksP47wBdIoy0MAU6z/auKB+vcZy9SlbUVSEPoHVoY/iuEEELoM/UGrx8CvlOukR5CWHoklX9O77K9Xx8ffk3gW8BPCsvOtf2jfO71gMcl3VHtJcE8GsYFpOprT0i6kDSk24DohhJCCGH5Um/w+t+kCkgxdmN4F0kdpDFf9wWGA9/MlbNGUBiiqjhfniYFPWNJRQMOIo1Nui3wBrCP7QU1zjuPNGbtJ0ljsH6bNBD/gcD7gEl5PFokfSa3cWVSqdhjbU9XKsvVSirLOhRotX2mpJWAH5GqXzWRysgeZftVSYcDxwLtpBE79rddseCBpCeA/WzPzvNfAVpsHyrpzHz8lUhDz02y/ZfCvWkFdiONHftmhcNXVcyuVpovmAKsKWkW8LrtHWwXxxBendQ9qNbIJONIX9dyedzzSE9rIngNIYTQ5+odKmtl4L8l3SHpsuJHIxsXlikv2x5DKkZwTp37DAem5YptFwG/BabY3pL06LnSm/pdNdneHvgcKRB+2/Y2wAmkUqtI2hj4HjDOdgup8MKv8/5HA1Ntj7TdnNsBKRv5ku1tbI8EngH+Pa87A9jN9ihSRar5Ndp3KamiWNmh5OppwOm2x+TjXwX8uLDdcOBB21vZPi8v2zkXBrhP0v+r497U48vAi7ZH2d6hvFDSkZIeB2aSuhX8o8YxNgT+UpifD3ywj9oXQgghLKbezOuj+SOEaq7On6cD60laudbG2au2b87TDwFPF4aQKpFLm3aj3BfzIVL2tDxfAjbJ03sAGwP3pUQrACtI+hdSBaufSFqVNH7x7/L6vYFhksqP6ZtIYxsD3A1cKuk3wM22n6rRvsuAByR9C9iU9Jj+f/K6cZK+TMpudv1ZfJPOABvgJuBXtt+QNBq4NVc6a8hwdTlgPk/SFsAVku7qJoANIYQQlop6x3n9QaMbEpZ5bwLYXpQDxBWAhSye3e8a0LYXphex+KPxRdT3/dn1vOVjFPcfAtxmu1I54+sk3U/qFvNt0kuJ4/M+R9u+u8I+nyVlXHcDfifpyFx2911sz88lW8cBu5C6JXRI+hCp4McY23Ml7QBcWdj1tWIfc9vPF6ZnSvo9sA1QLXjt7t7XxfYcSc/ktl9XZbP5wK6F+Q2Bv/bmfCGEEEJ36u02EEJvLABWlFTOgB7YT+24AxgrafPyAklj8udNgAW2W4EfkAJCgKnA1yWtkrdbQ9Km+eWpD9v+o+3T87FHd3P+VlJXhQNI3QgAhpH63i7IYycfWesAktYvTH8I2I7UD7eap4APS1pL0pB87kpeBlYtvBRGLkFcnt6IdH21nrzcBoyR9JE8fySLZ41DCCGEPhPBa2gY2wuBY4A7Jf2RlA3tj3Y8QcqmXiRptqTHgCPy6v2BOZJmAj/L7QU4ndRN4MFclnca6bH/UKBV0hxJs4EPAL/opgnXkzKXj9qen9s0B7iGFBQ+AMzt5hhflvRIPudU4ATbM2tc8zPAf5C6T/wB+FuV7f4PuIJ0D/6QF5+UzzULuBb4t1rdE2y/AhwO3CTpSeC9wJndXE8IIYTQK0M6OmL0qxBC/ymVSiOAuc3NzTQ1NfV3cxpqWSoPWyqVui13u6yIaxl4lpfrgLiWRmhvb6etrQ1go5aWlnld10fmNYQQ+kipVKq5flkJXEMIYSCrtzzsEDr77K1te0tJnwDWtR1920LDSJpM5SGzJhZGJuh3kmbw7p+n6bZr9mXtg/OeR+r/WrSwwniufXGuqaSXsYrm2967r8+1rKqVsXhr0SJWGhrVtEMIYUnVmwY4mTRs0dmkAcgBnia9LR3Ba2gY2xcCF/Z3O7rTiGCxzvM2NDjucq6GBqlfuvVGXlz4diNP0a9u2u+g/m5CCCEsF+rtNjAR2NP21aRqO5BeMPlwIxoVQgghhBBCJfVmXocCr+bpcvC6emFZCDVJOglY3fZxFdZNJP1ztF/Xdf2l2F5JRwKr2D6rn9t0AmnUhLeBV4AjbD9SY/t7SCVhb2pwu4bD/2fv3OO1LKu8/yUQNAMPpOPkIUxtGW6B2J6wMDHz9HrIUlNEQV7MtMxMTVPzLXPSGZ3GUXHM41YTycQRNE+ZoTGJyZaj5s80GcccR3ESFAWEeP9Y65GbzfM8+9kn9t6wvp/P/nAfrvs63Pe92ete17rWj9txIYhlwJ+ib292ZLtJkiTJ+kmtntcHgZ+aWR/4MAb2x8B9HdWxpP0xswy4awWSrutMw9XMepnZEDy9V0lO9i5cprYrsBL4J0kmaVfgJTzVWJIkSZK0O7V6Xs/EE60vBDbAPa6PAOUUi7oNZnYHYLj054u4utIvgaslTY4yhwJnSRoRydtvATYGZuHyo5dU8myZ2b54nPBTwDD8j/yxpZyZZnYucEIUfxo4XVJZb7aZHQL8pHBoIHC0pMlmNho4DX+eC4FTJSk8mqNwL91OwCgz2wq4FPemv4l7yF40l6dqwCVWe+JKUGVzdYaU6q3ALrgXUJKOibrvxBPwb4hLp36vzPW98Zyq+wELgJmFcz2BfwQOikMPAedKKpsj1swGADOAG+KajYDj8UT5ewLvA0dIej3Knwt8Ne7VX4CTJb1uZpsANwF1uLjCfwH/E9f8kFVe2Jb270ZgrqR/jf06PE/rDvgCyDOA3lH8bEm/iXLzccnd/YC5wFX4795HgcV4LtVXy7XZhC+Y2XnAJ4C7JJ0X9Z8FHBv3YQn+zsyKc8Nww7hv1HGOpEfiHbkS+Hj0+UpJt0Su2KmFNqcDp9bQtyRJkiRpMc16XuOP9VG4OtJ2+MrmHSQdGcnJuzNnSNotvEXPAufiBtzoQpmTcIMVfGr0akl1+B/x3WtoYxfgOkmDcG/ZhQBmdjBuuO4N7IobjD+oVImkByQNkTQEGI8bbA+b2XA80f4+kupxo+PmwqV74UZRHfBajOH46M8EPEE9uPE7RdLgKHtTlTEdCPSTNDC8gKWE/28Dh0U/hgC7mdlBZa4/BdgeN8C/yCpVK/Bk90OAofHz2ThWjf7ANEmfjX7/BhgfY2wkshWY2SjcaNxL0lDgATyRP8BFwCJJO+Pv+xcqtNXS/jWw5vvUENKvD0dfPosbkrc2ubafpD0k/V9Js4GfAvPN7C9R/vtV2i2xHbBP9HNcQQXrNkm7R9s/IBZimtnmwL8D34tnOxQXauiFvy9nStod+DxwnpntXGws1MJOxQ30JEmSJGl3mvW8hmb8TyXdjHto3uj4bq01TjSz43Ev0sbAC3hmhX+JOD5wI+ZEM+uHe+UmAEiaEcpLzaGCEtJ04LDY3h+YKGkRgJldD/xrc5WZ2YHAd4HhkpaY2WHAYOApd4zRA9iscMk0SS/F9p7AbEklqc9bgGvNrC/wBPBP4VX9bfxUYjbwGTMbj3vcfhXHewKXm9ne0Y+tcEPvoSbXjwBulfQB8IGZ/Rw3hkr3pUHSshjvLcCRwL9V6c+7kkp9eAZ4tZBGqxHPlAFwOLAb8Ezcq5KnutSn0wEkLTCzeyq01aL+SZoW0rK7An/Eva3D4vQOwJ0h/foBsJWZbVXyEgO3leoJSdgjgB0l/beZnYMbu4dWuS8Av5T0N2BhKIvtgMek1kcM7ebA34BPR/lhuBLY76P/K4C/xqzDZ4CJce/AZyw+AzxfaO9qfGbmmmb6lSRJkiStotawgfvM7DBJ60yMa3gsTwX2lvSmmY0Evi7pPTObjHuaASZLWhzGK6xasFYrSwrbK6j9npfr82DcQ3agpAVxuAdws6SLKlxW06I6SZPM7EngAOA8PIRiVIWyfzazXXCv6cHAT8I4+y5uOO8ZhvX1ePhAR7O0sL2Cyve8Bx7mUfRMrw1uxTN2TAX+KOk/4/ideEjKveGxfI/V71fx2R2Nhx+UZF5vA35YQ9tr3IsI27gb99Y/Y2afwEMoqtEDWBCe/7KY2RV4eMphYTAnSZIkSbtT64KtDYG7zWyqmd1uZreVfjqycx3MprjX7a1YiDa2cK4BNzbGECED4SF9FvecYWZD8en+1vIo8LXwypVEIH5dqXB45yYBoyS9UDh1H+4Z3ibK9TSzSpnSpwODC1O9o4GZkt4xsx2B1yU1AD9i9an8pn3ZBlgh6V48HnoL3IO3KfDfYbhujXsKy/EYcEIsRNqIVR8K4PdltJltYGYbRB8r3pcWMgU4zcw2i3H0iQ+CUp9OiuP9cW9qOVrTv9vw92Ycq0JQwO/Xy7E9FvdkVuJl4PNmtnHsHwLMa6bdSmyIG/T/FfunFc49CQyMuNfS+7QZIOA9MyvFaGNmO5c+6szsJ0A98GVJxY+JJEmSJGlXajVe5+GLhX6LL2x6qfDTXXkI7/8LwOP4dDPgU734oqN+sV3iROA7ZjYXOBtfSLOQViDpQeDnuLEwNw5fUuWScbiRON7MZsXPCElPABcAU8xsNv6syhqNkbroBGBChDyMYpV39RhgrpnNxKd+z6jSl12BJ6O9PwCXSnoNX1T0OTObx6rY03JcD7yCT6M/hi9WK56bgy/imhnbN1TpS81Iuh2P8X08xt8IfC5O/xjYzMyexz8SnqjS9xb1T9IrwHPAvkAxHOE7wL1m9gyeM/mtKtXcg7+zjXHfxxDGdkuJD7GL8FjWRnwBWOnc/wJfwbOLlO5RvaTleMjLsWY2x8yeBa4FeocX/vv4orDfx7v5763pW5IkSZI0R4+VK1s6C77+YmYfAxZLWhkxgFMBk/TXzu1ZknRfGhsbBwAv19XV0adPNedz96a7ycM2NjZWlbvtTuRYuh7ryjggx9IRLF26lHnz5gFsX19fP7/p+ZriL81sv0rnJD3W6t51P/bGFyT1iP2T03BNkqREtf/4u5PhmiRJ0pWpdfFQ07RJW+Ar9F9lPZKIlfQInt92NcxsCp6SqMgraoUWvJldhE/bNuUASWs100MX68t1eNqvIssl7bY2+1GJzuqfrZn/t8T5kh7oyLaTNalkuHY3r2uSJElXpibjVdL2xf3I/Xohnvx+vac1RmqVui7G03V1Ol2sL9/o7D5Uo7P6FwbqOmGk/t8Hf8Xby5d3djc6hPuPOqazu5AkSbLOUOuCrdWI3I//AKyhnpQkSZIkSZIkHUWrc47iid8zl2PS7pjZLGCYpPc7uy8twVwO+De4cts1cWw8MBz/XfkAOK8kAdvBfbkRF4L4nZl9GXhN0h/aUN984FBJ88xlh39fStkWadF+jqtx/amrhHIkSZIk6ya1Ltj6L1ZPzv9RPFfkNzuiU0n3w8x6hke+zVRLhN/VKI07VMr+EXiwSZHzJS2MsoOB35jZFiEP22FIGlfY/TIuJ9xq47UJY4AFeJo5cDGFi/D0cj9qpzaSJEmSpCy1el6bKi0tBl4oSZsma2JmdwCGJ55/EU9C/0vgakmTo8yhuMLSiEi9dQsuUzsL2BFXg7WFjGwAACAASURBVLq/Qv37AlcCT+GSniuBYyX9Mc6fi+d0Bc+jerqksmpbZRb9DASOljTZzEbjSexLUqqnSlJ430bhcc87AaPMbCvgUlwm9k3gFEkvmuuJNuAfPT1xedUrqty7lUBfSe+Gx+/nuJrX1rj615a4sMHmwFhJT5jZANxAuxWfFegBnCbpdxXa2A435rYNmVrM7G5c9OEOXPK2P7BRlDtF0rJy48af10+By2ki11oyXINNqEGhLZ7dV/F7/hc8q8XrZvZDYGfcSPw0noP1MuCfgU8C90g6J+qYClyBe3sPB/Y3s3HATyWVFRcpelfL7cexk3CJ3avM7BLgbEmPAr+LdzJJkiRJOpRaY153l/R44WeGpEVm9t0O7V335gxJu0naFVfmOhc34EYXypzEKsWl23HDtg43SnevoY1dgOskDQLuwhfRYWYH44br3rigQE/gB5UqkfSApCHh8RyPG4EPh4TuMbiMaD1unBWlVffCjZc64LUYw/HRnwm4EQhu/E6RNDjKNs1e0Rx9JA3DDbobgA8k7QGcz+pGd39gdrR/OnBnqKeVG/MruKDDwfChqta+uGzqCmBkTH/X4fevqMD24bglzYr7vYmku8u1ZWYXm9lLwL34R0FFA9bMRgE7AHtJGoovxvrnQpF6XK3LcEP2shjDIFz5a6cm43wYVxa7LJ5xm1TxJN2Cvx/fjvoebUt9SZIkSdJSajVeL6pw/ML26sg6yIlm1hhqXCOBIbhK0nAz6x/G0heASSGxWYcbfEiagSs3NYckzYzt6bjRA7A/MFHSojCUro9jVTGzA4HvAodLWoIrKg0Gnoo41MuAbQuXTJNUUlnbEzccn4v9W4AhMZ3+BDDOzH4cOYPfrmFsRX4R/z6De29L+424h7rEMtxLi6SpwPu4kVeJBnwKHPwZTZG0GP+9ODvGPAfYD39+JT4ct5ltit+Xb1VqRNJFknYAjgL+ycx6V+nT4fizeiba/yYwoHD+YUkLI0RjDvBrSUuj32LVO5AkSZIk6yRVwwYK4gQ9zWwEPhVb4lNkqqyyhMfyVGBvSW+a2Ujg65LeM7PJuKEEMFnS4pI+PDVMKTdhSWF7BW1YgBfxmNcBB0paEId7ADdLqvTxUjYMoSmSJpnZk8AB+LT/WNYMRanGkqhnhUcgfDjuNo0Z/5j4l/iQGIPLtYI/n88DwyW9Y2bn49P0JYrjrgP+HvhD9O3jwGFmtnmkGvsQSY/Fs94VN7zL0QMPF7m5wvmmz7zd3gFgOat/0G7YhrqSJEmSpENozvN6U/xsiE8Xl/ZvxA2Q0zu0d92XTfH40Ldi2ro45dyAG0pjiJCBiB1+Fp8OxsyG4gZOa3kU+JqZ9Q01sHHArysVjtXik4BRpRXkwX24B3mbKNfTzCrpxk0HBpvZzrE/GpgZxt+OwOuSGvAFPXu0YWzV6E18GMQHxEbA85UKS3oPmIzH6fYrxMduCiyIvm/Cqo+NcnVMk7SlpAGSBuBhB/9P0sVm1qNwPzCz3fB43T9XGcMU4DQz2yyu6RMfFm1hER5v2xwvEuEqZvZF4O/aWF+SJEmStDtVvTQlcQIzu03SiWunS+sED+GexRfwVdlPEAabpGklT6ukaYVrTgRuNrPvA3Pjp7jYp2YkPWhmg4An49AM4JIql4zDVdPGh/cQ4ExJvzWzC4ApIUzRG190tobXMDzMJwATzKwXvmCr5F09BjjezJbh3uUzWjOuGngLD1X4Hu7BPE7SsmauaQB+x+oxwbcBR5jZ88AbcX6jVvSnB3C9mW2OezXfB45RFUlhSbeb2ceBx+NZfAS4FpjdivZL3A40mNnRVFmwhd+DW83sdOAx4JUK5a4H/tnMzgHOBn4L/Ce+OHETM3sVuFHSD9vQ5yRJkiQpS4+VKzs0Y09SI2b2MWCxpJWReWAqYNUMnWQVpWwDkj7e2X1JWkZjY+MA4OW6ujr69Cm7vq7b0x3lYRsbGyvK3XY3cixdj3VlHJBj6QiWLl3KvHnzALavr6+f3/R8rXle+wE/xBcYfZxC7Kuk7dqjowl7A5fHND94eqQ0XJOkG1HpP/7uZrgmSZJ0ZWpd3HEtsA2uc/9zfDr4HDxOMmkHJD0CPNL0uJlNAZp+ILwi6fCWtmFmFwFfKXPqAElvtLS+ttDefZE0H/+watrOEDw0oCnXSLqxpe20F2Y2gzV//6ZL+kYHt9s0p2+J8yU90JFtJ0mSJEl7UKvxegDwGUlvmdmKSF4/A1/Q8y8d172kNUZqlbouxj9AOp211RdJs1g9zVWXQJ0koRoGahqpHUTR69odQwWSJEm6A7Uarx9h1eKhd2MF9n+zeo7NJEmSVjPuwUd4e3m7KAx3Ce476sud3YUkSZJ1klpFCmbj8a7gK6+vBf6NVdrmSZKsA5jZLDNrTWaFJEmSJFkr1Gq8ngzMj+0z8JQ/m+LpnZKk3YnUXEkNtOe9CsnX99urviRJkiRpb2oKG5D058L2G3he0GQtYmZ34FKnffBk8mPxnKtXS5ocZQ4FzpI0ItJt3QJsDMzCQzwukXR/hfr3Ba4EngKG4flYj5X0xzh/LnBCFH8aOF1SWYWtMouCBgJHR6z0aOA0/N1bCJwqSWY2Bl8I+A6wEzDKzLbCBQR64nljT5H0onkC1AZcKrYn0CDpiir3biVwAXAk0B84R9KkOHdQuTaq1DU1xj8M+ARwl6Tz4tyOwM/wnLnL8UVQD1XrQxe9V30lvWtm8/EFml8EtsbV0bbERRs2B8ZKeqKUpgy4FfgSno3ktILoQ5IkSZK0GzV5XkMp6GQze8zM5sSxfczsmI7tXlLgDEm7SdoVV+M6FzdKRhfKnESoduGJ6a+WVIcbpbvX0MYuwHWSBgF3ARcCmNnBuOG6N6781ZPVk/qvhqQHwoM3BBiPGzYPh+rVMcA+kuqBy3HlthJ7AWdHn1+LMRwf/ZkA3BHlTgOmSBocZW+qYWyLJO0e47gqxrVllTaqsR2wD/BZYJyZ7RTH7wAmRF2jgJ+b2RbV+tBF71WRPpKGAV8FbgA+kLQHcD6rG939gdnR/unAnaEulyRJkiTtSq1hAxcD/xdX1imlbXoVN6CStcOJZtZoZnNxz9cQ4B5guJn1N7P+eFzypMjLW4cbMUiaAcypoQ1Jmhnb04EdYnt/YKKkRZJW4u/B/s1VZmYHAt8FDpe0BDgMGAw8ZWazgMuAbQuXTJP0UmzviRtDz8X+Lbh6Vl9csWycmf3YzPYD3q5hbBML4/qEmW3YTBvV+KWkv0laCPwR2CGuGcIqyd/ncI/3Xs30Aehy96rIL+LfZ3DvbWm/kdUXbC7DvbRImoqHFhlJkiRJ0s7UaryOAQ6VNBGfTgZ4GfhUR3QqWZ3wwp0KHBSe1wuBDSW9B0zGjdmRwGRJiwuXtlQ+bUlhewW1Z6Mo1+fBwHXAEZIWxOEewM0lT2N4A4s5bMuGITQlpvyHAy/hU9m313DZkri2tJy91WOj9fepbB+64L2q1ufS2Nv0fiRJkiRJa6nVeO3Jqj+WJYPoY9T4BzRpM5viMY9vxVTs2MK5BvzjYgyrvH6L8NCC4wDMbCg+3d9aHgW+ZmZ9QwFsHPDrSoXNbGtcwGKUpGJGivtwD/I2Ua6nmVXSoZsODDaznWN/NDBT0jsRW/q6pAbgR8AerRxXxTZaWlFcMyvqwMw+g3tOp1e7rhvdq+bojX9AlT62NgKe76C2kiRJkvWYWj0nDwA/NbMzwWNggR/jf2CTjuchPIbyBWABPhW8B4CkaREmgKRphWtOBG42s+8Dc+NnIa1A0oNmNgh4Mg7NAC6pcsk4fNHSeF8vBMCZkn5rZhcAU2KFfG980VljmTbfNLMTgAlm1gtfhDQqTh8DHG9my/CPqTNaOa5qbbSG44Gfxe/JcuAESW82c023uFc18BYeqvA93Gt8nKRlHdRWkiRJsh7TY+XK5meWwzi6FTgY2ACfOnwEOLE1Xqqk4zGzjwGLJa2MzANTAZP0187tWbKuUco2IGkNed5aaGxsHAC8XFdXR58+684ar+6usNXY2LiaYlh3JsfS9VhXxgE5lo5g6dKlzJs3D2D7+vr6+U3PVw0biPQ7xEKdI/HFWnsBO0g6Mg3XLs3ewKzIDjERODkN1yTpWBobVznGu7PhmiRJ0pVpLmzgBaBfYf86SV/pwP4k7YSkR3Dv+GqY2RRWZYwo8Yqkw1vahpldBJR7Hw6IfMBrjfbsS5ncqyXOl/RAa/rXlWjv5yZpPtAqr2uSJEmStJTmjNceTfb37aB+JGuJ1hipVeq6GE+j1um0Z1/CQO32RmolutJzW5dYtmJF84WSJEmSNtOc8drSVEtJkiSt4uQHf8vC5X/r7G60mslHHdLZXUiSJFkvaM547WVmI1jlgW26j6THOqpzSecTCfKHaT3Uuw/p2H/EP+I2AO4FLoxFcB8DrsVVtjYAbqwmu9rdMbMfAh+TdLaZDQE+LemuwvkrcBWuAcCukuZ1SkeTJEmSdZ7mjNc3WF2S8q0m+ytJoYIuh5n1LCSVbxMhW7reESmnpgFDJa0wsw2A/wCeAqbg8qjLgEG48tTvzWyapKp5XTsbM/sIsDKU0lrLEOBQXEK4xL3AvwK/a0O9SZIkSdIsVY1XSQPWUj86FDO7A5eq7AO8iCf5/yVwtaTJUeZQ4CxJIyK11C3Axnji+R2BSyTdX6H+fYErccNmGG7UHyvpj3H+XFzPHuBp4HRJZQUeyiwWGggcLWmymY3Gtep74TlbT5UkMxuD5/V8B9gJGBWZIi7FBSbeBE6R9KJ5MtEG3ODqCTRU8xia2Uqgr6R3zWw+LgH6RWBrXLFpSzw5/ebAWElPlFIn4enVvoR76k+TVNGwMbO/w1Wmdojyl0u6Lc7NxzMmfAnYBLhS0jVV6voTcJSk2bH/LaBe0knhIfwCnjd1QfT5Pwt9bgD2A66XdF2h2g3jmtK89uC4dyuBxWb2OJ7ntaLxambDgMuBkvzsOZIeMbPdgavw920x8G1JTxf6dANwEJ74/3jgG7gk7Pu4KtfrUf+5uPezF/AXPMPE6+E13SXu3Xb4O7pG5omid7Xcfhzrj8fL9guv/BOSvl3KMVzIVZskSZIkHUKtClvdnTMk7RbSqs8C5+JGyuhCmZMIhSpcQvNqSXW4Ubp7DW3sgmdjGIR7pC4EMLODccN1b1zlqifwg0qVSHqgJAkKjMeNl4dDtegYYB9J9bgRVPSC7wWcHX1+LcZwfPRnAnBHlDsNmBJyo3XATTWMrUgfScNwI+kG4ANJe+CeyKLR3R+YHe2fDtwZ6mCVuAqYF+UPAC4zs7rC+S1j3J8Dzg/RhErcSuVne5mk3SUNBu7EwwKKfX5a0tCS4Wpmu0W6sTeAx4BfRdlG4Cgz28DMPg4cCHyyUofMbHPg34HvRdtDgafNrDeusHVhjP0HwKQ4XurTNEmfxZ/Vb4DxUbYR+FbUPwo3/PeSNBRfcPbPhS7sCYyUtHNbUqZJegu4CHg03tNvt7auJEmSJGkN64vxeqKZNZrZXNxLOAS4BxhuZv3Dm/QF3GjoB9ThBh+SZgBzamhDkmbG9nTckADYH5gYuXJXAtfHsaqY2YHAd4HDJS0BDsO9fU+Fx+syYNvCJdMkvRTbe+KG43OxfwuuftQXV+caZ2Y/NrP9gLdrGFuRX8S/z+De29J+I+6hLrEM99IiaSruJazmltsf+FmU/2/c+BpROH9TnPsf3IDct0pdtwHHmVkvM9sVl9cteX0PNrPpZjYPOBt/F0osYfWpcCTNCENxW6AeGB6nLsM9tzNwr/BUXFWrEsOA5yT9PupdEUakAcsk/SaOP4rfu9K9eldSyWB+BnhV0qzYL97zw/F7+Ey8H9/E409LPCBpQZX+JUmSJEm3oFZ52G5LeCxPBfYOGc2RwNclvWdmkwk9dmCypMUlqVVanmlhSWF7BW24t2Y2GJ9CP7BgcPQAbpZ0UYXLyoYhNEXSJDN7EvdunoeHULREEnVJ1LMipohL427TmNsTSa+Y2bO4Ity+xPS+mX0S+Bdgd0kvm9nexEdKsLhSLKikBWb2IHA0PlX+Hm4gAmBm1wLPlbu2jSwtbK+g8nvWAw9tKXrji9Tyfixn9Q/aDWvtZJIkSZKsLdYHz+umeHzoWzFtPbZwrgEYEz+3gKuJ4aEFxwGY2VB8ur+1PAp8zcz6mlkPXMv+15UKm9nW+DTyKEkvFE7dh3uQt4lyPc2skobbdGCwme0c+6OBmZLeMbMdgdclNQA/AvZow9iq0Zv4MIgPiI2A56uUfxQ4OcpvBRyCT9OXGBPntohzv22m/Qb8Xh+HhxGAC24sA16PhUvfqFaBmX06ymFmG+PG8NzY72dmG8X2IOBIPPtAJZ4EBkbca+n5bQYI6B1ZPAhv+AZxvCVMAU6LOjGzPvER1BJeBOrN7CPhpT+0QrlFePxskiRJkqx11gfj9SHgJVwt7HF86hWAWGTSD+hXWnASnAh8J8IMzsYNloWtaVzSg/j0+ZNRD8AlVS4ZB2wBjDezWfEzQtITwAXAFDObDcwDjqjQ5pt4nO2EiNccxSrv6jHAXDObCVwNnNGacdXAW3iowhzcqDtO0rIq5b+NG9xzcOP+PEnPFs4vMLNG/D5eKmluuUoK3IN7XZ+T9ApAXPNL3EP6FPByM3Ucgd+r2VH+CeDGOPcpYLaZPYcbysdLeq1SRZL+F1e1+mmMsRFfRLYMjx/+SRz/B3yxWbV7Va7+2/G45scL9X+uJXXg9+x/gT/GdmOFcr8BNjaz2WZ2FYCZXWVmrwLbAI+G5ztJkiRJ2p0eK1emDkFTzHN4Lo6p5oF4PKO1ZaHL+kRplbykdpEMjWwDh2bu0HWTxsbGAcDLdXV19OlTbU1f12bZihXMnTWL+vpKEyLdi8bGxhxLF2RdGcu6Mg7IsXQES5cuZd68eQDb19fXz296vkvEKHZB9gYuj2l+8JRDabgmSVKR3j17dnYXkiRJ1gvSeC2DpEeAR5oeN7MpeJ7MIq9IOrylbZjZRfg0clMOkPRGS+trC+3dF0nzgTW8rubKTA1lLrlG0o1ljpfqG1DuuJnNYM13eLqkqrGsHUlXeq6V6Ir3LUmSJElqJY3XFtAaI7VKXRfjyd47nbXVl0jx1G6KXZJ2a6+62ouu9Fwr0RXvW3dm2YoV6XVNkiRZi6TxmiRJl+DrD/4HC5f/rfmCXYx7j2o2bXOSJEnSjqTxmnQJIrH+MEnvd3A7XwZek/SHjmxnbWFmFwBfw3O+9sAzMfwizj3KqvCNXrgK3GBJtYhuNG1nDL5o7qhYkHeApOsL58/GU53thAtrlJVSTpIkSZK2sj6kyko6CDNrt7nSkBrtUMM1+DJVctu255g6mujrNZIGhXzsIcANpVyvkvYvSA1fCDzbGsO1DAOArzc59ni0/0Q71J8kSZIkFUnPaztiZnfgsp598ITvY/G8oldLmhxlDgXOkjQi0nDdAmwMzMKlPi+p5LUys32BK/Gco8NwFbBjJf0xzp+L53cFeBo4XVJZZSUzOwT4SeHQQOBoSZPNbDRwGv5+LAROlaTwvo0C3sE9bKNCUOBSoCfwJnCKpBfN5bcacAnZnrjK1RVV7t1KoK+kdyM11s+BLwJb40pgW+KiB5sDYyU9UUrJhYsQfAn3PJ4m6XdrNMCHkruHA/ub2Tjgp8ArwFV4TtPPAheGytoZuNACwNkl+dbo223R3t8DV0i6JsQMrgH2w1Wx3pVUNs+qmW0H/AHYVtIHcexuXIjiDlz+tj8u7PCHuKfLyt3/glQswMfwd6LcR+lYoJL6VqlfYwjvarn9AuOB7cNb/qKkoyQ9HddUayJJkiRJ2kx6XtuXMyTtJmlXXKXrXNyAG10ocxKh5gXcjhu2dbhRunsNbewCXCdpEHAX7lHDzA7GDde9cUWwnsAPKlUi6YGCV248bgQ+HGpYxwD7SKoHLmd1o2cv3JirA16LMRwf/ZmAG1/gxu8USYOj7E01jK1IH0nD8AT+NwAfSNoDOJ/Vje7+wOxo/3TgzlBSKzfmh3Elqsti7LfFqV2A6+PY/cDDwF7hzTyWVQpdJT4afdsXuCzyAg8GRgADJQ2msjoVIZowD1fswsz6R11349P/I2NRVR3+HIuqcB/e/5LhambfMLPngZm49PFbxfbiA2N//Fm1B9/ExR+GlDFskyRJkqRDSeO1fTnRzBpDmWskvrL+HmC4mfUPI+ULwKTw7tXhBh+SZgC1TOlK0szYng7sENv7AxMlLZK0Erg+jlUlvJHfxeMUlwCH4YbYU+FZuwzYtnDJNEkvxfaeuOH4XOzfgqtq9cWnj8eZ2Y9D8vTtGsZW5Bfx7zO497a034h7qEssw720SJoKvI97v1vCnyQ9WdjfATfkn412twoDsMTEaG8+8FdcVerPuKzrTWZ2As3TQEje4u/KFEmL8d/Js+Pez8E9ucUMDcX7T/TjOkk744btBfGeFTkReCiU15IkSZKkW5PGazsRHstTgYPC83ohsKGk94DJuIEyEpgcRkqJlkqcLSlsr6ANoR9mNhi4DjhC0oI43AO4ueSVDc9pMbdt2TCEpkiaBAzHpXnPo+VevyVRz4riPm0ccwWajulO4FpJuwBDgeXAhk37VuyPpIW4B3ciMAh4tonB25QPP2pwI7bkjR8JfB4YHu/RtU3arnj/Q/72NdyLW+QkmgkZCJaz+v8JG1YqmCRJkiSdRRqv7cemeHzoWzFtXZzqbcANlDGEkSJpER5acByAmQ3Fp/tby6PA18ysbyiDjQN+XamwmW0NTMLjJl8onLoP9yBvE+V6mlklrbjpwGAz2zn2RwMzJb1jZjsCr0tqAH5ElUVSbaQ3bvCVPiA2Ap6vUn4RsEkzdW4KvBzbY/EY5qqY2RZ4OMHDuLG+EPhUpfKFj5pLgX6FON1NgQVxDzchxlal3YGF7e3xuN3nCsf2xsf7YHNjwOO0B5lZHzPrDVQKCajlHiZJkiRJh5ALttqPh/DFNC8AC/Bp8z0AJE2LMAEkTStccyJws5l9H5gbPwtb07ikB81sEFCa/p4BXFLlknHAFsD4wiKbMyX9NtIvTYnV7L3xRWeNZdp8M6bIJ5hZL3zB1qg4fQxwvJktw73LZ7RmXDXwFh6q8D3ca3ycpGVVyt8ONJjZ0axasNWU7wD3mtlf8ef6VpkyTdkWX+nfC/+9ehA37qvRAPyO1WOTbwOOiBjWN+L8RlXq+KGZ7QJ8gHuBv11awBecBNxW8GBXRNL0SK/1LO7BnY0vSmvKHEBmNg94PtJnnYM/4y3w+7sEj/9d1Fy7SZIkSdISeqxc2dJZ66S9iIU+iyWtDA/aVMAk/bVze9Y9KGUbkLSGFG3SfWhsbBwAvFxXV0efPs06ubscRYWtxsZG6usrTVR0L3IsXZN1ZSzryjggx9IRLF26lHnz5gFsX19fP7/p+fS8di57A5fHND/AyWm4Jkn3IqVhkyRJ1i5pvHYikh4BHml63MymANs1OfyKpMNb2oaZXQR8pcypAyS90dL62kJ79yVW+6/hdTWzIfiUfFOukXRjS9tpLV2lH01pz/crSZIkSdY2abx2QdrTiJB0MXBxe9XXFtZWXyL/6ZBmC64n/WhKGqktY9mKv9G7Z65tTZIk6Sqk8ZokSZfglIdmsHB514vB//evfr6zu5AkSZIUSOM1Sbo4VWRaS5LBV4QiV6dhZmOBM/GMB8vxzBVlZXqTJEmSpC3kXFjSrYj0XUkXwcx6hdDClcD+ITd8MfCzzu1ZkiRJsq6Sntdujpndgcuh9sGTzI/F87JeLWlylDkUOEvSiEjJdQuwMTALl1q9RNL9FerfFzdMngKG4Tlbjy3lEjWzc4GSHOrTwOmSyqpAmdkhwE8KhwYCR0uabGajgdPwd3IhcKokhddxFPAOsBMwKpSrLgV64rllT5H0onnC2gZcTrYn0CDpiir3biVwAXAk0B84R9Kkpim4ivulbeAG4CA8B+vxwDdwudz3ccWy1yu0+RHgGlz2dSnwrqTPRX7YX0U/NgL+EONaI2etmV0CHItL005tcq4lz+NCoL+kM2O/PyDgk7jU7CW4ylYv4B8kTYxyU/F3Zy/gf/F8xT2AvsD/4EILr5ZrM0mSJEnaSnpeuz9nSNotpESfBc7FDbjRhTInsUp+9HbcsK3DjdLda2hjF+A6SYOAu3DpW8zsYNxQ2htXB+vJ6gn3V0PSAyXZWWA8bgQ+HMpYxwD7SKoHLmd1OdO9gLOjz6/FGI6P/kwA7ohypwFTQtK2DriphrEtkrR7jOOqGsqDG5jTJH022vgNMD760wh8q8q1g4EReAL/wcChcXwFMDKm/+vwezm26cVmdhhwOL4QbC9g58K5Fj0PXBDh2DCcwdW8poR88TPA52OM+wNXmNlmhWs/FecPCWnhU4BnzOwV/APltCrtJkmSJEmrSeO1+3OimTWa2Vzc+BgC3AMMN7P+4U37AjApVL7qcIMPSTNwtaTmkKSZsT0d2CG29wcmSlokaSVwfRyripkdCHwXOFzSEuAw3Kh7ysxmAZfhilUlpkl6Kbb3BGZLKkmg3oIrbPXFVc3GmdmPzWw/4O0axjaxMK5PmNmGNVzzrqRfxfYzwKuRWQDceN2xyrV/BjYAbgp1shIfAc6O8c/BPbPlMhWMAH4h6d1QzSoa6C16HpJewT94DolDY1iV2msL4O5Q0XoY2Bz38JeYIGk5QLxX3wJ2l7Qd/mz/vZC/OEmSJEnajTReuzHhsTwVOCg8rxcCG0p6D5iMG7MjgcnhTSvR0iXdSwrbK2hDuImZDQauw6fWF8ThHsDNJa9seE6LeUjLTns3RdIkYDjwEnAe7qFtjiVxbUk+tRe+4Kj4u9HUoF1a2F5BC+6PpIW4J3siMAh4NsIgRgKfB4bHs7y2TLsdQQMw2sx2BTbB5WgB/g0PSdg1POWvNulP8ZkcALwtSQCS7sI/cFL5LEmSJGl30njt3myKx4e+ZWZ95wCh3wAAIABJREFUWH2auQH3pI0hQgZCZ/5Z4DgAMxuKTy+3lkeBr5lZ3/CyjQN+XamwmW0NTAJGSXqhcOo+3IO8TZTraWaV9OmmA4PNrDRdPhqYKekdM9sReF1SA/AjYI9Wjut1YIOoD9ywbBfMbAvgo5Iexg3shfgU/KbAghjHJlXafAw4xsw2jsVrJxXOteh5BPcA+wBn4THCpQ+bTYH5IV38Jap7k18GhprZljHGEcAiYEGVa5IkSZKkVaTx2r15CPcyvgA8jk9hAyBpGtAP6BfbJU4EvhNhBmcDc3EDqsVIehD4OfBk1AO+yKcS4/Dp6PFmNit+Rkh6Al84NcXMZgPzgCMqtPkmHtc5wczm4Iu5RsXpY4C5ZjYTuBo4o5XjWh7X/trM/oB7U9uLbYFHY5xzgAdxg/w2oK+ZPY8b82XTTMXCuvuB2XHdC4VzLX0eFLz0J0QfSpyHx7nOwu9rxfASSY3APwGPx7j+ETiqYAgnSZIkSbvRY+XK/PuyPmFmHwMWh0dtID41bJL+2rk9S9ZXGhsbBwAv19XV0adPn87uzhq0RGGrsbGR+vpKkwbdixxL12RdGcu6Mg7IsXQES5cuZd68eQDb19fXz296PlNlrX/sDVxeWExzchquSVKZlIZNkiTpWqTxup4h6RHgkabHzWwKsF2Tw69IOrylbZjZRcBXypw6QNIbLa2vLXRWX8xsHOVTZo0pZCZYK3Sl55EkSZIkbSWN1wSA1hipVeq6GFdZ6nQ6qy+SbgRuXNvtlqMrPY/uREvCBZIkSZK1RxqvSZJ0CU59aC4Ll3d2L1Yx6au7dXYXkiRJkjKk8Zok3YBY9T9M0vstvG4+cKikeR3Rr0I7Y3DFtvlx6GVJR3Zkm0mSJMn6SRqvSdINCKGALklBXvZRSUd1ameSJEmSdZ40XpOkG2BmK4G+kt4Nb+ptwJeAvweukHRNlBuOq3OB5/6tKtFqZn/Cc7LOjv1vAfWSTjKzK3Bp4d644MBYSf9pZgOAGbgQxn64DO2SMtUnSZIkSbuTqxGSpHvyUUnDgH2By8zsY6GyNhE4PSRmn2DNDBJNuRVXKStxEqHIBlwmaXdJg4E7cfGBEv2BpyUNlXRdHPtCCE88YWb/p02jS5IkSZIKpPGaJN2TiQCS5gN/BbYBDHhP0tQ4dxfNq6fdBhxnZr3MbFdcFrak7nWwmU03s3m4GlsxdGEJcFdh/35guwhvOAO4ycw+0/rhJUmSJEl50nhNku5JcZp+BZVDgKpK6El6BXgWOBgYAzSE+tongX8BjpNUB4wFNixcurgo/yppQWkxmaSZwH8Ae7RoREmSJElSA2m8Jsm6g4CNIu4VMzsK96Q2RwMwDjgODyMA6AcsA143s48A36hWgZltXdj+JLAXMKeF/U+SJEmSZknjNUnWESQtxQ3Qa81sDh4P+0oNl94TZZ8LTyyS5gK/BJ4DngJebqaOb5rZs2Y2G5gCnB8e2CRJkiRpVzLbQJJ0AyT1KGwPaHJuQGH7d8CuhdPlJGqb1v0esEmZ42fg8asl/l8cnw98vEnZ84Hzm2urGv920K706dOnLVW0K6mwlSRJ0jXJ/5mTJEnKkIZrkiRJ1yQ9r0myHmBmM1jz9326pKqxrEmSJEnS1UjjNUnWAyTt1tl96E5kyECSJEnXJY3XJEm6BN986E8sWt41DMa7vjqws7uQJEmSVCCN16RLYmZjgEMlHVXm3L64JGqX8SYW+2tmhwPDJZ3TyX1aCcwF/haHTogsAq2paz4+vnkx1t9LeiHODQFuxmPoN8BzvJ4e2Q+SJEmSpF1J47UbYGY9Ja3o7H4ktSFpCp4uqlMws16Slsfu3pLebecmxgALgBdiX8BekpZFTthfAqcAV7Vzu0mSJEnSPY1XM7sDl8LsA7yIq//8Erha0uQocyhwlqQRZjYQ12vfGJgF7AhcIun+CvXvC1yJ57cchqsUHSvpj3H+XOCEKP407mUqayCY2SHATwqHBgJHS5psZqOB0/DnsBA4VZLCszUKeAfYCRhlZlsBlwI9gTeBUyS9aGaGJ5n/aJxrkHRFlXu3ErgAOBLXpz9H0iQzGwDMkPTxKPfhfmkbuAE4CNgIOB5PXL8n8D5whKTXK7T5EeAaYD9gKfCupM+ZWS/gV9GPjYA/xLiWlanjEuBYXAp1apNzNT+Pwj24EPhytH0ysH+MbQP8+ZSedaVn1Bu4Osa0AJhZqH8MBa9xC9+XC4H+ks6M/f64cVhK/H8JrnTVC/gHSROj3FT83d4L+F/gkErjr3Jf5ke/55Xbj2MnAbsBV8UzOVvSo4VqNsCf5d9IkiRJkg6gawSYtZwzJO0maVdc2vJc3IAbXShzEm6wAtyOG7Z1uFG6ew1t7AJcJ2kQruF+IYCZHYwbInvj+TR7Aj+oVImkByQNCc338bgR+HCoIB0D7COpHrgcn3otsRduGNQBr8UYjo/+TADuiHKnAVMkDY6yN9UwtkWSdo9x1Ood6w9Mk/TZaOM3wPjoTyPV84kOBkYAAyUNBg6N4yuAkTH9X4ffy7FNLzazw4DDgSH4fdm5cK5Fz6PA23EPzgUmA/8RY7sNN+5p5hmdAmyPf4x8kQpSqK3o323AsWHYA4zEn+9i4Bng89HP/YErzGyzwrWfivNFw3Wqmc0ys0vNrM1JVCXdgr/D3473+tEY5yfMbBZuyL8DXN/WtpIkSZKkHN3VeD3RzBrNbC7+x30IrhI03Mz6h7fqC8AkM+uHG0YTACTNoDbZShUUgqYDO8T2/sBESYtC2/36OFYVMzsQ+C5wuKQlwGG4UfdU/NG/DNi2cMk0SS/F9p7AbEnPxf4twBAz6ws8AYwzsx+b2X7A2zWMbWJhXJ8wsw2rFQ7elfSr2H4GeFXSrNhvxL3Zlfgz7pG7ycxOKBz/CHB2jH8O7sUcUub6EcAvJL0b4RNFA71VzwP4RWEsKwte+OJYqj2jEcCtkj6IJP8/r9BOi/oXClfPsspzOgb/MAPYArjbzOYBDwOb4zMQJSYUwgUAtosPg31wI7sWo75VSHotPtC2AnoDX+motpIkSZL1m25nvIY37FTgoPC8XghsGAbEZNyYHQlMDm9ViZUtbGpJYXsFbQixMLPBwHX41PqCONwDuLnklQ3P6XaFy2qKU5Q0CRgOvASch3tom2NJXFuKo+0FLGf196GpQVtcfLOCFtwfSQtxT/ZEYBDwbIRBjAQ+jy9u2hW4tky7HUWp/ytYc2ylsTT3jDqKBmC0me2KK1/9Lo7/Gx4ysWsYiq+y+v1a7Z2R9F/x7yLgRuBzzbTb3DvQLPE79ws8rCRJkiRJ2p1uZ7wCm+Kxh2/FNGhxmrkB91SNIUIG4g/3s7jmO2Y2lNXlM1vKo8DXzKyvmfUAxgG/rlTYzLYGJgGjSquzg/twD/I2Ua6nmdVXqGY6MNjMStPlo4GZkt4xsx2B1yU1AD+iwvR1DbwObBD1gRuW7YKZbQF8VNLDuIG9EJ/i3hRYEOPYpEqbjwHHmNnGZtYTDwkp0aLn0UKqPaPHgBPMrJeZbVSl763p3z24t/QsPIa59OG1KTBf0koz+xJVvN1mtln0iwhBOAqPia3Gi0RIjZl9Efi7CuUWUZCTNbNPlUISIhb4CDzLQZIkSZK0O93ReH0I9zK+ADyOT/sCIGka0A/oF9slTgS+E2EGZ+N/WBe2pnFJD+JTxE+y6g/0JVUuGYdP946P2MNZZjZC0hN4bOUUM5sNzMP/6Jdr8008bnKCmc3BF3ONitPHAHPNbCa+gOiMcnXUMK7lce2vzewPuAeyvdgWeDTGOQd4EDfIbwP6mtnzuKH4u3IXx5T+/cDsuO6FwrmWPo+aaeYZXQ+8AvwRN2SfrlBHi/tXmEU4Ab9HJc7D41xn4c+9WvjLzni4Q+mef0DzYQM/AM6K+v8PPr5yXA9cFO/y/ng874xo6xl8wdiPm2krSZIkSVpFj5UrWzqb3v0ws48Bi8NjNRCfejVJf+3cniVJ0tjYOAB4ua6ujj592rymrF1orcJWY2Mj9fWVJlC6FzmWrsm6MpZ1ZRyQY+kIli5dyrx58wC2r6+vn9/0fLdMldUK9gYuj2lbgJPTcE2SpBIpDZskSdJ1WS+MV0mPAI80PW5mU4CmC3BekXR4S9sws4sov8L6AElvtLS+ttBZfTGzcZRPmTWmkJlgrdCVnkc5OvEZNc07XOJ8SQ90VLtJkiRJ0l6sF8ZrJVpjpFap62Lg4vaqry10Vl8k3Yivau90utLzKEcnPqMHgDRSq/DBipVs0LNH8wWTJEmSTmG9Nl6TJOk6nPPQa7yzomdnd4NbvrI2sqElSZIkrSUDu5IkqYiZ3Ri5lVt6XYOZVVNdS5IkSZJWkZ7XJFnHMLNeTZS2Wo2kce1RT5IkSZK0F2m8Jkk7YGYr8ZywRwL9gXNC/QwzOwi4FOgJvAmcIunFCvVsyeqLC7fDZWjPNLM9cYnafnHuIkm/MrMBwAxcpGM/4HozexT4GZ5jeDm+IOshM/socCuuePYBLoN8TJVxTQWukHS/mTXgymSfxnP3PgmMjhR0W+M5af8emA/8rYbbliRJkiQtJsMGkqT9WCRpd1xc4Cr40Bi9HThe0iBgAnBHpQokvVGSo8WVxN4GrjGzTXGJ4ZGS6oFDgZ/FcXCD+WlJQyVdF21MiDZHAT8PpbMDcRGPgZIGA6e0cIx1wCG48VsP7B/HrwKekDQQzzjxhRbWmyRJkiQ1kcZrkrQfE+Pf6cAnzGxDYE9gtqTn4twtwBAz61utIjPbFrgbN3pfwnMVbw88GApYDwIrWSURuwS4K67tCwxhlUTyc7g07F64StlnzGy8mR0NLG3hGO+VtETSMlxNa4c4PoLINCHpz8BvWlhvkiRJktREhg0kSfuxBEDSCjODVv5+mVk/XA73XElPxuEewBxJ+5QpP4BQkGuubkl/NrNdgC8CBwM/MbNdJS2psXvFcivI/0OSJEmStUx6XpOkY5kODDaznWN/NDBT0jvlCptZL9zjerukuwunfg/sZGYjCmV3L6jGfUjUPSvawsw+AwwGppvZNsAKSfcCZ+IxsZu3cYwAj+FhDpjZ9rhxnCRJkiTtTnpNkqQDkfSmmZ0ATAjD9E08BrUSn8MNvy3NrFTuDkmXm9nhuMzxlUBv4M/AYRXqOR6PiT0TX7B1QvTlYOCy8Az3BC6V9FobhwlwBnCbmY0EXgamtkOdSZIkSbIGabwmSTsgqUelfUkPAQ/VWM/juFFZ7tzTwL5lTs0HPt6k7IuU8X5KehCPl60JSfsWtsc0OTemsP2Xcu21hMsP+gR9+vRpSxXtQipsJUmSdG0ybCBJkqRAGq5JkiRdm/S8JkknYGaHAD8pc+p8SQ+sr31JkiRJkuZI4zVJOoEwCruEYdiV+tKZZLhAkiRJ9yCN1yRJugRXPvw/vL+i8/5L+uGRn+i0tpMkSZLaSeM1aVcigf4wSe93dl/aipntC/SW9EhzZdupvVHA94CBwHckXVM4dwHwNTy3ag88S8AvqtS1Ly7ruluHdpoP5W//ERdN2AC4F7iwlryzSZIkSdJScsFWgpmVXd3eGkLadF0wXHvhK/sPWEvt9cRzsx6LS8g25RpJgyR9FpdnvcHMNlsbfauBacDQkLQdAnyJyim8kiRJkqRNpOe1DGZ2B2BAH+BFYCzwS+BqSZOjzKHAWZJGmNlAXIpzY9wA2RG4RNL9FerfF7gSeAoYhnusjpX0xzh/LnBCFH8aOF3SuxXqarrYZiBwtKTJZjYaOA1/zguBUyXJzMbguUbfAXYCRpnZVsCleJqmN4FTJL1onhC0AfhonGuQdEWVe7cS6CvpXTObD/wcT6G0NXAesCUwEk+MP1bSE6EQNQO4FTd8egCnSfpdlXZG4rlFe8ehsyX9Js7Nx6VavwRsAlxZ8mKa2e7AVfizWgx8W9LThT40APvhz/MbwEfMbH9goqTLyvRjO+APwLaSPohjdwP3AXcAvwL6AxtFuVMkLSv3DCTNiuv/1rQdSQsLux/D35nmPj57mdnPaPKOxbO+E+gHbAj8StL3ou3e+Pt0EO7l/bOkI+PcucBX8ffpL8DJkl5v8m5uiD+TNcaQJEmSJO1Bel7Lc4ak3STtCjwLnIsbNaMLZU4itOOB23HDtg43SnevoY1dgOskDcI16S8EiCTyJ+Ba9rviBuMPKlUi6YHwdg4BxuMG2MNmNhw4BthHUj1wOXBz4dK9cIOvDngtxnB89GcCbniBG79TJA2OsjfVMLYifSQNw42eG4APJO0BnM/qRnd/YHa0fzpwp5lVS/r5MLBXeCKPxQ3fIlvGuD8HnG9mg8Iwm4RPaQ/C7+ukOF7qw9OShkq6GrgOuC3u7xqGK4CkV4B5uNQqZtYf99jejRt/I2Pqvg5/lmMLl3/4DEqGazXM7Btm9jwwE/i6pLeauaTsOwa8DRwW92cIsFtM/QN8H/gU7kkdDJwcbY8CdsDv+VB8gdc/F/q2m5nNAd7A1bZ+1dx4kiRJkqQ1pPFanhPNrNHM5uJewiHAPcBwM+sfBsoXcMOnH26YTACQNAOYU0MbkjQztqfjhgFAycu3KGIGr49jVTGzA4HvAoeHTv1huCToUxGHehmwbeGSaZJeiu09ccPxudi/BRhiZn2BJ4BxZvZjM9sPN3xaQiku8xnce1vab8Q91CWW4V5aJE0F3se935XYATfSn406twqPYomboq7/wQ2pfaO+ZSUPraRHo91SO0twI6+lNABjYnskbuwvxn+/zo77Pwf36A4pXFd8Bs0i6TpJO+NG7wXxHjZzSdl3rCeu1DUbfw51hX4dinuql0UFC+L44fh7+EyM55vAgEJDM8JI3haoB4bXOq4kSZIkaQlpvDYhPJanAgeF5/VCYENJ7wGTceNkJDA5DJQSLV2csqSwvYI2hHCY2WDcS3hEwdjoAdxc8sqG53S7wmVlwxCaImkSboi8hE/7397C7i2JelYU92njmPFp72sl7QIMxSVQN2xDfQCLW7nI6MMPG9yILXnkRwKfB4bHu3Rtkz7W9AyaImku7i3ft5mild6x7wKbAXuGwXkvzd+7HngoTOl9qpP0uTJ9W4AreB3d7ECSJEmSpBWk8bomm+LxoW/FtHVxmrcBN07GEAaKpEV4aMFxAGY2FJ/uby2PAl8zs75m1gMYB/y6UmEz2xqfCh8l6YXCqftwD/I2Ua6nmdVXqGY6MNjMdo790cBMSe+Y2Y7A65IagB8Be7RhbNXojRt7pQ+IjYDnq5TfFHg5tsfi8clFxkRdW+ALnH4LCOhtZiPi3H746nhVaGMRHjNblcKHzaVAv0Ks7qbAgriPm5TG1xoirrq0vT3wWeC5yldUZVPgvyUtiffniMK5+4HvlEIpzKwkOzsFOK20SMzM+sRHE2b2aTP7SGxvjIdQzG1l35IkSZKkKrlga00ewhfSvAAswKfN9wCQNC3CBJD0/9l78zC7qmpf+w0JhC4RDSgCYlB0aKiQQBFAlCaINPlojgpIE0jA0B6RLjQK8ilGwAP3qkA4kbYAaRSDJNJ3IsZLaDYkpAL+EE3M4WKOgEiAQwoS6v4xxiYrxd67mlTPeJ+nnqy1ZjfmWrsqY4815/jNKrQ5HLjazL6D/6c9D3eA242ku8xsS+CRuPQEMKVGk0nABsBU31sFwMmSfhfplWbGTvY18E1npQpjvmRmhwE3xi77l/B7AL5u9lAzexuPLp/YkXm1gVfwpQqn41G+g8uvrqtwEnCbmb2KP7OW6z9fNrMS7nyeH9FKzOzrwMXhZL0J7B8bqCqN8Rv8C8AcqmzYKtAA/IGV1ydfB+wX61T/EeVrVevAzA7G1yZ/ONqdCeweyzm+b2ZbAO/gUdRvlzf4dYCLgVvMrBF4AXigUHYB7oTPiWf+PH6Prg9H9vdxr1bDI8lzced3opktw5ck/Aa4soO2JUmSJElNBjQ3ZyrGVcXM1iVeOUeE7CHAJL3as5b1Dco7/SWt31rdNva3ENhbUmNn9Jd0LaVSaTiwoK6ujsGDa+3R61o6Q2GrVCpRX1/tBUffIufSO+kvc+kv84CcS1fQ1NREY2MjwGb19fULW5Zn5LVz2AHfAFP+n++odFyTpG+R0rBJkiR9g3ReOwG5AtP7VJjMbCawaYvLiyTt294xzOwc4GsVinaX9I/29rcqdLYtkhYC74u6mtlo/HV8Sy6VVPW1tKTh7bWhNTpqS1fTmZ+xJEmSJOkLpPPahXSmAyHpXODczupvVeguWyL36ehWK3YDvcmWIumkrhrLljczKCOuSZIkfYp0XpMk6RVcf8/LNC3v3j9J//7Vj3XreEmSJMmqk85rknzAMbP/H/g+MFJSo5l9Fvg58HE8f+7juFzvWzX6GA+cjssTn1SW402SJEmSzibzvCbJB4xIh1Y+3hpX7PpbocrbwCmh5rUlrow2uZVu5+AyvTd2rrVJkiRJsjIZeU2SDmJmzcBZwFeBYcBpkqa3TP1VPC8fA1cAe+J5Xw8FjsVlet/CldIWVxlzPPB1SV+N80HAIuCLwLp47tV1cMWsyyX9NOo14FFUA4bgOXUHA1NxgY2HymPEBrqFcfyumT0GfL7WvSinJTOzd1u5bUmSJEmySmTkNUlWjSWSxgCH4cn/28IwYJakrYCrcJGAqSHVWgK+VaNtWYq2nJ1hL+BPkhbgDudukrbGhTWONrOi0zkalz0ubzw7F/hFOKsVMbO1cAWzmW2cW5IkSZJ0Kem8JsmqcXP8OxvYyMzWbEObNyTdEcdPAi9ENgNw53Xzag1DivY2VkjNTmRFCq+1gavMbB7wR2AjYFSh+a8lvQlgZl8AtsEjtRWJqO7NwIOS0nlNkiRJegXpvCbJqrEUQNLyOB+Ev54v/m61dGibCsfLy30UzltbztMATDCzYcDOwK/j+nnAYmArSaOAx1qM/UbheGd8KcCCUCTbBLjHzHYHCEnhG4BXgW+3Yk+SJEmSdBvpvCZJ57MYWN3MyhHUQ2pVbi+SZgFDgfOB2yIaC7Ae8F+SlplZHbBjjT4ukLSRpOEh6vACsIeke81sNdxBXg58U1JqSCdJkiS9hnRek6STkbQMOBG4LzY7LW+lSUe4FjiKlVW/pgBHmdnTeOqrhzvY917AeGAkUDKzOWY2tVYDMzvYzF4ADgB+aGYvmNmIDo6fJEmSJFUZ0NycQZUkSXqOUqk0HFhQV1fH4MGDu3XszlbYKpVK1NfXd1p/PUnOpXfSX+bSX+YBOZeuoKmpicbGRoDN6uvrF7Ysz8hrkiQfWFIaNkmSpO+ReV6TpBdiZpOonDJrYiEzQbdiZuPwTWEt+a6kO7vbniRJkuSDSTqvSdILkXQlcGVP21EkHNR+5aQuX97MwIy+JkmS9CnSeU2SpFcw4+5/8s6y7v2TdOjXN+jW8ZIkSZJVJ53XJOkFmNn3gXUlTa5QNhHYW9L+3W1XwYaPA7/F/2YMBP4EHC3pVTPbHzi7UH0T4GFJX+t+S5MkSZL+TjqvSRKY2cCC2EAShNLWy8BO5ZyyZvYT4HvAKZJ+zQqhBMzsKeDGnrA1SZIk6f+k85pUxcxuAAwYDDyPa9zfAlwiaUbU2Rs4VdLYyOt5DbAOMAeXOZ0i6fYq/e8C/BR4FPgC0AwcJOnZKD8DOCyqPw6cIOmNCl1V2kw0AjhA0gwzmwAcj3/eXwOOk6SIaI4HXgc+A4w3sw3x5P8DgZeAYyQ9b2aG51RdO8oaJF1UxZa18TysWwDvAJJ0YPR9Ey4wsCZwh6TTK7RfA7gE2BV3Gp8qlA0EfgzsGZfuBs6o5nSb2ZXAPEk/i/M6YCbwaeBgPB/tGlF9sqQHot5CXBp212j/zZhL2YZ14162HG9rPPKacrJJkiRJl5CpspJanChpG0kjgfnAGYQ0aaHOEbjDCnA97tjW4U7pmDaMsQUwTdKWwK+I189mthfuuO6AJ8sfiEf6KiLpTkmjJY0GpgJP4HKnOwIH4lHDeuBC4OpC0+1xp60OeDHmcGjYcyMukQru/M6UNCrqXlVjTnsAQyWNCJnWY+L6v4B9wo7RwDZmtmeF9scAm+EO+JeBbQtlR0fbreNnq7hWjQbe/7waQjXrHmB7SVsBB+EOd5GhkrYNxxUAM5uDO/WfAc6tMN6RwA2S3q5hU5IkSZJ0mHRek1ocbmYlM5uHS5yOBm4FdjSzYWY2DNgZmG5mQ4E64nWxpCeAp9swhiSVI4uz8YggwG7AzZKWhKN1eVyriZntAZwC7CtpKbAPMAp4NByvC4BPFJrMkvSXON4OmCvpmTi/BhhtZkNwtapJZvZDM9sVd0SrMRf4vJlNNbMDgKa4PhC40MzmAiX8fo2u0H4scK2kd+I1/S8KZbvhzufb4SBeU+u+hJTsEDMbGa//D2aFk/pp3MGfD/wS2DCiw2Wuq9DfaOBjwLPAscUyMxuMf06ubtkuSZIkSTqLdF6TikTE8jhgz4i8ng2sGc7UDNxJOQSYIenNQtP2SrYtLRwvZxWWspjZKGAasJ+kl+PyAODqclQ2IqebFppVXIbQEknTgR2BvwBn4hHaanX/ikeU78Mdy7lmtibuVH8Y2C4iu7fhywe6mmuBibjs67OS/hbXbwIuk7QFHsVd1sKeivdG0jvR52Etir4K/FVSW760JEmSJEmHSOc1qcZ6+JrGVyKidmShrAF3hiYSSwYkLcGXFhwM7619HLkK498PfMPMhpjZAGAS7gxWxMw2BqYD4yU9Vyj6LR5B3iTqDTSzatp3s4FRZva5OJ8APCXpdTPbHFgsqQH4ASu/ym9pyybAckm3AScDGwAfwe/p3yUtDXv3q9LFg8BhZjbIzNbCvySUuR+YYGarm9nqYWPV+xJchz+XSaxY4kHYsyCOj8TXNleb0yfMbN04Xg34OjCvRbUjyahrkiRJ0sWk85pU4248yvgc8HvgyXJBvIoeiq+JnFVoczhwUiwzmIw7N+/b1NMWJN2Fvy5/hBVO0pQaTSbhTuIJmG7SAAAgAElEQVRUM5sTP2MlPQycBcyM1/WNVHEaJb2ERxNvNLOn8c1c46P4QGBe7KS/BN/oVI2RwCMx3mPA+ZJeBC4Gvmhmjfia2QeqtL8cWIS/mn8Q36xWLHsa38T1VBxfUcMWJC0CngF2wZd9lDkJuM3MngQ+BbxSoxsD/hj35Wng48C33ys0+wTwRTLLQJIkSdLFDGhubu9b3iSpTETm3pTUHJkHHgJM0qs9a1nSmymVSsOBBXV1dQweXDX42yV0tsJWqVSivr5aYL9vkXPpnfSXufSXeUDOpStoamqisbERYLP6+vqFLcszVVbSmeyAb0gqewNHpeOa9GZSGjZJkqTvkc5r0mlIuhe4t+V1M5sJbNri8iJJ+7Z3DDM7B6ik3LS7pH+0t79VoZfZMg1P+1VkmaRtutOOJEmSJOlq0nlNupyOOKk1+jqXyvlFu51eZsuxrddKWtLZywaSJEmSried1yRJegX33/kqy5d175+kfQ9Yv1vHS5IkSVaddF6TpJ2EKtaP8Zy2q+P5Ws+OjWrrApfhylerA1dWk5FtwzjHAmtJ+knnWL5S383AEElvmNlJwI3lpQ615lejv4vw9FnDgZGSGjvb5iRJkiSBTJWVJO0iVKpmAVuH2tRo4Cu4khfAd4G3gS2Bejxfa8u1qG1C0rSucFwrcBLw0cJ5rflV4zZgJ+BvrdRLkiRJklUiI6/9jIionYWrHQ0DTpM03cyGA09IWj/qvXdePsbzhe4JrAUcist/bge8hatWLa4x7kI8L+uXgY1xFaqP4gn2PwIcGTlXMbNxYeOauKN3sqTZZma4AMLauJRqg6SLzGwN4Ee4FO1gPM/ocRE1PBoXAmjCv4wdKOlPVWz8M7C/pLlx/i2gXtIRETncGVgDeDns/Vvh3jQAuwKXS5pW6HbNaPNunI8Ku5uBN83s93EvZ1ex6WxgmKST43wYIOCTwGnAupImxz24FM/V+g9gDrChpP0r9Rt9vRddrXQe184CNgJ+bWZLgUMK8riV5leRcr5ff4RJkiRJ0nVk5LV/skTSGDzh/sVtbDMMmCVpK1Yk0J8aMqYl4Ftt6GOwpC/gr4+vAN6RtC0ejTwPwMw+DXwP2EtSPS4u8KtofzwwMyRc68IOgNOB1yRtK2kU8CLwnSi7ENg1ooRj8OT+1bgWV6QqcwQrFKcukDQm+r8Jf21evDePS9q67Lia2TaRsP8fuJDAHVG3BOwfCljrA3vgjmg1rgMOioguuLM/s4XkLsAxeMaGEbjkbKdkEZD0I/x+7h/yuc9AzfklSZIkSY+Skdf+yc3x72xgIzNbs1bl4A1JZQflSeAFSXPivIS/Om6NXxbar104LwGbx/EewKeBhwtRukFm9jHgYeA/zGxt4HfxA7AvMNTMylHGwcDcOH4QuNbMfgvcIemvNey7DnjUzE4HPo/Lo/4hyvYys38H1uX9vxdLWeFgAyDpCWDLcFCnAzuG/RfgDvUTwEu4UMMG1QyStMjM5gPjgJm45O7JFaqOBa6XtAxYZmY3xZhdQo35JUmSJEmPks5r/2QpgKTl4SAOApaxcqS9pUPbVDheXu6jcN6Wz0rLcct9FNsPAO6WdHiF9tPN7BFgd3zZwZG4POsA4HhJD1Zo8zU84ror8DszOzakZd9HwVHcC3/93hCbrD4J/AQYI2mBme3AyjKnb1bbrCTpZTO7CzgAeFjS/wD/Xi43s8twadZaNAATzGwB8CFWONSrynLimbfxC8z7aDm/TrIrSZIkSTpMLhv44LAYWN3MyhHQQ3rIjnuBPc1si/IFMxsT/24OLJbUAPwA2DaqzAROMbO1ot4QM/t8vGr/lKTHJF0QfW/VyvgN+FKFg/FlBABD8bW3i81sNXytb1XM7LNRDzNbB3eG58X50IKdW+Jrjy9rxaZb8c1Op7JivWxLHgIONbNB4Yh+o5U+AZ7HHXuo/byX4E4zYXfV+SVJkiRJT5PO6weEeN18InCfmT2GR+V6wo4/49HUq8xsrpk9i6/nBDgQmGdmTwGXhL3gr+LnAo/HOsxZ+Gv/gUCDmc0zs7nAx4Gft2LCrXjU9RlJi8KmecAteIT0UWBBK33sF3bOjfoPA1dG2aeAuWb2DO4oHyrpxVqdRbR2Br5G+boq1aYBfw8bHwSeBV5rxc5TgJ+bWYkaSxfwddHXmNkcMxvRyvwqYmYXm9kLwCbA/RHhTpIkSZJOZ0Bzc9XUjUmS9CLMbIik181sMB6NvkVSTaeyL1AqlYYDC+rq6hg8eHC3jt3ZClulUon6+vpO668nybn0TvrLXPrLPCDn0hU0NTXR2NgIsFl9ff3CluW55jVJ+g73h+O6JnA/HtlNVoGUhk2SJOl7pPOatBkzm0TllFkTC5kJehwze4L3f7ZnS6q5lrUrMbNz8M1lLdm9rGzVGpK2q9Bvtz2TzphDkiRJkqwq6bwmbSZeUff619SSOiUHamci6Vzg3C7ot9ueSVfNoafo7CUDSZIkSfeQzmuSJL2CP874J+8u674/SbsdUmsPW5IkSdJbSec16XWY2Z3ACZL+0tO2tIeQt30KuEzS5Lh2Fp7Wajmer/Z8Sb+s3kun2XIuMF/SL81sF2ANSfeuQn8PARdJut3M/g14UdJjUTYYz5awDUBZgjhJkiRJuoJMldXPMLOBPW3DqiJpXF9xXMv3O/79OXBbiyqXStoyZHfHAVeY2Ye72i5J5xSc5F1w4YfO4t9YkYMX3DG/CJetTZIkSZIupd9HXs3sBsBwSdHncdWmW4BLJM2IOnsDp0oaG3kurwHWAebgsqZTJN1epf9dgJ/i+TC/ADQDB0l6NsrPwPN3AjyORxTfqNLXOOC8wqURwAGSZpjZBOB4/Jm9BhwnSWY2Ec+b+jrwGWC8mW0InI/nQX0JOEbS8xEZbMClWwfiCfEvqnHvmoGz8ET7w4DTJE03s+HAE+UIW/G8fAxcAewJrAUciif+3w54C9hP0uIa4y4E9pbUGBG/Eu4sDQd+Bvxf4ARgo7DploK95+J5StcCvitpepUx1gYWAZ+T9HJcuwh4XdIPKn1uJL0az/visGkr4GzgdlwR7HZcXnbd8jiSirlY18U/HzW/NLbyrA8B/gVsWbgPF+Gf08eB8aEa1oA/h9/j9341M9sNuDkEHSqN+xARXa10Htf2wOV6d4vNYv9b0nV4JoThteaVJEmSJJ3BByHyeqKkbSSNBOYDZxBynIU6R+AOK8D1uGNbhzulY2idLYBpkrYEfoU7NJjZXrjjugMwEncYv1etE0l3ShotaTQwFXc+7jGzHfEE/jtJqgcuBK4uNN0emBw2vxhzODTsuRG4IeodD8yUNCrqXtWGuS2RNCbmcXEb6oM7urMi2ngV8AAwNewpUXl3fC02AXbGnd9zgTpJO+D35Cct6i6P+7cvcLmZfbRShyEMcBuhPBVqXYewQnWr0uemzBbA5fGsbjezUcAeFWwh+j7WzP6ELyk4WtIr1Sbahmc9BjhF0ufwLwI3ht0j8M/Yl1vMcx4ucHBd2FvRcW0rku7Bc8xeEP1VE1VIkiRJki7hg+C8Hm5mJTObh/8nPxpXWdrRzIaZ2TDcMZpuZkOBOkLXXtITwNNtGEOSnorj2cCn47gc6VoSkp+X04ZXqxHdOgXYV9JSYB9gFPComc3BFac+UWgyq/CafTtgrqRn4vwaYLSZDcGVkiaZ2Q/NbFc8gtcaNxfmtVFIk7bGG5LuiOMngRcKaZtKeJSwPdwi6d1QqnoF+E2hr41b2HQV+AOJsbev0W8DMDGO9wL+JGlhnFf63JT5s6RHAMxsdfy5HiupomqZpGnhbG4PnBWfuWq09qz/KOmFOH4Kf/b/CgW1ubT/3iZJkiRJn6JfO68RxToO2DMiaGcDaxbkOA+JnxmS3iw0ba/s2NLC8XJWYTlGRPGm4a/WX47LA4Cry1HZiJxuWmhWcRlCS+IV+o7AX/DX3Ne3odnSaFt2zAYBy1j5s9PSoW0qHC9n1e9Py/aVbGo3kmYBQ8xsJO7EXgPVPzeFpsX7/XH8y8qdsdzhJOAoM7u8wnjz8Mj4LjXMau1ZV7wXhfNVWQrU2nNNkiRJkh6nXzuvwHr4msFXYkf0kYWyBtxhmUg4LZKW4K+IDwYws63xV7Ed5X7gG2Y2xMwGAJOA+6pVNrONgen4usXnCkW/xSOBm0S9gWZWTb9tNjDKzD4X5xOAp0JWdHNgsaQG4AesvOmmPSwGVo/+IF699xKOADCzz+BrUme3Uv9a4FRgJ/zeQ+3PzUpIWiRpfUnDJQ3Hl5pcIenosGNEua6ZbRY2PVOxM6c9z7qtLAE+1IZ6zxPLZMLu0VXqtbW/JEmSJOl0+rvzejceZXwO37jyZLkgom5DgaFxXOZw4KR4XTwZmIc7Mu1G0l3AL4BHoh+AKTWaTAI2AKaa2Zz4GSvpYXzj1Ewzmws04puSKo35Er4+9UYzexrfzDU+ig8E5pnZU8AlwIkdnNeyaHufmT2GR/x6C4NifrfjG9VaU366Dr9fMyIiDzU+Nx3g+2Y2P5YA/Br4dnkzXyXa86zbwW+AMfF5OrNGvf8AxsVn/wx8WUIlrgcOif4OBzCzx/HP+YfN7AUz6/ViFkmSJEnfZEBzc3vfkPdvzGxd4M3YsT0CeAgwSa/2rGVJa0S2gSHVsjkkvZNSqTQcWFBXV8fgwYO7bdyuUNgqlUrU169qoLx3kHPpnfSXufSXeUDOpStoamqisbERYLP6+vqFLcv7faqsDrADcGG85gc4Kh3XJOl/pDRskiRJ3ySd1xbIVYjep0RkZjOBTVtcXiRp3/aOYWbnAF+rULR7G15zdyo9ZUvkCK2UMmtiITNBu5D0Pm8kUmVVUpa6VdK5HRmnM+jMz1M7xx2Nr/duyaWS8lV/kiRJ0utJ57WNdKZTEU5TjzlORXrKlnCUutxZCge82sajHqOrndQa486hF96PnuDd5c2sltHXJEmSPkc6r0mS9ApKt/4T3um+P0k7HL5Bt42VJEmSdB7pvCbdQkib7i1p/wplu+AypNt0t13VKNprZvsCO0o6rYdtOhI4Gc/usAw4WdIfatRvwGV7L+1iu1bDJZfr8Lyz/8BFG/5Ss2GSJEmSdID+niqrV2JmA3vahqTtSJrZk46rmQ0KVa6fAruF/O25wM97yqYKXAt8XtIoXADkfSINSZIkSdIZ9IrIq5ndABgwGE+UfiQeyblE0oyoszdwqqSxkcLqGmAdYA4uiTlF0u1V+t8F/4//UeALuILWQeV8m2Z2Bp7rE+Bx4IRq6ZbMbBxwXuHSCOAASTPMbAJwPH5fXwOOk6SI4o0HXgc+A4w3sw2B84GBwEt4TtLnzczwDTVrR1mDpItq3LtmPC/oV4FhwGmSppvZcDzqtn7Ue++8fAxcAewJrAUcChyLy8u+hSt8La4y5mrApcCuuJrWG5K+aGaDgDvCjrWAx2Jeb1foYwpwEPAqno6sWNbm51G4B2cD/xZjH4XL8O4JrI4/n/KzrvaM1sBz3+4KvEwhx2nLqHE7Py9nA8MknRznwwABn8TlYqfgSlaDgB9JujnqPYR/trcH/onnHx4ADAH+GxdSeIHWqTOzB3GJ2UeACZEG7hA8V+8aUW+ypAdi7M8DPwM2jDEvknStmX087tGm+PO9SdJ5kt4FZhbGfARXGkuSJEmSTqe3RF5PlLRNSHHOxxOkN+DqUGWOIJSw8CTpl0iqw53SMW0YYwtgmqQtgV/hzg5mthfuiOyAq2kNBL5XrRNJd5alO4GpuBN4T0iKHgjsJKkeuBC4utB0e9xBqMMlQq8HDg17bgRuiHrHAzNDFrQOuKoNc1siaUzM4+I21Ad38mZJ2irGeACYGvaUqJwJoMwoYCwwIiJte8f15cAh8fq/Dr+X71OnMrN9gH3xjUPbA58rlLXreRT4V9yDM/DI3x9jbtfhzj2tPKNjgM3wLyNfpor6WAfsuw44KBx7cDWymSFH/CTwpbBzN+AiM/twoe2nonxcSAUfAzxpZovwL1DHt+G+1AHj8M9/fYwDcA+wfYx9EB45JeycgauEbRm/k+UvhdcBF0vaNvray8y+UmHMb7GyM5skSZIknUZvcV4PN7NSKPscgjs1twI7mtmwiFbtDEw3s6H4f8g3Akh6Ani6DWNIUjmaNhvXowf/z/xmSUskNeOvO3er1EERM9sDOAXYV9JSYB/cqXs01JQuwKNdZWYV1gBuB8yVVJYJvQYYbWZDgIeBSWb2QzPbFfhXG+Z2c2FeG5lZWzTp35B0Rxw/CbxQSFFVwqPZ1fgrHtG8yswOK1xfDZgc838aj2JW2tk+FvilpDckLWdlB71DzwP4ZWEuzYUofHEutZ7RWOBaSe+E0tYvqozTLvskLcK/kI2LSxNZkapqA+DXZtaIO5Mfwd9AlLkx1MyIz/23gDGSNsU/e78p5COuxm2Slkb0+0lWfO4/jX/pmo/fuw3jbYABgyTdUpjDK2a2DrALcHHcu8eAjYDPFwczs9Pj2tmt2JUkSZIkHaLHndeIhh0H7BlRnrOBNcOBmIE7s4fg8p1vFpq2VxpsaeF4OauwZMLMRgHT8FfrL8flAcDV5ahsRE6LeTzbpPokaTqwIy5PeiYeoW2NpdG2LNM6CN/QU3y+LR3apsLxctpxfyS9hkfybga2BOaH43MI8CV8c9NI4LIK43YVZfuX8/65lefS2jPqKhqACWY2EvgQUN5k9Z/4komREcl/gZXvV/EzszseXRaApF/hDuj6rYxd7bneBFwmaQtga/zzUutZrYb/zo0p3L9PS3ov0m9mJ+CfgXFaIbWbJEmSJJ1Kjzuv+Nq914BXzGwwK79mbsAjVROJJQOSluCRrIMBzGxr/PVtR7kf+IaZDYko1iTgvmqVzWxjYDowXtJzhaLf4hHkTaLeQDOrprE2GxhlZuXX5ROApyS9bmabA4slNQA/oMrr6zawGFg9+gN3KjoFM9sAWFvSPbiD/Rr+ins94OWYx4dqjPkgcKCZrROb144olLXrebSTWs/oQeCw2By1Vg3bO2LfrcBOwKn4GubyF6/1gIWxBvUr1I52LwC2DtEFzGwssARfn9sR1os+wX/nyrqsApaZ2QHlimY2TNLruNN9ZuH6J+JLC2Z2DHA08BVJ/+ygTUmSJEnSKr3Beb0bjzI+B/wef7UJgKRZwFBgaByXORw4KZYZTAbm4Q5Uu5F0F/6K+JHoB3wTTTUm4a97p5rZnPgZK+lhfG3lTDObCzQC+1UZ8yV83eSNZvY0vplrfBQfCMwzs6fwzTEndnBey6LtfWb2GB516yw+Adwf83wauAt3yK8DhpjZn3BHsWIap3ilfzswN9o9Vyhr7/NoM608o8uBRcCzuCP7eJU+2m1f4S3CYfg9KnMmvs51Dv7cqy5/kVQC/gP4fdj+Y2D/giPcXk4CbjOzJ/EvHq/EOMvwe3Ksmc2LscpLHg4FRsT1efhyg/Viuct/Auvin7c5ZvZoB+1KkiRJkpoMaG7u6P99PYeZrQu8GRGrEfirV5P0as9aliRJeymVSsOBBU3PbdTnRQpKpRL19dVeuPQtci69k/4yl/4yD8i5dAVNTU00NjYCbFZfX7+wZXmvSJXVAXYALixsVjkqHdck6dvUf+0jDB48uPWKnUTKwyZJkvRN+qTzKule4N6W181sJp6DssgidUBH3szOAb5WoWh3Sf9ob3+rQk/ZYmaTqJwya2IhM0G30JueRyV68BmNZkX2giKXSrqyq8btD6TjmiRJ0jfpk85rNTripNbo61xcxajH6SlbwvnpFQ5Qb3oelejBZzSHyunIkiq8u6yZ1Qal45okSdJX6VfOa5IkfZf5v3yFAW93/Z+krSZ9tMvHSJIkSbqOdF6TpA8TAh7X4zlf3wb+jEvyvmRmnwV+Dnwcz+P6OHC8pLc6ONZCXCa3MSRz/085XVwsX7gaz2CyOvBHXDa3qXJvSZIkSdIxekOqrCRJOkBIuTYD/yHJQhjiL7hyGLgze4qkz+FiEmvjqeU6g4nAZwvnwuVmR+N5l4fhcrZJkiRJ0qlk5DX5wGNmzXj+16/iTtdpkqab2XDgCUnrR733zsvHwBXAnsBaeB7UY3H537dwBbbFVcYcD3xd0lfjfBCeZ/aLeL7Uy4B1cNWryyX9NOo14FFUA4aEs/hQoevZuGIdkhYCC+P43cj3u5KcawW7FhLR1Urnce0IYBtcKnYKMFnS/YVuVo/78W6tsZIkSZKkI2TkNUmcJZLG4EICF7dWORgGzJK0FXAV8AAwVdKWQInKmRrK3ArsaGZlede9gD9JWoA7nLtJ2hpXWDvazIpO52hcTnmljVpmthruuM5sOViohh1Zqay9SLoGd9y/HTKx98cYG4XgwsvA67jwQ5IkSZJ0Kum8Jolzc/w7G9jIzNZsQ5s3JN0Rx08CLxRSiJWoIfcaqlu3sUKGdiIrUl6tDVwVKlZ/BDYCRhWa/1rSmxW6vQR4A7i0eDGiujcDD0paZee1GpJeDId6Q2ANKqcOS5IkSZJVIp3XJHGWAkgqy+gOwl/PF39HWjq0xc1Iy8t9FM5bW5bTAEyITVc7A7+O6+cBi4GtJI0CHmsx9hstOzKzi4DPAN+Q9G7h+kDgBuBV4Nut2AOtz7lVwrH+Jb6MIkmSJEk6lXRek6Q6i4HVzawcQT2kVuX2ImkWMBQ4H7gtorEA6wH/JWmZmdUBO9bqx8zOA+qBfyvu7o9lBA24I/1NSW3Rgn4eGBPtvwx8rEq9JcCHCmN9yswGx/EawH7AvDaMlyRJkiTtIp3XJKmCpGXAicB9sdlpeStNOsK1wFGsrJI1BTjKzJ4Gvg88XK2xmW0BfAdfWvB/zGyOmf0mivcCxuO7/0tRNrUVe74HnBprV/8/fBNZJS4Hzok+d8Mlm58ws7n4Eop/Aj9sZawkSZIkaTcDmpvbEoxJkiTpGkql0nBgwYD5H+/zIgWlUon6+vou6787ybn0TvrLXPrLPCDn0hU0NTXR2NgIsFl9ff3CluWZKitJkl7BFt8YxuDBg7t8nJSHTZIk6duk85okXYiZTaJyyqyJhcwE3YqZjcM3hbXku5Lu7G57upt0XJMkSfo26bwmSRci6Urgyp62o0g4qP3eSa1GRl6TJEn6Num8JknSK/jr9S+zWlPX/0myf6+WQCFJkiTpC2S2gSRJkiRJkqTPkJHXpMNEOqUvSHqrp23pKcxsA6AR+IOk/ePa7via0pHAJZImr+IYU4D9gZck1cz5WqX9cGB3SR2WazWz7wPrSppsZqOBz0r6VaH8NmAz4F1cROGEnlrTmyRJkvRvMvKadJjQtf/AOa4ht1rmMt6/fvSvwCTgwk4a8lRgx444rsFw4OhOsgVgNHBgi2sTJI2StBVwEXB1J46XJEmSJO+Rkdekw5hZMzBE0htmthD4BfBlYGPgTOCjuCrVR4AjJT0cUcAn8OT8XwEGAMdL+kONcT4GTAM+HfUvlHRdlC0Ebo6+PgT8VNKlNfr6M7C/pLlx/i2gXtIRIbG6M7AG8HLY/LeCzQ3ArniC/mlmdijw31G2d3kMSc9H3//Wyi0s2jUOOAuXY30bOFnSbDP7Q1x7wMzukXSamU0Ajsd/f18DjpOk6Oc7+D1/F3gT+BIwFdgsIuXPlyPEFWz4PhFdrXQe14YB5wJDo7+HJX1b0muFrj4U4ydJkiRJp5OR16QzGSzpC8DXgSuAdyRtC3yXlVMzDQPmStoSOAG4qSwtWoWLgcaovztwQcimlvmopHrgi8B3zWzLGn1dC0wonB8BXBPHF0gaI2kUcBPw4xY2Py5pa0nTzGwj4BTcSV8lzOzTuLLVXjGPScCvAArR1h3Ccd0Rj3ruFHUvJKKc4dTuG3VHAftIehf4d+CZiJRXdFzbiqRXgHOA+6O/bxfmcaWZLQJ+xMr3OEmSJEk6jXRek87kl/Hvk8DahfMSsHmh3tt4lBZJDwFvAVaj392An0f9v+Ov6ccWyq+Ksv8G7gB2qdHXdcDBZjbIzEYC6wHlqO9eZjbbzBqByfjr8TJLCYcyuAI4XdIbNcZqK3vgUeWHI5p5AzAoIs4t2QcYBTwadS8APhFlewP/Kel1eM/R7DYkTZK0Kf5lpbOWTCRJkiTJSuSygaQzWQogabmZvXcOLKeXfNYkLTKz+cBeuJPbIKnZzD4J/AQYI2mBme0A3Fho+qakopbyF4CrYp7rAmuZ2Z2SxnXArAHA3ZIOb2PdqyWd04FxWmMZK3+hXbMjnUi63swuN7Nh3e1AJ0mSJP2fjLwmPcEa+LpM4jX4WsCfatS/Hzgq6m8IjAMeLJRPjLINoux3rYzfgL+aPxhfRgAwFI8ILzaz1YBja3Ug6SOShksajkdp7+qg4wpwL7CnmW1RvmBmY6rU/S1wuJltEvUGmllZiPp24DgzGxJlw+L6Enwdams8D9Sb2WrRx95V6q3Un5mta2afKJzvA/wzfpIkSZKkU+kV0bDkA8crwGgzOx2PJB4s6e0a9b8N/NzMno76Z0qaXyh/2cxKuEN1vqR5rYx/K76J6XFJiwAkzTOzW4Bn8M1adwI7dWBumNmX8E1kQ4EBZnYQ8E1J91SqL+nPZjYej+SuhTv3fwQer1D3YTM7C5hpZgOj7i340ozr8M1ys83sHeANM9sJeBpQLIf4U411r7cC3wCeBRZFn5V4AJhsZnOB3+NrXG8xs3XwKPs/8fW2zVXaJ0mSJEmHGdDcnP+/JN1Heee+pPU7qb+FwN6SGjujv6T7KZVKw4EFdXV1DB5ca99e59CV8rClUon6+vrWK/YBci69k/4yl/4yD8i5dAVNTU00NjYCbFZfX7+wZXkuG0iS5ANFVzmuSZIkSfeQywaSbkXSQuB9UddQbWqo0ORSSVfW6G94petm9gTv/3zPllRzLWtXYmbTgO1bXF4maZtutqPX3ZuupnlZMwPSaU2SJOkXpPOa9ApCSnR0qxXb3l+3OoRtobc4h73x3gD8/cqXGPU6RTkAACAASURBVLi0a/4kfeLUDbuk3yRJkqT7yWUDSZIkSZIkSZ8hI69J0guoJMVaKJuIb0pbJXWsVSEU0GYA2wAUN9xFTtyLcMEHcKGI0zPbQJIkSdIVZOQ1SYJIPZW0wMwG4SmwLsLVzlqyBJggaQSwFS7gML77LEySJEk+SGTkNamKmd2Ay7YOxhPYH4nnFL1E0oyoszdwqqSxZjYCuAZYB5iDS8JOkXR7lf53AX4KPIo7PM3AQZKejfIzgMOi+uPACdXkWM1sHHBe4dII4ABJM8xsAnA8/nl/DThOkiKiOR54HfgMMD5EEM4HBgIvAcdIet5cSqsBl70diCtzXVTFlrVx8YMtgHcASTow+r4Jz/+6JnCHpNMrtF8DuATYFc85+1ShbCDwY2DPuHQ3cIak5VVsuRKYJ+lncV4HzMTlaA8GTsRzxQJMlvRA1FuI56rdNdp/E7g/Up2tRDFNmaQmM3sK+GQle5IkSZJkVcnIa1KLEyVtI2kkMB84A3fgJhTqHIE7rADX445tHe6UVlOJKrIFME3SlsCvgLMBzGwv3HHdARiJO4zfq9aJpDsljZY0GhcgeAK4JxS8DgR2klQPXAhcXWi6Pe601QEvxhwODXtuBG6IescDMyWNirpX1ZjTHsBQSSMkjQKOiev/wpP31+Ob07Yxsz0rtD8G2Ax3wL8MbFsoOzrabh0/W8W1ajTw/ufVEK/07wG2l7QVcBAr1MbKDJW0bTiubcLMPgp8HV86kCRJkiSdTkZek1ocbmaH4pG5dYDngHOBnxSkR3eOekOBOtzhQ9IToYjVGpJUjizOBvaJ492AmyUtATCzy4GftdaZme0BnALsKGlpSJWOAh714CkDgA8XmsyS9Jc43g6YK+mZOL8GuCykUh8G/iOiqr+jtgTtXODzZjYVeIgVjtxA4MJYIzoA2BB3RO9u0X4scK2kd4B3zOwXwJcK96WhrEhmZtcAXwX+s5IhkmaZ2RAzG4krZx2MR7nBo683mdnGeIR4QzPbUNLiKL+uxhzfR9ynmcD/KjzTJEmSJOlUMvKaVCQilscBe0bk9WxgTUn/g2/cOSR+Zkh6s9C0vZt0lhaOl7MKX6jMbBQwDdhP0stxeQBwdTkqG5HTTQvNKi5DaImk6cCOwF+AM/EIbbW6f8UjyvfhzuZcM1sTd6o/DGwXkd3b8OUDXc21wERgL+BZSX+L6zcBl0naAo/iLmthT5vuDby3VOJ24F5J/6szjE6SJEmSSqTzmlRjPXx96Cux0/zIQlkD7gxNJJYMRIR0Ph7Zw8y2xl/3d5T7gW9E1HAAMAl3BisS0cPpwHhJzxWKfotHhjeJegPNrJr23WxglJl9Ls4nAE9Jet3MNgcWS2oAfsDKr/Jb2rIJsFzSbcDJwAbAR/B7+veICG8M7FeliweBw8xskJmthX9JKHM/MMHMVjez1cPGqvcluA5/LpNYscSDsGdBHB+Jr21uN+GY/xYXOjinI30kSZIkSVtJ5zWpxt14lPE54PfAk+UCSbPwTUdD47jM4cBJZjYPmAzMwx3gdiPpLuAXwCPRD8CUGk0m4U7iVDObEz9jJT0MnAXMNLO5QCNVnEZJL+HrbG+MJQ/jWbFr/kBgXmxGugTf6FSNkcAjMd5jwPmSXgQuBr5oZo34mtkHqrS/HFiEv+Z/EN+sVix7Gt/E9VQcX1HDFiQtAp4BdgFuLRSdBNxmZk8CnwJeqdWPmT2OP48Pm9kLsRkM4JvR9x6Fe39Wrb6SJEmSpKMMaG7OVIxJ52Bm6wJvSmqOzAMPASbp1Z61LOnNlEql4cCCuro6Bg/uUPC3VbpLHrZUKlFfXy2w37fIufRO+stc+ss8IOfSFTQ1NdHY2AiwWX19/cKW5blhK+lMdsA3JJW9hKPScU16A93huCZJkiTdQzqvSach6V7g3pbXzWwmsGmLy4sk7dveMczsHOBrFYp2l/SP9va3KvQyW6bhab+KLJO0TXfa0VtpXvYuAwblKqkkSZL+QDqvSZfTESe1Rl/n4um6epxeZsuxPW3DqvLS5S8waGnXiJxteNrwLuk3SZIk6X4yFJEkSZIkSZL0GTLymiS9nJCx3VvS/hXKdgEu6unlAWbWjGeFeDcuHSZpXo0mSZIkSdIh0nlN+hRmNlDS8p62I3HMbJCkZXG6g6Q2CxskSZIkSUdI57WPY2Y3AIYnmH8eTzZ/C3CJpBlRZ2/gVEljI4XVNbjc6xxgc2CKpNur9L8L8FPgUVxWtBk4SNKzUX4GnhsVPB/pCdUcGDMbB5xXuDQCOEDSDDObAByPfyZfA46TpIg6jgdeBz4DjDezDYHzcbnVl4BjJD1vrv/aAKwdZQ2SLqpx75rxHLBfBYYBp0mabmbDgSckrR/13jsvH+O5VfcE1gIOBY7F5WXfwhW+FlMBM1sNuBTYFWgC3pD0RTMbhMvIDos+H4t5vV2hjynAQcCreDqyYll7nsfZwDBJJ8f5MEDAJ/HNX1Nwxa1BwI8k3Rz1HsI/O9sD/wTGVeo/SZIkSbqCXPPa9zlR0jYh4TofOAN34CYU6hzBCmWl63HHtg53Sse0YYwtgGkhaforXCoWM9sLd5R2wBPzDwS+V60TSXeWZVqBqbgTeE9I0R4I7CSpHrgQuLrQdHtgctj8Yszh0LDnRuCGqHc8MDMkYOtwIYDWWCJpTMzj4jbUB3cwZ0naihViA1PDnhLwrRptRwFjgRGSRgF7x/XlwCHx+r8Ov5dHtmxsZvsA+wKj8fvyuUJZu54Hrrx1UDjO4EpeM0Pu90ngSzHH3YCLzOzDhbafivKi4/pQCBScH6psSZIkSdLppPPa9znczEqhanUI7tTcCuxoZsMimrYzMN3MhuKO0Y0Akp7AFZpaQ5KeiuPZwKfjeDfgZklLJDXj6k+7tdaZme0BnALsK2kpsA/u1D1qZnOAC4BPFJrMkvSXON4OmCvpmTi/BhhtZkOAh4FJZvZDM9sV+Fcb5nZzYV4bhdRpa7wh6Y44fhJ4QdKcOC/h0exq/BVYHbjKzA4rXF8NmBzzfxqPzI6u0H4s8EtJb8TyiaKD3q7nEcpb81kROZ2If/EBVyv7daiB3YPL21qh+Y2F5QIAm4bjvRMeUa/lNCdJkiRJh0nntQ8TEcvjgD0j8no2sKak/wFm4M7sIcCMiKaVaa+s2tLC8XJWYbmJmY0CpuGv1l+OywOAq8tR2YicFvPCtmkdpaTpwI64rO2ZeIS2NZZG2/I62kHAMlb+3Wjp0DYVjpfTjvsj6TU8kn0zsCUwP5ZBHAJ8CdgxnuVlFcbtChqACWY2EvgQ8Ie4/p/4koSRESl/oYU9Kz0TSf8V/y4BrgS+2KVWJ0mSJB9Y0nnt26yHrw99JV7TFl8zN+CRtInEkoFwLOYDBwOY2db46+WOcj/wDTMbEqpak4D7qlU2s42B6cB4Sc8Vin6LR5A3iXoDzayaPt1sYJSZlV+XTwCekvS6mW0OLJbUAPwA2LaD81oMrB79gTuWnYKZbQCsLeke3MF+DX8Fvx7wcszjQzXGfBA40MzWMbOB+JKQMu16HsGteLT0VHyNcPmLzXrAwpD6/Qo1oslm9mEzWyuOBwH742tikyRJkqTTSee1b3M3HmV8Dvg9/gobAEmzgKHA0DguczhwUiwzmIynN3qtI4NLugv4BfBI9AO+yacak/DX0VNjbeQcMxsr6WF849RMM5sLNAL7VRnzJXxd541m9jS+mWt8FB8IzDOzp4BLgBM7OK9l0fY+M3sMj6Z2Fp8A7o95Pg3chTvk1wFDzOxPuDP/h0qNY2Pd7cDcaPdcoay9z4NClP6wsKHMmfg61zn4fa21vORz+JKP8pzeIZcNJEmSJF3EgObm9r5BTvoyZrYu8GZE1Ebgr4ZN0qs9a1nyQaVUKg0HFtTV1TF4cNfs8+ouedhSqUR9fbWXBn2LnEvvpL/Mpb/MA3IuXUFTUxONjY0Am9XX1y9sWZ6psj547ABcGK+VAY5KxzXp73SH45okSZJ0D+m8fsCQdC9wb8vrZjYT2LTF5UWS9m3vGGZ2DvC1CkW7S/pHe/tbFXrKFjObROWUWRMLmQm6hd70PHqC7oq6JkmSJN1DOq8JAB1xUmv0dS5wbmf1tyr0lC2SrsR33fc4vel51OKlK//MoKbOdzI3PGVEp/eZJEmS9BwZjkiSJEmSJEn6DBl5TZJejpl9H1hX0uQKZROBvSXt3912tbDju3jWh3dwKd9jJM3vSZuSJEmS/klGXpOkDURO1aQFZjbIzEYDxwBjQvL2V7jEb5IkSZJ0Ohl5TTqEmd2Ay4UOBp7HBRJuAS6RNCPq7A2cKmlspOW6BlgHT2C/OTAl8pZW6n8X4KfAo8AXcFWwgyQ9G+Vn4LlJAR4HTpBUUYnLzMYB5xUujQAOkDTDzCYAx+O/C68Bx0lSRDTH41HEzwDjQwnrfGAg8BIeXXzezAwXhVg7yhokXVTFlrWBa3GVrXdw6d0Do++b8Ny8awJ3SDq9Qvs18By2uwIvA08VygYCPwb2jEt3A2cU1MNa9nUlME/Sz+K8DpiJy/8ejOe6XSOqT5b0QNRbiCuE7Yrnk70Yl7xdG3gTV+p6odKYSZIkSbKqZOQ16SgnStompEznA2cQUqOFOkcQ6l64VOslkupwp3RMG8bYApgmaUs8mnc2gJnthTuuO+AKYQOpkRRf0p1l6VlgKvAEcE/I6x4I7CSpHo8WXl1ouj3utNUBL8YcDg17bgRuiHrHAzND1rYOuKrGnPbAhSNGRJTymLj+L2CfsGM0sI2Z7Vmh/THAZrgD/mVWVhE7OtpuHT9bxbVqNPD+51VW2boH2F7SVsBBuMNdZKikbSV9U9Jc4H8DC83s/0b979QYN0mSJEk6TDqvSUc53MxKodR1CO403QrsaGbDzGwYsDMw3cyGAnW4w4ekJ6it2FRGksqRxdl4RBBgN+BmSUvC0bo8rtXEzPYATgH2lbQU2AcYhatDzQEuwBWwysyS9Jc43g6YK+mZOL8GGG1mQ4CHgUlm9kMz2xV3RKsxF/i8mU01swOAprg+EM+/Oxco4fdrdIX2Y4FrJb0T6li/KJTthjufb0t6O2ysel9CeW2ImY0MWdeDWeGkfhp38OcDvwQ2jOhwmffUuMzsk7gi2uaSNsad4pbObpIkSZJ0Cum8Ju0mIpbHAXtG5PVsYM2C1Ogh8TND0puFpu2Vc1taOF7OKixzMbNRwDRgP0kvx+UBwNXlqGxETou5bisuQ2iJpOnAjrhU75l4hLZa3b/iEeX7cMdyrpmtiTvVHwa2i8jubfjyga7mWmAisBfwrKS/xfWbgMskbYFHcZe1sKd4bw7Alx/8Pc6vw53sJEmSJOl00nlNOsJ6+PrQV8xsML7etUwD7gxNJJYMSFqCLy04GMDMtsZf93eU+4FvmNmQUAqbhDuDFTGzjYHpwHhJzxWKfotHkDeJegPNrJou3mxglJl9Ls4nAE9Jet3MNgcWS2oAfsDKr/Jb2rIJsFzSbcDJwAbAR/B7+ndJS8Pe/ap08SBwWGyUWgv/klDmfmCCma1uZquHjVXvS3Ad/lwmsWKJB2HPgjg+El/bXI0FwJfMbJ04Hwc0tjJukiRJknSI3LCVdIS78c1Mz+Gbhh4mHDZJs2KZQPm1dJnDgavN7Dv4Jp95uAPcbiTdZWZbAo/EpSeAKTWaTMKdxKm+twqAkyX9zszOAmbGZqc18E1npQpjvmRmhwE3xiv2l/B7AL5u9lAzexuPLp9Yw5aRwAVhx0DgfEkvmtnFwC1m1ohvdnqgSvvLgS2BZ/F7/zjwsULZ5qzYxHUPcEUNW5C0yMyeAXYhvlwEJwG3mdmr+PN+pUY3t+LLKkpm1oQvmzii1rhJkiRJ0lEGNDe3901ukrQfM1sXeFNSc2QeeAgwSa/2rGVJT1MqlYYDC+rq6hg8uFaAt2N0pzxsqVSivr5a8L5vkXPpnfSXufSXeUDOpStoamqisbERYLP6+vqFLcsz8pp0FzvgG5IGxPlR6bgm3UF3Oa5JkiRJ95DOa9ItSLoXuLfldTObCWza4vIiSfu2dwwzOwf4WoWi3SX9o739rQq9zJZpeNqvIsskbdOddvQE3Rl1TZIkSbqHdF6THqUjTmqNvs4Fzu2s/laFXmbLsT1tQ1t4+epGBjW1Xq89fOyknn/9lSRJknQuGZJIkiRJkiRJ+gwZeU2SXkDI0e4taf8KZbsAF/X0a/6QBB4LfBwYUpbjNbPPAj+P68vwDAjHS3qrp2xNkiRJ+i8ZeU16HZG2KuklRGowcNnbSqpfbwOnSPocnsZrbWByN5mXJEmSfMDIyGsfICJehieKfx5PGn8LcImkGVFnb+BUSWMjFdU1wDrAHDz35xRJt1fpfxfgp8CjwBfwXKUHSXo2ys8ADovqjwMnlKNuFfoaB5xXuDQCOEDSDDObAByPf+5eA46TpIg6jgdeBz4DjA8p0vPxXKgvAcdIet48QWoD7iANxOVQL6px75qBs4CvAsOA0yRNN7PhwBOS1o96752Xj/EcqXsCawGHAsfi+UzfwpW6FlcZczXgUmBXXP71DUlfDCfwjrBjLeCxmNfbFfqYAhwEvIqnFSuWted5nA0Mk3RynA8DBHwS38Q1BVfOGgT8SNLNUe8h/LOzPfBPYJykB6NspTEkLQQWxvG7ZvYY8PlK9iRJkiTJqpKR177BiZK2CSnW+cAZuAM3oVDnCFYoJF2PO7Z1uFM6pg1jbAH/r707D5OrqvY+/g0JhClhiHiVMQiyMHQI0EQGmYLMTwjqlQAhJAHC6ABIFK4CFxUVX/CqQBQZA0oIIlwTGQwgFwElCA2ZAH8BJEQuRgG5hMEECHn/WKvISaeq0kN1d3WzPs/TD1Vnn7PP3nUq9Op99tmLyyM16S/xlK+Y2cF4oLQ7vsB+b+DcSpVIuqOUbhWYiAeB0yOl7EhgL0mNwEXANYVDdwUmRJtfjD4cHe2ZDNwQ+50KTItUrg34aOCqLJI0NPpxSQv2Bw8wH5S0Y5zjd8DEaE8T8MUqxw7Bb68PkjQEGB7blwKj4vZ/A/5ZHtf8YDM7FBiBj3LuCmxbKGvV9cAzaB1ZGD0dhX9+bwKPAXtEH/cDLjazDQrHfizKD6lSf/O2rxV9mtbSY1JKKaXWyOC1exhjZk1mNgcPPnbAsxrtaWYDYjRtb+CWyG7VgAd8SHoUmN2Cc0hSKTPTDGCreL0fMEXSIknL8CxO+62qMjM7EPgKMELSYuBQPKh72MxmAhcCmxUOeVDSs/F6F2CWpCfj/bXADmbWD8/mNd7Mvm1m++LZnFZlSqFfG5vZmi045g1Jt8frx4AXJM2M9034aHYlfwFWB66OrFwlqwETov+z8ZHZcrfhhwE3SXpD0lJWDNBbdT0kLcD/4CkFoOPwP3zAs479KrJ6TcfT1BaHVSdLerdKP1cQAfIU4F5JGbymlFLqEBm81rkYsTwFOChGXs8B1pT0FjAVD2ZHAVNjNK2ktanTFhdeL6UdU0rMbAhwOX5r/eXY3Au4pjQqGyOnxfVdy972bk7SLcCewLPA2fgI7aosjmOXxvs++INFxe9/84C2uGjTUlrx+Uh6DR/JnoLPAX0ipkGMAvYA9oxr+ZMy5+0Ik4CxZjYYWA94ILb/FJ+SMDhGyl9o1p4WXRN4f57yDfg0hy+3v8kppZRSeRm81r/18fmhr5hZX1a8zTwJH0kbR0wZkLQIH2k7CsDMdsJvL7fVPcARZtYvsmONB+6utLOZbQLcAoyWNK9Q9Bt8BHnT2K+3mVVahHMGMMTMSrfLxwKPS3rdzLYGFkqaBHwT+GQb+7UQWD3qAw8sa8LMNgLWljQdD7Bfw2/Brw+8HP1Yr8o57wVGmtk6ERQeWyhr1fUItwJ7AWfic4RLf9isD8yPlL37U300uaKY4zsJD+qPL9SfUkop1VwGr/Xvt/go4zzg9/gtbAAkPQj0B/rH65IxwOkxzWACMAcPoFpN0p3AL4CHoh7wh3wqGY/fjp5oZjPjZ5ik+/EHp6aZ2SxgLnBYhXO+hM/rnGxms/GHuUZH8Uhgjpk9DlwKnNbGfr0bx94dDxgtXcUhrbEZcE/0czZwJx6QXw/0M7M/48H8A+UOjgfrbgNmxXHzCmWtvR4URumPiTaUnI3Pc52Jf65Vp5eY2a1m9kKpWjObHq8Pxq/PYKAprvnEanWllFJKbdVr2bIcJOlpzGxd4M0YURuE3xo2Sa92bctSWllTU9NA4LmGhgb69u1b07o7Oz1sU1MTjY09I6tX9qU+9ZS+9JR+QPalIyxZsoS5c+cCbNnY2Di/eXkuldUz7Q5cFLeVAU7IwDV9EHVm4JpSSqlzZPDaA0m6C7ir+XYzmwZs3mzzAkkjWnsOMzsP+FyZogMk/aO19bVHV7XFzMZTfsmscYWVCTpFPV2PetHZo64ppZQ6RwavHyBtCVKr1PUt4Fu1qq89uqotkq4Crurs85ZTT9ejrV6+tok+S2o3jenfTvtUzepKKaVUP3JYIqWUUkopdRs58ppqJtK8Dpf0+TJl+wAXR3apulBsr5mNwNdf/WoXt+kGPEnBR4F+pbSvZrYN8LPY/i6eFvZUSf+qUtd9+GdeNi1wjdv9a2BL4D18fdgvdfbUiZRSSh8MOfLaQWJ9ztRNSJrWlYFrIX3r1ZTPuvU28BVJ2+KJD9bGl0GrF2Mj8cSOwMWsmPo3pZRSqplOG3mNESUD+gLP4Ivt3wxcKmlq7DMcOFPSsFji6VpgHWAmvoD6BZVGkWJk70fAw8BueIapIyU9FeVn4etcgo9afak0qlWmrkOA7xY2DQIOlzTVzMYCp+Kf3WvAKZIUo3ijgdeBjwOjI6vS9/D88y8BJ0l6xswMX9R97SibJOniKp/dMnyN1M8CA4CvSrrFzAYCj0r6UOz3/vvSa+BK4CBgLeBo4GQ8/eq/8AxYCyucczXgMjyF6RI8XeqnIsi6PdqxFvCn6NfbZeq4ADgSz7p0X7OyFl+PwmdwDvCZOPcJeFrUg/BUrIcXrnWla7QGvjbsvsDLwOOF+sdRGDVu5fflHGCApDPi/QBAwBbArvg6rGtGe74jaUrsdx/+3d4V+CdwiKR7o2yFc0iaD8yP1+/F2rSfqPR5FextZmcDGwO/lHR21H8mfm364NnDTimNlJrZbsBFQL+o46uS7orv7Y+ADwFrAD+SVEqOUVxHeD18BDallFKquc4ceT1N0s6RFvMJ4CwibWVhn2OJTFF42s9LJTXgvzCHtuAc2wGXS9oe+CUe7GBmB+OByO74Quq9gXMrVSLpjlIaU2AiHgROj1StI4G9JDXiv+CLI0y7AhOizS9GH46O9kzG02eCB1bTYqSqgRVz11eySNLQ6MclLdgfPMh7MEbDrgZ+B0yM9jRR/kn5kiH47etBkoYAw2P7UmBU3P5vwD/L45ofbGaHAiPwUcRdgW0LZa26HgX/F5/BWfii+3+Ivl2PB/es4hqdhN/aHgR8mgrZudrQvuuBIwujp6Pw6/smnlRij2jnfnhSgA0Kx34syg9pQf9L7VsL/8yntWD3zfHsWjsC483s46U2Sxoa7ToXT+eLmW0I/DfwtbjuOwGPRN8mA2fENdgDOLuQBQ0zu8rMFgDfYcV/1ymllFLNdOac1zFmdjQ+YrMOnjXoW8APY6QKYO/Yrz8eGE0GkPRoZFpaFUkqjabNAA6N1/sBUyJ1KmZ2BfDjVVVmZgcCX8HnQi6OgGwI8HCMjPUCioHIg5Kejde7ALMkPRnvrwV+Ymb9gPuB/2dmawP/Ez+rMqXQr43NbM1qO4c3JN0erx8DXijMQ2wC9q9y7F/wEc2rzexePOMT+B88EyLA6433/60yxw8DbirM2bya+GOCNl4P4KZCX5YVRuGbWL5MVLVrNAy4TtI7wDtm9gs8CGuuVe2TtMDMngAOwQPKccAZUbwRcE0Eje8CG+J3IGZE+eTI9tUiEUROAe6V1JLg9WZJ7wGvmdlTwFbA00CjmX092vMesE3svxvwpKQ/Rt+WAq/GnZBPAFMKo8J9Y9ufY9/x0cZj8D8aWhyQp5RSSi3VKcFrjIadAuwu6SUzGwWcKOktM5vK8hzvUyW9GcEr+K3/1lhceL2UdvTPzIbgo1EHSno5NvcCrpF0XoXDKt72Lopb/g8BB+ApOo9jefrTShbHsUsjeOiDB0PF0fPmAe2SwuultOLzkfSamW0H7IMHc983s53wgHcPPKB/PQKgbSrVU2Ol9i9l5b6V+rKqa9RRJgFjzew5/LZ5KfXrT/GA9nOR8WweK16nFn1n4P151Dfg0zC+3MLDVrrmMX3iV/jo9GNmtjHwv6uopxfwctyNqErSz83sCjMbIOmVFrYzpZRSapHOmjawPj738BUz68uKt5kn4SNV44gpAzHi9QRwFEAETYPbcf57gCPMrF9knRoP3F1pZzPbBLgFGC1pXqHoN/jI8KaxX28zq5RHbQYwpHBbdSzweAR8WwMLJU0CvkmF29ctsBBYPeqD5X8EtJuZbQSsLWk6HmC/ht/iXh8PYl43s/WqnPNeYKSZrRNB17GFslZdj1aqdo3uBY4xsz5x671S29vSvlvx2/Nn4nOYS394rQ/Mj8B1f3zudqvFHORJeAB6fKH+tijNv/1rvD+1UPYQMCjmvZY+vw3wObxvxahqqU3bmll/M1vXzDYrbD8Un8P7z3a0MaWUUiqrs4LX3wLP4lMFfo/f9gVA0oNAf6B/vC4ZA5xuZnPwp6rn4AFUq0m6E/gF/ot5Tmy+oMoh4/HbvRPNbGb8DJN0Pz63cpqZzQLmAodVOOdL+LzJyTHlYTTLR1dHAnPM7HH8AaLT2tivd+PYu+MBnqVtqaeCzYB7op+zgTvxgPx6oJ+Z/RkPFB8od3Dc0r8NmBXHzSuUtfZ6tNgqrtEVwALgKTyQfaRCHa1un6S38Hm4x+CfUcnZ+DzXmfh1rzr9xcxuNbMXStWa2fR4fTD+/RkMNMV3cmK1uqq0kQJzzgAAHVNJREFUdRFwHj6XtQl4s1D2T3wKxn/F97YJaIzv2qH43N7ZMU3iJyyfBnSzmc2Jfp4BHNrOADullFIqq9eyZfX5+8XM1gXejBGrQfjT6ibp1a5tWUqplpqamgYCzzU0NNC3b9+a1dsV6WGbmppobKx0M6Z7yb7Up57Sl57SD8i+dIQlS5Ywd+5cgC0bGxvnNy+v5yQFuwMXxW1bgBMycE0ptVRnB64ppZQ6R90Gr5LuAu5qvt3MpuHL/xQtkDSitecws/NY/pR60QGS/tHa+tqjq9piZuMpv2TWOHVyhqR6uh7l1GP7bOU1iUu+LumOzm5PvVj27lJ69ck8ISml1BPVbfBaSVuC1Cp1fQtfrqvLdVVbJF0FXNXZ5y2nnq5HOfXYvghQe0SQ+sp1f6TPktpMY/rwl/atST0ppZTqT95XSymllFJK3Ua3G3lN9SOeLN9N0r+6ui1dJZYUmws8UEgtewB+K38wniVuQjvPcQHweeAlSXu24fiB+NSGK9rRhvOBdSVNMLMdgG0k/bJQfjHw78BAYLCkuW09V0oppVRNjrymNosUuh+4wLWQBhZ8uajmt+3/gi+3dlGNTnkmnhSi1YFrGAicWKO2gKf8Hdls26/xdW6fr+F5UkoppZXkyGtqMzNbBvST9IaZzcfXRv00sAm+vumH8UQAGwLHSbo/RgEfBa7Ds3X1Ak6VVHa92DjPv+HZzraK/S+SdH2UzcfTpe6PZ7b6kaTLqtT1NPB5SbPi/RfxdUyPjdHDvfG1S1+ONj9faPMkYF98vdjLzdMd/z3KhpfOIemZqPszq/gIi+06BF+fdk3gbeAMSTPM7IHY9jszmy7pq2Y2Fk8s0Adf+/gUSYp6/gP/zN/D12/dA5gIbBkj5c+URojLtOF8YnS13PvYNgCf99s/6rtf0pdLazQXUsemlFJKHSJHXlMt9ZW0G377+ErgHUmfBL7Oik/EDwBmSdoe+BJwY2Req+QSYG7sfwBwoZk1FMo/LKkR+BTwdTPbvkpd1+HZzkqOJTK7ARdKGippCHAj8P1mbX5E0k6SLo+Uql/Bg/R2MbOtgHOBg6Mf44FfAhRGW3ePwHVPfNRzr9j3IuCaqGcsMCL2HYInCngP+ALwZIyUlw1cWyrSvZ4H3BP1tTRNbUoppVQTGbymWrop/vsYsHbhfRMrpkV9Gx+lRdJ9wL+AakN2+wE/i/3/ht+mH1YovzrK/g7cDuxTpa7rgaMiRexgPH1radT3YDObYWZz8axuOxSOW0wElOFK4GuS3qhyrpY6EB9Vvj9GM28A+sSIc3OHAkOAh2PfC/FsaOCjvz+V9Dq8H2imlFJKPUpOG0i1tBhA0tK4fbw4ti+lTr5rkhZEatOD8SB3UmRx2wL4ITBU0nNmtjswuXDom83Sne4GXB39XBdYy8zukHRIG5rVC/itpDEt3PcaSee14Tyr8i4r/kG7ZgecI6WUUmqXHHlNXWENfF4mcRt8LeDPVfa/Bzgh9v8IcAhwb6F8XJRtFGX/s4rzT8JvzR+FTyMA6I+PCC80s9WAk6tVIGlDSQMlDcRHae9sY+AKnozjIDPbrrTBzIZW2Pc3wBgz2zT2621mpVx+twGnmFm/KBsQ2xfh84FX5Rmg0cxWizqGV9ivpfWllFJKNZfBa+oKrwA7mNls/Gn9oyS9XWX/LwNDYv+7gbMlPVEof9nMmoCHgO9JmrOK89+Kj7o+KWkBQBxzM/Ak8DDwXOu75cxsDzN7AZ8Te5KZvWBmB1baX9LTwGh8JHeWmT0FnFRh3/vxB7ummdksfJmuw6L4ejy4nRFTCqZGID4bkJnNNbNfVWn6rcA/gafidVOF/X4HrBNtvST6fEn0eVPgnhjdTimllGqu17Jltclok1JLlJ7cl/ShGtU3Hxie64p2X01NTQOB5xoaGujbt9pzey3XVelhm5qaaGxsXPWO3UD2pT71lL70lH5A9qUjLFmyhLlz5wJs2djYOL95eY68ppR6nK4IXFNKKXWOuniIJn1wSJoPrDTqGlmbJpU55DJJV1Wpb2C57Wb2KCt/v2dIqjqXtSOZ2eXArs02vytp505uR919NimllFJLZfCa6oKkmay4NFV76+vUgLAl6iU4rMfPppa6aspASimlzpHBa0qpLrxy/f/QZ8l77a7nw19s66IPKaWUuoMMXlOqA2Y2Dn/wbKUMWGa2D3BxV46Ymtk2eKKIj+LrwT6Cp/X9V5QPxFeO2Apf1/cHkq7umtamlFLqyfKBrVR3zCzv+dYRM+uDr4H7FUnbAtvjGdQmRHkv4L+BKyQZsB2+ZFdKKaVUczny2g2Y2Q14+tS++ELyx+Frkl4qaWrsMxw4U9IwMxsEXAusA8zEU7NeIOm2CvXvA/wIX990N2AZcKSkp6L8LOCY2P0R4EuV0qKa2SHAdwubBgGHS5pqZmOBU/Hv3WvAKZIUo46jgdeBjwOjIxnB94DewEvASZKeMU9pNQkPnnrjGbIurvLZLcPXRf0sMAD4qqRbmi/ZVXxfeo2ngD0IT6JwNJ64YBc8ne1hkhZWOOdqwGXAvsAS4A1Jn4og8PZox1rAn6JfK61xa2YXAEcCrwL3NStrzfU4Bxgg6Yx4PwAQsAX+8NgFeCatPsB3JE2J/e7Dvzu7Av+MBAzzASS9Z2Z/Aj4Rp9kPeF3Sr6N8GfCPcu1JKaWU2itHXruH0yTtLGkw8ARwFh7AjS3scywesAL8HA9sG/CgtFK2pqLtgMslbQ/8EjgHwMwOxgOl3YHBeMB4bqVKJN0haQdJOwAT8SBwemTSGgnsJakRuAi4pnDorsCEaPOL0Yejoz2TgRtiv1OBaZKGxL4tuTW9SNLQ6MclLdgfPMB8UNKOcY7fAROjPU3AF6scOwQYBgySNITlmaqWAqPi9n8D/lke1/xgMzsUGIE/wLYrsG2hrFXXA09ccGQEzuCZzaZJehN4DNgj+rgfcLGZbVA49mNRvsIkUjNbK9o9LTYNAl4xs5vN7PH472ZV2pRSSim1WQav3cMYM2syszl48LEDngFpTzMbEKNpewO3mFl/PDCaDCDpUTzD0qpI0uPxegY+dxE8qJkiaVGMqF0R26qKjFJfAUZIWgwcigd1D0f2pwuBYoDzoKRn4/UuwCxJT8b7a/GMXP2A+4HxZvZtM9sX+L8W9G1KoV8bm9maLTjmDUm3x+vHgBdiRQTw4HXrKsf+BVgdz5h1TGH7asCE6P9sfGS23AoLw4CbJL0haSkrBuituh6RQewJPG0ueCrdSfF6I+BXZjYXmA5siI/wl0yW9G6xvgiCpwD3SioFr72jL+dGIPwQy9PuppRSSjWVwWudixHLU4CDYuT1HGBNSW8BU/FgdhQwNUbTSlqbOm1x4fVS2jGlxMyGAJfjt9Zfjs29gGtKo7Ixcrp54bCyt72bk3QLsCfwLHA2PkK7Kovj2KXxvg/+0FHx+988oF1SeL2UVnw+kl7DR7Kn4PNDn4hpEKOAPYA941r+pMx5O8IkYKyZDQbWAx6I7T/FpyQMjpHyF5q1Z4VrEnORb8CnMny5ULQAaJL053j/C+CTte1CSiml5DJ4rX/r4/NDXzGzvqx4m3kSPpI2jpgyIGkRPtJ2FICZ7YTfXm6re4AjzKxfPJgzHri70s5mtglwCzBa0rxC0W/wEeRNY7/eZlYpB90MYIiZlW6XjwUel/S6mW0NLJQ0CfgmbQ+SFgKrR33ggWVNmNlGwNqSpuMB9mv4Lfj1gZejH+tVOee9wEgzWycCxmMLZa26HuFWYC/gTHyOcOkPm/WB+ZKWmdn+VBlNjnm8k/DA/fhCHQB3ApuZ2Ufj/UHArFW0KaWUUmqTDF7r32/xUcZ5wO/xW9gASHoQ6A/0j9clY4DTY5rBBGAOHkC1mqQ78ZG0h6Ie8Id8KhmP346eaGYz42eYpPvxB6emmdksYC5wWIVzvoTP65xsZrPxh7lGR/FIYI6ZPQ5cCpzWxn69G8feHQ8fLV3FIa2xGXBP9HM2HtzNwOef9jOzP+PB/APlDo4H627DA8AZ+LUvlbX2elAYpT8m2lByNj7PdSb+uVabXnIwfg0GA01xXSdG/W8CXwLujD6Pi5+UUkqp5notW9bau8up3pnZusCbMaI2CL81bJJe7dqWpbSypqamgcBzDQ0N9O3bt931dWWGraamJhobK91Q6F6yL/Wpp/Slp/QDsi8dYcmSJcydOxdgy8bGxvnNy3OprJ5pd+CiuK0McEIGrumDIlPDppRSz5bBaw8k6S7grubbzWwasHmzzQskjWjtOczsPOBzZYoOkNSpa3x2VVvMbDzll8waV1iZoFPU0/VIKaWUOlIGrx8gbQlSq9T1LeBbtaqvPbqqLZKuAq7q7POWU0/Xo6t15bSBlFJKHS+D15RSXXjl57+lz5L2Pzf34S+UG4BOKaXUU2TwmmrGzM4H1pU0oUzZOGC4pM93drsqKbbXzE4G1pL0wy5sT198VYCdAUqpa6Nsd+BifHkr8DSzX2u2ZFXz+ubjn/ncDmpy6TyrAX/AU/YC/A04WdL8jjxvSimlD6ZcKquHiXVBUytJuryLA9c++HJdF1M+Y9YiYKykQcCOwG4sXz6sS0l6D0+iMSTS4d4J/FcXNyullFIP1eNHXs3sBjzlZV/gGXyR/5uBSyVNjX2GA2dKGhZLS10LrAPMxBduvyDW3ixX/z7Aj4CH8YBiGXCkpKei/Cx8fU2AR4AvSSqbTcrMDgG+W9g0CDhc0lQzGwucil+z14BTJClGNEcDrwMfB0ZHNqfv4Wk7XwJOkvSMmRm+0PzaUTZJ0sUV2rI2nuJzO+AdPH3syKj7Rnx92TWB2yV9rczxa+DrsO4LvAw8XijrDXwfX8wefC3bswoZsJrXNRB4FLgyjlkLOBo4GU8l+y88m9fC2P8s4N/js/pffLWFhZEY4Go8fe5C4K/A3+OY81k+Ctva9l0FzJH043jfAEzDU+weha8nu0bsPkHS72K/+XgWrn3j+OPx9WEHNj9HcfRU0pJY53aLcu1pZqSZXQl8FLhY0mVx7ovxlMJr4NfnOEnPR9lw4Hw8xe17eNA828x2wdP69o+6zyul0I2sYiX947iUUkqp5j4II6+nSdo50nE+AZxFpMss7HMskaEKTzd6qaQGPCgd2oJzbAdcLml74Jd4ClfM7GA8cN0dX9y9N3BupUok3VFKnwpMxAO26ZEidiSwl6RG4CLgmsKhu+JBUQPwYvTh6GjPZDylJ3jwOy1GyBrwQK6SA/HkB4NiNO2k2P5/wKHRjh2Anc3soDLHnwRsiQfgn2bFTFgnxrE7xc+Osa2aAcCDknaMdv8OmBh9bCKe+jez0XjQuKuknYA7gB9EHecBiyRtC3weD97KaW37JrHy96mUyWp6tGVH4Ej8D4Ki/pI+GYFri5jZh/Hg/PYW7L62pN2AfYALYw1ggAslDY1reyMerGNm2+APoR0VZbsCz5nZ+njK31Fx7YcDP4vtpXbdYWYLgSNYMX1sSimlVDMfhOB1jJk1RbapUXhQciuwp5kNMLMBeBBzi5n1x0flJgNIepTqWYdKJKk0sjgDD57Ab/9OkbQoApkrKH9LeAVmdiDwFWCEpMXAocAQ4OHIhnQhnsWp5EFJz8brXYBZkp6M99cCO5hZP+B+YLyZfdvM9sUD0UpmAZ8ws4lmdjiwJLb3xteQnYUHjQ34Z9rcMOA6Se9EhqdfFMr2w4O7tyW9HW1c1efyRmmUD88y9kJhOaomlqc2HRF1PRaf1ReAgYU2XQ0g6WX8e1BOq9oX2c36mdnguP1/FMuD1K3wP0CeAG4CPhKj1yXX0wpxHacBPyh856qZEm2cD7wKbBrbDzazGWY2F8/CVrqG+wN3SHo6jlsi6XX8D7At8SxaM/GpAcsopJSVdAiwMR4Mn9OafqWUUkot1aOnDcSI5SnA7pJeMrNRwImS3jKzqSzPLT9V0psRvIL/Um6NxYXXS2nH52pmQ/ARrgMjwALoBVwj6bwKh5WdhtCcpFvM7CHgADw16HFUmDcp6S9mth0+anow8F0zG4wH1RsAu0habGZX4NMHOtqSwuulVP7Me+HTPIoj053hOjwl6n3AU6Vb8Hggd6akX8eDTW+x4ufVomsH70/luA24S9IPVrV/WOlzMrMtgB8CQyU9Fw+DTV5FPb2A2ZL2qraTpPfM7GrgaXykP6WUUqqpnj7yuj4+P/SVeJL7uELZJJbnYL8WQNIifGrBUQBmthN+u7+t7gGOMLN+ke1qPHB3pZ3NbBPgFmC0pHmFot/gI8ibxn69zaxS/rYZwBAz2zbejwUel/S6mW0NLJQ0CfgmK97Kb96WTYGlkn4NnAFsBGyIf6Z/i8B1E+CwClXcCxxjZn3MbC2W/6EA/rmMNbPVzWz1aGPFz6WVpgGnmtkG0Y++8QdBqU3HxvYBwGcr1NGW9l2Pf2/Gs3wKCvjn9Vy8Pg6fe91qZrYm/j2YUeWPmJbqD7wNLIyA+uRC2V3AIWb28Thv3xjt/SPwcTMbVmjTUDPrZWYbmdmHCnUcDsxpZxtTSimlsnp68Ppb4FlgHvB7/HYz8P6t3v74nMMHC8eMAU6PaQYT8F/CxYdRWkzSnfjt8odY/sv8giqHjMeDxIlmNjN+hkm6H/gGMC1u18+lQtAo6SV8nu1kM5uNj6yWRldHAnPiYZ9L8QeJKhkMPBTn+xPwPUkvApcAn4rbzaW5p+VcASwAnsKDxkealc3GH+J6PF5fWaUtLSbp5/gc399H/5uAT0Xxt4ENzOzP+B8J91dpe6vaJ2kB8CQ+t7Q4HeF04Ndm9hjwMeCVavWY2SP492UDM3shHgYDOD7qPrDw3fhGtbqqtHUO/tDik/iDhs8Vyp4GTgBuimv/EDBQnl54BPCfZjbLzJ7CH+rqBXwEnxoxO/7d7E+drISQUkqp5+m1bFlr75D3bPFAy5uSlpmvPHAfYPHLO6VUY01NTVsBz2yzzTasscYaq9x/VZYtXUqv3l2zYtzcuXNpaGjoknPXWvalPvWUvvSUfkD2pSO8/fbbzJs3D2DrxsbGZ5uXZ/DajJkdgD/N3ys2nVtaUiulVHtNTU17AA90dTtSSinVnT0bGxsfbL6xRz+w1RaS7sLn/a3AzKYBmzfbvEDSiNaew8zOA8rlsDxA0j9aW1971FlbLseXZip6V9LOndmOSuqxfWY2nlgmrJlxhdUY6t0jwJ54Zq7254dNKaXU3fXG1yd/pFxhjrymlFJKKaVuo6c/sJVSSimllHqQDF5TSimllFK3kcFrSimllFLqNjJ4TSmllFJK3UYGrymllFJKqdvI4DWllFJKKXUbGbymlFJKKaVuI5MUpJQ6hJltA1wHDABeAcZIerrZPr2BS4CDgGXAhZKuWlVZZ6tBXw4AvgsMBi6VNKETm7+CGvTlXOBIPKHEO8DXJU3vvB6838b29uNY4AzgPXxB9CslXdJ5PVihne3qS2EfAx4HftJV37EaXJfzgVOBF2P3P0j6Que0fkW1uC5mNhI4F8/auQzYT9LfO6cH77ehvdfkemD7wu7bA5+RNK0Tml9WjrymlDrK5cBESdsAE4GfldnnaGBr4OPAbsD5ZjawBWWdrb19+QswHk893dXa25c/AUMlbQ8cB9xkZmt1eKtX1t5+3AIMkbQDsDtwppltX6aOztDevpSCj58Bv+7w1lbX7r4A10vaIX66JHAN7eqLme0MnA/sL6kB2AN4reObvZJ29UPSmNL1AMYCrwKd/gdrUQavKaWaM7MPAzsBN8amG4GdzGyjZrsegY94vSfpJfwX7+EtKOs0teiLpGciXe+7ndTssmrUl+mS3or9ZuMjSgM6vPEFNerHIkmlFJNrA6vjI06dqkb/VgDOBm4D5nVwkyuqYV+6XI36cgZwsaSFAJJek7S441u/XAdck+OBGyQt6ag2t0QGrymljrAZ8L+SlgLEf1+M7UWbA88X3i8o7FOtrDPVoi/1otZ9GQM8K+mFDmhrNTXph5mNMLMnYp+LJM3p0FaX1+6+mNkQ4EDghx3e2upq9f060sxmm9ldZrZbRza4ilr0ZRDwMTO738weM7NzzKxXB7e7uZr9mzezNYBRwDUd1toWyuA1pZRSq5nZ3sC3gaO6ui1tJWmapO2AbYBjYs5ot2JmqwNXACeXApRu7nJgy5iWchEw1cw6dWS/hnrj80P3B/YGDgaO6dIWtc9ngAVxF6lLZfCaUuoIfwU2iXl4pfl4G8f2ogXAFoX3mxf2qVbWmWrRl3pRk77EaNgv8Ic21KEtLq+m10TSAnwu7/AOaW117e3LR4GtgDvMbD5wOnCCmV3Rsc0uq93XRdJCSe/E67tje0MHt7ucWv0/7FeSlkh6HZgKfLJDW72yWv5bOY46GHWFDF5TSh1A0j+AmSwflTsKeDzmUhXdjP+iXS3mYH0G+FULyjpNjfpSF2rRFzMbCtwEfF7SY53T8hXVqB+fKO1kZh8ChgGdPm2gvX2RtEDShyQNlDQQ+BE+d/HETurC+2p0XTYp7WRmOwADgU7/A6lG/+4nAweYWa8YIf80MKvjW79crf7/ZWabAnsCN3R8q1ctl8pKKXWUk4HrzOw8/OnUMQBmdgdwnqRHgZ8DuwClZVu+Jem5eF2trLO1qy9mtgcwBegP9DKzI4Hju2KJKdp/XX4CrAX8rHCX/ZgumC/a3n6cGEuYvYM/dHaZpLs6swMF7e1LPWlvX75rZo34Umxv49+thZ3ZgYL29mUKsDPwJL4k23Tg6s5r/vtq8f0aC/xG0qud1+zKei1b1ukPV6aUUkoppdQmOW0gpZRSSil1Gxm8ppRSSimlbiOD15RSSiml1G1k8JpSSimllLqNDF5TSimllFK3kcFrSimllFLqNjJ4TSmllAAzm29m+9VBO+4zs/Fd3Y6U6lUmKUgppZTqgJn1whMmpJSqyCQFKaWUEj7yCowHNgVOAP4EHAv8ExgNbAN8G+gLfFXSdXHcJGAxsBWwK/AYMEbS81G+O/DjOH4ecJqkP0bZfcAfgH2AnYBbgSPxzF/vApMkfdHMfgx8DlgPz4J0uqQHoo7zgUHRhs/ieerHRuYkzGyzOP+e+B3XGyV9McqOA74KfCT6e2Kp3SnVq5w2kFJKKa1sF2A2MADPUT8FGApsjQeyl5nZuoX9j8YD2w/hueRvADCzDYHbgUuirv8CbjezAYVjjwFOBPoB44AHgC9KWrcUZAKPADsAG0Z7bjazNQt1jIg2rg9MAy6L8/cGbgOeBwYCm8R+mNlhwNfxoHijOO+Nrf+oUupcOW0gpZRSWtlzkq4FMLObgG/g+d6XAHeZ2dt4IDsz9r9d0v2x/zeA12LEcx/gaUk/j/1uNLMvA4cCk2LbJElPlE5sZis1RtIvCm9/YGbnAAbMim0PSrojjv85cHps/ySwMT5S/G5p3/jvycD3JD0Vx30X+LqZbZGjr6meZfCaUkoprezvhdf/ApDUfFtx5PWvpReS3jCzf+JB48b4qGfR8/gI6ErHVmJmE4Djo75lQH98lLdkYeH1W8CaZtYH2Ax4vhC4Fm0B/NjMflDY1ivalsFrqlsZvKaUUkrtt1npRUwn2BB4MX62aLbv5sBvC++bP3yywnsz2xP4GvBp4AlJ75nZq7Ts4a6/ApubWZ8yAexfge9IuqEF9aRUN3LOa0oppdR+h5jZHma2Bj73dYakvwJ3ANuY2Sgz62NmR+APV91Wpa6/Ax8rvO+HP7z1EtDHzM7DR15b4k/A34ALzWwdM1vTzD4VZZcD/2Fm2wGY2XpmdngL602py2TwmlJKKbXfZOA/8ZUJGvGHupD0CjAcOBN4BR9BHS7p5Sp1/Rj4vJm9amaXANPxkdp5+O38xbRgqkGcfyk+v3ZrfBWCF4Ajouy/ge8DU8xsETAXOLjlXU6pa+RSWSmllFI7xFJZL0g6p6vbktIHQY68ppRSSimlbiOD15RSSiml1G3ktIGUUkoppdRt5MhrSimllFLqNjJ4TSmllFJK3UYGrymllFJKqdvI4DWllFJKKXUbGbymlFJKKaVuI4PXlFJKKaXUbfx/5IIASZK5RMkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 504x864 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XxwkRX5gnXWi",
        "colab_type": "text"
      },
      "source": [
        "We will create new datasets with less features."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CfArYVshrZNh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2b1b60ee-54b9-4b04-f033-c9d3aeab7b83"
      },
      "source": [
        "#getting top k features\n",
        "top=250\n",
        "features = get_top_features(X_train,model.feature_importances_,top=top)\n",
        "print(\"Taking top %i features\"%(top))\n",
        "X_train_200 = X_train[features].copy()\n",
        "X_val_200 = X_val[features].copy()\n",
        "X_test_200 = X_test[features].copy()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Taking top 250 features\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "xA9bSzUARRZ_"
      },
      "source": [
        "# XGB Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "3UWwclfyRRaC",
        "colab": {}
      },
      "source": [
        "model = xgb.XGBClassifier(n_jobs=-1,\n",
        "                      nthread=-1,\n",
        "                      scale_pos_weight=1,\n",
        "                      learning_rate=0.02,  \n",
        "                      colsample_bytree = 0.5,\n",
        "                      subsample = 0.9,\n",
        "                      objective='binary:logistic', \n",
        "                      n_estimators=301,\n",
        "                      reg_alpha = 0.3,\n",
        "                      max_depth=5, \n",
        "                      gamma=5,\n",
        "                      random_state=42\n",
        "                     )\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T0CEyW37Dai9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "3f158cb4-0845-464f-d1ba-ae1a1af3c308"
      },
      "source": [
        "#https://towardsdatascience.com/fine-tuning-xgboost-in-python-like-a-boss-b4543ed8b1e\n",
        "\"\"\"\n",
        "manually changed each parameters 1 by 1 and compared the validation auc\n",
        "and selected the best parameter value\n",
        "\"\"\"\n",
        "eval_metric = ['auc','error']\n",
        "%time model.fit(X_train,y_train,eval_set=[(X_val,y_val)],eval_metric=eval_metric,verbose=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0]\tvalidation_0-auc:0.808495\tvalidation_0-error:0.039551\n",
            "[1]\tvalidation_0-auc:0.810294\tvalidation_0-error:0.039551\n",
            "[2]\tvalidation_0-auc:0.81366\tvalidation_0-error:0.039551\n",
            "[3]\tvalidation_0-auc:0.815861\tvalidation_0-error:0.039551\n",
            "[4]\tvalidation_0-auc:0.815493\tvalidation_0-error:0.039551\n",
            "[5]\tvalidation_0-auc:0.819946\tvalidation_0-error:0.039551\n",
            "[6]\tvalidation_0-auc:0.818774\tvalidation_0-error:0.039551\n",
            "[7]\tvalidation_0-auc:0.818047\tvalidation_0-error:0.039551\n",
            "[8]\tvalidation_0-auc:0.817864\tvalidation_0-error:0.039551\n",
            "[9]\tvalidation_0-auc:0.819897\tvalidation_0-error:0.039551\n",
            "[10]\tvalidation_0-auc:0.81771\tvalidation_0-error:0.039551\n",
            "[11]\tvalidation_0-auc:0.820205\tvalidation_0-error:0.039551\n",
            "[12]\tvalidation_0-auc:0.820502\tvalidation_0-error:0.039551\n",
            "[13]\tvalidation_0-auc:0.821995\tvalidation_0-error:0.039551\n",
            "[14]\tvalidation_0-auc:0.822118\tvalidation_0-error:0.039551\n",
            "[15]\tvalidation_0-auc:0.824068\tvalidation_0-error:0.039551\n",
            "[16]\tvalidation_0-auc:0.822585\tvalidation_0-error:0.039551\n",
            "[17]\tvalidation_0-auc:0.823451\tvalidation_0-error:0.039551\n",
            "[18]\tvalidation_0-auc:0.823644\tvalidation_0-error:0.039551\n",
            "[19]\tvalidation_0-auc:0.823173\tvalidation_0-error:0.039551\n",
            "[20]\tvalidation_0-auc:0.822466\tvalidation_0-error:0.039551\n",
            "[21]\tvalidation_0-auc:0.823177\tvalidation_0-error:0.039551\n",
            "[22]\tvalidation_0-auc:0.823359\tvalidation_0-error:0.039551\n",
            "[23]\tvalidation_0-auc:0.823537\tvalidation_0-error:0.039551\n",
            "[24]\tvalidation_0-auc:0.823866\tvalidation_0-error:0.039551\n",
            "[25]\tvalidation_0-auc:0.823872\tvalidation_0-error:0.039551\n",
            "[26]\tvalidation_0-auc:0.823551\tvalidation_0-error:0.039551\n",
            "[27]\tvalidation_0-auc:0.823646\tvalidation_0-error:0.039551\n",
            "[28]\tvalidation_0-auc:0.82497\tvalidation_0-error:0.039551\n",
            "[29]\tvalidation_0-auc:0.825182\tvalidation_0-error:0.039551\n",
            "[30]\tvalidation_0-auc:0.82505\tvalidation_0-error:0.039551\n",
            "[31]\tvalidation_0-auc:0.82509\tvalidation_0-error:0.039551\n",
            "[32]\tvalidation_0-auc:0.824789\tvalidation_0-error:0.039551\n",
            "[33]\tvalidation_0-auc:0.825102\tvalidation_0-error:0.039551\n",
            "[34]\tvalidation_0-auc:0.82478\tvalidation_0-error:0.039551\n",
            "[35]\tvalidation_0-auc:0.825454\tvalidation_0-error:0.039551\n",
            "[36]\tvalidation_0-auc:0.824934\tvalidation_0-error:0.039551\n",
            "[37]\tvalidation_0-auc:0.825259\tvalidation_0-error:0.039551\n",
            "[38]\tvalidation_0-auc:0.82527\tvalidation_0-error:0.039551\n",
            "[39]\tvalidation_0-auc:0.825285\tvalidation_0-error:0.039551\n",
            "[40]\tvalidation_0-auc:0.824697\tvalidation_0-error:0.039551\n",
            "[41]\tvalidation_0-auc:0.824437\tvalidation_0-error:0.039551\n",
            "[42]\tvalidation_0-auc:0.824922\tvalidation_0-error:0.039551\n",
            "[43]\tvalidation_0-auc:0.824837\tvalidation_0-error:0.039551\n",
            "[44]\tvalidation_0-auc:0.82603\tvalidation_0-error:0.039551\n",
            "[45]\tvalidation_0-auc:0.826818\tvalidation_0-error:0.039551\n",
            "[46]\tvalidation_0-auc:0.827084\tvalidation_0-error:0.039551\n",
            "[47]\tvalidation_0-auc:0.82738\tvalidation_0-error:0.039551\n",
            "[48]\tvalidation_0-auc:0.827452\tvalidation_0-error:0.039551\n",
            "[49]\tvalidation_0-auc:0.827421\tvalidation_0-error:0.039551\n",
            "[50]\tvalidation_0-auc:0.827958\tvalidation_0-error:0.039551\n",
            "[51]\tvalidation_0-auc:0.828341\tvalidation_0-error:0.039551\n",
            "[52]\tvalidation_0-auc:0.828001\tvalidation_0-error:0.039551\n",
            "[53]\tvalidation_0-auc:0.828216\tvalidation_0-error:0.039551\n",
            "[54]\tvalidation_0-auc:0.828387\tvalidation_0-error:0.039551\n",
            "[55]\tvalidation_0-auc:0.828123\tvalidation_0-error:0.039551\n",
            "[56]\tvalidation_0-auc:0.828346\tvalidation_0-error:0.039551\n",
            "[57]\tvalidation_0-auc:0.828617\tvalidation_0-error:0.039551\n",
            "[58]\tvalidation_0-auc:0.828353\tvalidation_0-error:0.039551\n",
            "[59]\tvalidation_0-auc:0.828681\tvalidation_0-error:0.039551\n",
            "[60]\tvalidation_0-auc:0.828711\tvalidation_0-error:0.039551\n",
            "[61]\tvalidation_0-auc:0.829128\tvalidation_0-error:0.039551\n",
            "[62]\tvalidation_0-auc:0.82992\tvalidation_0-error:0.039551\n",
            "[63]\tvalidation_0-auc:0.829949\tvalidation_0-error:0.039551\n",
            "[64]\tvalidation_0-auc:0.830168\tvalidation_0-error:0.039551\n",
            "[65]\tvalidation_0-auc:0.830214\tvalidation_0-error:0.039551\n",
            "[66]\tvalidation_0-auc:0.830805\tvalidation_0-error:0.039551\n",
            "[67]\tvalidation_0-auc:0.83111\tvalidation_0-error:0.039551\n",
            "[68]\tvalidation_0-auc:0.831255\tvalidation_0-error:0.039551\n",
            "[69]\tvalidation_0-auc:0.831025\tvalidation_0-error:0.039551\n",
            "[70]\tvalidation_0-auc:0.83143\tvalidation_0-error:0.039551\n",
            "[71]\tvalidation_0-auc:0.831802\tvalidation_0-error:0.039551\n",
            "[72]\tvalidation_0-auc:0.831777\tvalidation_0-error:0.039551\n",
            "[73]\tvalidation_0-auc:0.831829\tvalidation_0-error:0.039551\n",
            "[74]\tvalidation_0-auc:0.831964\tvalidation_0-error:0.039551\n",
            "[75]\tvalidation_0-auc:0.832187\tvalidation_0-error:0.039551\n",
            "[76]\tvalidation_0-auc:0.832346\tvalidation_0-error:0.039551\n",
            "[77]\tvalidation_0-auc:0.832293\tvalidation_0-error:0.039551\n",
            "[78]\tvalidation_0-auc:0.83277\tvalidation_0-error:0.039551\n",
            "[79]\tvalidation_0-auc:0.832578\tvalidation_0-error:0.039551\n",
            "[80]\tvalidation_0-auc:0.832958\tvalidation_0-error:0.039551\n",
            "[81]\tvalidation_0-auc:0.833233\tvalidation_0-error:0.039551\n",
            "[82]\tvalidation_0-auc:0.833312\tvalidation_0-error:0.039551\n",
            "[83]\tvalidation_0-auc:0.833081\tvalidation_0-error:0.039551\n",
            "[84]\tvalidation_0-auc:0.833123\tvalidation_0-error:0.039551\n",
            "[85]\tvalidation_0-auc:0.833139\tvalidation_0-error:0.039551\n",
            "[86]\tvalidation_0-auc:0.833318\tvalidation_0-error:0.039551\n",
            "[87]\tvalidation_0-auc:0.833137\tvalidation_0-error:0.039551\n",
            "[88]\tvalidation_0-auc:0.833691\tvalidation_0-error:0.039551\n",
            "[89]\tvalidation_0-auc:0.833322\tvalidation_0-error:0.039551\n",
            "[90]\tvalidation_0-auc:0.833834\tvalidation_0-error:0.039551\n",
            "[91]\tvalidation_0-auc:0.834009\tvalidation_0-error:0.039551\n",
            "[92]\tvalidation_0-auc:0.833944\tvalidation_0-error:0.039551\n",
            "[93]\tvalidation_0-auc:0.834016\tvalidation_0-error:0.039551\n",
            "[94]\tvalidation_0-auc:0.833996\tvalidation_0-error:0.039551\n",
            "[95]\tvalidation_0-auc:0.834073\tvalidation_0-error:0.039551\n",
            "[96]\tvalidation_0-auc:0.834095\tvalidation_0-error:0.039551\n",
            "[97]\tvalidation_0-auc:0.83387\tvalidation_0-error:0.039551\n",
            "[98]\tvalidation_0-auc:0.83386\tvalidation_0-error:0.039551\n",
            "[99]\tvalidation_0-auc:0.834023\tvalidation_0-error:0.039551\n",
            "[100]\tvalidation_0-auc:0.833919\tvalidation_0-error:0.039551\n",
            "[101]\tvalidation_0-auc:0.834134\tvalidation_0-error:0.039551\n",
            "[102]\tvalidation_0-auc:0.834484\tvalidation_0-error:0.039551\n",
            "[103]\tvalidation_0-auc:0.834167\tvalidation_0-error:0.039551\n",
            "[104]\tvalidation_0-auc:0.834053\tvalidation_0-error:0.039551\n",
            "[105]\tvalidation_0-auc:0.833898\tvalidation_0-error:0.039551\n",
            "[106]\tvalidation_0-auc:0.83399\tvalidation_0-error:0.039551\n",
            "[107]\tvalidation_0-auc:0.833883\tvalidation_0-error:0.039551\n",
            "[108]\tvalidation_0-auc:0.833918\tvalidation_0-error:0.039551\n",
            "[109]\tvalidation_0-auc:0.833973\tvalidation_0-error:0.039551\n",
            "[110]\tvalidation_0-auc:0.834088\tvalidation_0-error:0.039551\n",
            "[111]\tvalidation_0-auc:0.834242\tvalidation_0-error:0.039551\n",
            "[112]\tvalidation_0-auc:0.834271\tvalidation_0-error:0.039551\n",
            "[113]\tvalidation_0-auc:0.834225\tvalidation_0-error:0.039551\n",
            "[114]\tvalidation_0-auc:0.834143\tvalidation_0-error:0.039551\n",
            "[115]\tvalidation_0-auc:0.834394\tvalidation_0-error:0.039551\n",
            "[116]\tvalidation_0-auc:0.834359\tvalidation_0-error:0.039551\n",
            "[117]\tvalidation_0-auc:0.834625\tvalidation_0-error:0.039551\n",
            "[118]\tvalidation_0-auc:0.834471\tvalidation_0-error:0.039551\n",
            "[119]\tvalidation_0-auc:0.834675\tvalidation_0-error:0.039551\n",
            "[120]\tvalidation_0-auc:0.834515\tvalidation_0-error:0.039551\n",
            "[121]\tvalidation_0-auc:0.834614\tvalidation_0-error:0.039551\n",
            "[122]\tvalidation_0-auc:0.834638\tvalidation_0-error:0.039551\n",
            "[123]\tvalidation_0-auc:0.834307\tvalidation_0-error:0.039551\n",
            "[124]\tvalidation_0-auc:0.834435\tvalidation_0-error:0.039551\n",
            "[125]\tvalidation_0-auc:0.834628\tvalidation_0-error:0.039551\n",
            "[126]\tvalidation_0-auc:0.834449\tvalidation_0-error:0.039551\n",
            "[127]\tvalidation_0-auc:0.834592\tvalidation_0-error:0.039551\n",
            "[128]\tvalidation_0-auc:0.834845\tvalidation_0-error:0.039551\n",
            "[129]\tvalidation_0-auc:0.835123\tvalidation_0-error:0.039551\n",
            "[130]\tvalidation_0-auc:0.83509\tvalidation_0-error:0.039551\n",
            "[131]\tvalidation_0-auc:0.835082\tvalidation_0-error:0.039551\n",
            "[132]\tvalidation_0-auc:0.835205\tvalidation_0-error:0.039551\n",
            "[133]\tvalidation_0-auc:0.835421\tvalidation_0-error:0.039551\n",
            "[134]\tvalidation_0-auc:0.835506\tvalidation_0-error:0.039551\n",
            "[135]\tvalidation_0-auc:0.835535\tvalidation_0-error:0.039551\n",
            "[136]\tvalidation_0-auc:0.835866\tvalidation_0-error:0.039551\n",
            "[137]\tvalidation_0-auc:0.835883\tvalidation_0-error:0.039551\n",
            "[138]\tvalidation_0-auc:0.83606\tvalidation_0-error:0.039551\n",
            "[139]\tvalidation_0-auc:0.836175\tvalidation_0-error:0.039551\n",
            "[140]\tvalidation_0-auc:0.836113\tvalidation_0-error:0.039551\n",
            "[141]\tvalidation_0-auc:0.836044\tvalidation_0-error:0.039551\n",
            "[142]\tvalidation_0-auc:0.836263\tvalidation_0-error:0.039551\n",
            "[143]\tvalidation_0-auc:0.836404\tvalidation_0-error:0.039551\n",
            "[144]\tvalidation_0-auc:0.836548\tvalidation_0-error:0.039551\n",
            "[145]\tvalidation_0-auc:0.836918\tvalidation_0-error:0.039551\n",
            "[146]\tvalidation_0-auc:0.836853\tvalidation_0-error:0.039551\n",
            "[147]\tvalidation_0-auc:0.837005\tvalidation_0-error:0.039551\n",
            "[148]\tvalidation_0-auc:0.837285\tvalidation_0-error:0.039551\n",
            "[149]\tvalidation_0-auc:0.83738\tvalidation_0-error:0.039551\n",
            "[150]\tvalidation_0-auc:0.837429\tvalidation_0-error:0.039551\n",
            "[151]\tvalidation_0-auc:0.837792\tvalidation_0-error:0.039551\n",
            "[152]\tvalidation_0-auc:0.837751\tvalidation_0-error:0.039551\n",
            "[153]\tvalidation_0-auc:0.838067\tvalidation_0-error:0.039551\n",
            "[154]\tvalidation_0-auc:0.837956\tvalidation_0-error:0.039551\n",
            "[155]\tvalidation_0-auc:0.83801\tvalidation_0-error:0.039551\n",
            "[156]\tvalidation_0-auc:0.838008\tvalidation_0-error:0.039551\n",
            "[157]\tvalidation_0-auc:0.838043\tvalidation_0-error:0.039551\n",
            "[158]\tvalidation_0-auc:0.838075\tvalidation_0-error:0.039551\n",
            "[159]\tvalidation_0-auc:0.838141\tvalidation_0-error:0.039551\n",
            "[160]\tvalidation_0-auc:0.838101\tvalidation_0-error:0.039551\n",
            "[161]\tvalidation_0-auc:0.838011\tvalidation_0-error:0.039551\n",
            "[162]\tvalidation_0-auc:0.837935\tvalidation_0-error:0.039551\n",
            "[163]\tvalidation_0-auc:0.837996\tvalidation_0-error:0.039551\n",
            "[164]\tvalidation_0-auc:0.838069\tvalidation_0-error:0.039551\n",
            "[165]\tvalidation_0-auc:0.838192\tvalidation_0-error:0.039551\n",
            "[166]\tvalidation_0-auc:0.838171\tvalidation_0-error:0.039551\n",
            "[167]\tvalidation_0-auc:0.838293\tvalidation_0-error:0.039551\n",
            "[168]\tvalidation_0-auc:0.838554\tvalidation_0-error:0.039551\n",
            "[169]\tvalidation_0-auc:0.838463\tvalidation_0-error:0.039551\n",
            "[170]\tvalidation_0-auc:0.838667\tvalidation_0-error:0.039551\n",
            "[171]\tvalidation_0-auc:0.838597\tvalidation_0-error:0.039551\n",
            "[172]\tvalidation_0-auc:0.838739\tvalidation_0-error:0.039551\n",
            "[173]\tvalidation_0-auc:0.838776\tvalidation_0-error:0.039551\n",
            "[174]\tvalidation_0-auc:0.838911\tvalidation_0-error:0.039551\n",
            "[175]\tvalidation_0-auc:0.839068\tvalidation_0-error:0.039551\n",
            "[176]\tvalidation_0-auc:0.839122\tvalidation_0-error:0.039551\n",
            "[177]\tvalidation_0-auc:0.839178\tvalidation_0-error:0.039551\n",
            "[178]\tvalidation_0-auc:0.839419\tvalidation_0-error:0.039551\n",
            "[179]\tvalidation_0-auc:0.839421\tvalidation_0-error:0.039551\n",
            "[180]\tvalidation_0-auc:0.839584\tvalidation_0-error:0.039551\n",
            "[181]\tvalidation_0-auc:0.839683\tvalidation_0-error:0.039551\n",
            "[182]\tvalidation_0-auc:0.839889\tvalidation_0-error:0.039551\n",
            "[183]\tvalidation_0-auc:0.839935\tvalidation_0-error:0.039551\n",
            "[184]\tvalidation_0-auc:0.839973\tvalidation_0-error:0.039551\n",
            "[185]\tvalidation_0-auc:0.840155\tvalidation_0-error:0.039551\n",
            "[186]\tvalidation_0-auc:0.840122\tvalidation_0-error:0.039551\n",
            "[187]\tvalidation_0-auc:0.840092\tvalidation_0-error:0.039551\n",
            "[188]\tvalidation_0-auc:0.840247\tvalidation_0-error:0.039551\n",
            "[189]\tvalidation_0-auc:0.840327\tvalidation_0-error:0.039551\n",
            "[190]\tvalidation_0-auc:0.840439\tvalidation_0-error:0.039551\n",
            "[191]\tvalidation_0-auc:0.84041\tvalidation_0-error:0.039551\n",
            "[192]\tvalidation_0-auc:0.840359\tvalidation_0-error:0.039551\n",
            "[193]\tvalidation_0-auc:0.840418\tvalidation_0-error:0.039551\n",
            "[194]\tvalidation_0-auc:0.840442\tvalidation_0-error:0.039551\n",
            "[195]\tvalidation_0-auc:0.840609\tvalidation_0-error:0.039551\n",
            "[196]\tvalidation_0-auc:0.840785\tvalidation_0-error:0.039551\n",
            "[197]\tvalidation_0-auc:0.840733\tvalidation_0-error:0.039551\n",
            "[198]\tvalidation_0-auc:0.840799\tvalidation_0-error:0.039551\n",
            "[199]\tvalidation_0-auc:0.840816\tvalidation_0-error:0.039551\n",
            "[200]\tvalidation_0-auc:0.840932\tvalidation_0-error:0.039551\n",
            "[201]\tvalidation_0-auc:0.841063\tvalidation_0-error:0.039551\n",
            "[202]\tvalidation_0-auc:0.841073\tvalidation_0-error:0.039551\n",
            "[203]\tvalidation_0-auc:0.841127\tvalidation_0-error:0.039551\n",
            "[204]\tvalidation_0-auc:0.84119\tvalidation_0-error:0.039551\n",
            "[205]\tvalidation_0-auc:0.841337\tvalidation_0-error:0.039551\n",
            "[206]\tvalidation_0-auc:0.841394\tvalidation_0-error:0.039551\n",
            "[207]\tvalidation_0-auc:0.841391\tvalidation_0-error:0.039551\n",
            "[208]\tvalidation_0-auc:0.841417\tvalidation_0-error:0.039551\n",
            "[209]\tvalidation_0-auc:0.841449\tvalidation_0-error:0.039551\n",
            "[210]\tvalidation_0-auc:0.84142\tvalidation_0-error:0.039551\n",
            "[211]\tvalidation_0-auc:0.841408\tvalidation_0-error:0.039551\n",
            "[212]\tvalidation_0-auc:0.841431\tvalidation_0-error:0.039551\n",
            "[213]\tvalidation_0-auc:0.841421\tvalidation_0-error:0.039551\n",
            "[214]\tvalidation_0-auc:0.841494\tvalidation_0-error:0.039551\n",
            "[215]\tvalidation_0-auc:0.841635\tvalidation_0-error:0.039551\n",
            "[216]\tvalidation_0-auc:0.841675\tvalidation_0-error:0.039551\n",
            "[217]\tvalidation_0-auc:0.841789\tvalidation_0-error:0.039551\n",
            "[218]\tvalidation_0-auc:0.841808\tvalidation_0-error:0.039551\n",
            "[219]\tvalidation_0-auc:0.841824\tvalidation_0-error:0.039551\n",
            "[220]\tvalidation_0-auc:0.841655\tvalidation_0-error:0.039551\n",
            "[221]\tvalidation_0-auc:0.841678\tvalidation_0-error:0.039551\n",
            "[222]\tvalidation_0-auc:0.841723\tvalidation_0-error:0.039551\n",
            "[223]\tvalidation_0-auc:0.841814\tvalidation_0-error:0.039551\n",
            "[224]\tvalidation_0-auc:0.842122\tvalidation_0-error:0.039551\n",
            "[225]\tvalidation_0-auc:0.842117\tvalidation_0-error:0.039551\n",
            "[226]\tvalidation_0-auc:0.842309\tvalidation_0-error:0.039551\n",
            "[227]\tvalidation_0-auc:0.842324\tvalidation_0-error:0.039551\n",
            "[228]\tvalidation_0-auc:0.842401\tvalidation_0-error:0.039551\n",
            "[229]\tvalidation_0-auc:0.84251\tvalidation_0-error:0.039551\n",
            "[230]\tvalidation_0-auc:0.842526\tvalidation_0-error:0.039551\n",
            "[231]\tvalidation_0-auc:0.842533\tvalidation_0-error:0.039551\n",
            "[232]\tvalidation_0-auc:0.842569\tvalidation_0-error:0.039551\n",
            "[233]\tvalidation_0-auc:0.842567\tvalidation_0-error:0.039551\n",
            "[234]\tvalidation_0-auc:0.842655\tvalidation_0-error:0.039551\n",
            "[235]\tvalidation_0-auc:0.842655\tvalidation_0-error:0.039551\n",
            "[236]\tvalidation_0-auc:0.842773\tvalidation_0-error:0.039551\n",
            "[237]\tvalidation_0-auc:0.842802\tvalidation_0-error:0.039551\n",
            "[238]\tvalidation_0-auc:0.842837\tvalidation_0-error:0.039551\n",
            "[239]\tvalidation_0-auc:0.842978\tvalidation_0-error:0.039551\n",
            "[240]\tvalidation_0-auc:0.84304\tvalidation_0-error:0.039551\n",
            "[241]\tvalidation_0-auc:0.843046\tvalidation_0-error:0.039551\n",
            "[242]\tvalidation_0-auc:0.842982\tvalidation_0-error:0.039551\n",
            "[243]\tvalidation_0-auc:0.843073\tvalidation_0-error:0.039551\n",
            "[244]\tvalidation_0-auc:0.843037\tvalidation_0-error:0.039551\n",
            "[245]\tvalidation_0-auc:0.843088\tvalidation_0-error:0.039551\n",
            "[246]\tvalidation_0-auc:0.84311\tvalidation_0-error:0.039551\n",
            "[247]\tvalidation_0-auc:0.843219\tvalidation_0-error:0.039551\n",
            "[248]\tvalidation_0-auc:0.84332\tvalidation_0-error:0.039551\n",
            "[249]\tvalidation_0-auc:0.843411\tvalidation_0-error:0.039551\n",
            "[250]\tvalidation_0-auc:0.843352\tvalidation_0-error:0.039551\n",
            "[251]\tvalidation_0-auc:0.843325\tvalidation_0-error:0.039551\n",
            "[252]\tvalidation_0-auc:0.843333\tvalidation_0-error:0.039551\n",
            "[253]\tvalidation_0-auc:0.843341\tvalidation_0-error:0.039551\n",
            "[254]\tvalidation_0-auc:0.843415\tvalidation_0-error:0.039551\n",
            "[255]\tvalidation_0-auc:0.843514\tvalidation_0-error:0.039551\n",
            "[256]\tvalidation_0-auc:0.843468\tvalidation_0-error:0.039551\n",
            "[257]\tvalidation_0-auc:0.84347\tvalidation_0-error:0.039551\n",
            "[258]\tvalidation_0-auc:0.843448\tvalidation_0-error:0.039551\n",
            "[259]\tvalidation_0-auc:0.843469\tvalidation_0-error:0.039551\n",
            "[260]\tvalidation_0-auc:0.843574\tvalidation_0-error:0.039551\n",
            "[261]\tvalidation_0-auc:0.843805\tvalidation_0-error:0.039551\n",
            "[262]\tvalidation_0-auc:0.843797\tvalidation_0-error:0.039551\n",
            "[263]\tvalidation_0-auc:0.843773\tvalidation_0-error:0.039551\n",
            "[264]\tvalidation_0-auc:0.84382\tvalidation_0-error:0.039551\n",
            "[265]\tvalidation_0-auc:0.843815\tvalidation_0-error:0.039551\n",
            "[266]\tvalidation_0-auc:0.843861\tvalidation_0-error:0.039551\n",
            "[267]\tvalidation_0-auc:0.843976\tvalidation_0-error:0.039551\n",
            "[268]\tvalidation_0-auc:0.843936\tvalidation_0-error:0.039551\n",
            "[269]\tvalidation_0-auc:0.843994\tvalidation_0-error:0.039551\n",
            "[270]\tvalidation_0-auc:0.84408\tvalidation_0-error:0.039551\n",
            "[271]\tvalidation_0-auc:0.844085\tvalidation_0-error:0.039551\n",
            "[272]\tvalidation_0-auc:0.844056\tvalidation_0-error:0.039551\n",
            "[273]\tvalidation_0-auc:0.844111\tvalidation_0-error:0.039551\n",
            "[274]\tvalidation_0-auc:0.844161\tvalidation_0-error:0.039551\n",
            "[275]\tvalidation_0-auc:0.844236\tvalidation_0-error:0.039551\n",
            "[276]\tvalidation_0-auc:0.844294\tvalidation_0-error:0.039551\n",
            "[277]\tvalidation_0-auc:0.844283\tvalidation_0-error:0.039551\n",
            "[278]\tvalidation_0-auc:0.844367\tvalidation_0-error:0.039551\n",
            "[279]\tvalidation_0-auc:0.844373\tvalidation_0-error:0.039551\n",
            "[280]\tvalidation_0-auc:0.844531\tvalidation_0-error:0.039551\n",
            "[281]\tvalidation_0-auc:0.844573\tvalidation_0-error:0.039551\n",
            "[282]\tvalidation_0-auc:0.844587\tvalidation_0-error:0.039551\n",
            "[283]\tvalidation_0-auc:0.844636\tvalidation_0-error:0.039551\n",
            "[284]\tvalidation_0-auc:0.844734\tvalidation_0-error:0.039551\n",
            "[285]\tvalidation_0-auc:0.844755\tvalidation_0-error:0.039551\n",
            "[286]\tvalidation_0-auc:0.844802\tvalidation_0-error:0.039551\n",
            "[287]\tvalidation_0-auc:0.844763\tvalidation_0-error:0.039551\n",
            "[288]\tvalidation_0-auc:0.844816\tvalidation_0-error:0.039551\n",
            "[289]\tvalidation_0-auc:0.844784\tvalidation_0-error:0.039551\n",
            "[290]\tvalidation_0-auc:0.844799\tvalidation_0-error:0.039551\n",
            "[291]\tvalidation_0-auc:0.844765\tvalidation_0-error:0.039551\n",
            "[292]\tvalidation_0-auc:0.844782\tvalidation_0-error:0.039551\n",
            "[293]\tvalidation_0-auc:0.844832\tvalidation_0-error:0.039551\n",
            "[294]\tvalidation_0-auc:0.844906\tvalidation_0-error:0.039551\n",
            "[295]\tvalidation_0-auc:0.844821\tvalidation_0-error:0.039551\n",
            "[296]\tvalidation_0-auc:0.844969\tvalidation_0-error:0.039551\n",
            "[297]\tvalidation_0-auc:0.845004\tvalidation_0-error:0.039551\n",
            "[298]\tvalidation_0-auc:0.845053\tvalidation_0-error:0.039551\n",
            "[299]\tvalidation_0-auc:0.845045\tvalidation_0-error:0.039551\n",
            "[300]\tvalidation_0-auc:0.845052\tvalidation_0-error:0.039551\n",
            "CPU times: user 4min 1s, sys: 992 ms, total: 4min 2s\n",
            "Wall time: 2min 5s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
              "              colsample_bynode=1, colsample_bytree=0.5, gamma=5,\n",
              "              learning_rate=0.02, max_delta_step=0, max_depth=5,\n",
              "              min_child_weight=1, missing=None, n_estimators=301, n_jobs=-1,\n",
              "              nthread=-1, objective='binary:logistic', random_state=0,\n",
              "              reg_alpha=0.3, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
              "              silent=None, subsample=0.9, verbosity=1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "q3LRv6T8RRaN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "408e999b-73d8-4541-a2d9-4d836a644105"
      },
      "source": [
        "y_pred = model.predict_proba(X_val)[:,1]\n",
        "label=\"XGBoost Classifer\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "i+=1\n",
        "result = plot_auc(y_val,y_pred,label)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t XGBoost Classifer on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.134 and 0.845 for XGBoost Classifer on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "qUwaU5nuRRaR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "outputId": "31155738-e818-45f1-ad0b-10dd8567e491"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133739</td>\n",
              "      <td>0.845052</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                      Model                        Dataset  log_loss       auc\n",
              "0        Logisic Regression  Normal with Response Encoding  0.523996  0.823515\n",
              "1  Decision Tree Classifier  Normal with Response Encoding  0.149110  0.769599\n",
              "2  Random Forest Classifier  Normal with Response Encoding  0.136829  0.836315\n",
              "3         XGBoost Classifer  Normal with Response Encoding  0.133739  0.845052"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "LeV2q75ULzvm"
      },
      "source": [
        "# LightGBM Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "32dzyVRQLzvn",
        "colab": {}
      },
      "source": [
        "model = lgb.LGBMClassifier(n_jobs=-1,\n",
        "                      nthread=-1,\n",
        "                      scale_pos_weight=1,\n",
        "                      learning_rate=0.01,  \n",
        "                      colsample_bytree = 0.5,\n",
        "                      subsample = 0.8,\n",
        "                      objective='binary', \n",
        "                      n_estimators=227,\n",
        "                      reg_alpha = 0.3,\n",
        "                      max_depth=7, \n",
        "                      gamma=5,\n",
        "                      random_state=42\n",
        "                     )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "yJS6sV0qLzvr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "fa325440-0775-4e4d-9887-94281f8b101d"
      },
      "source": [
        "#https://towardsdatascience.com/fine-tuning-xgboost-in-python-like-a-boss-b4543ed8b1e\n",
        "\"\"\"\n",
        "manually changed each parameters 1 by 1 and compared the validation auc\n",
        "and selected the best parameter value\n",
        "\"\"\"\n",
        "eval_metric = ['auc','error']\n",
        "%time model.fit(X_train,y_train,eval_set=[(X_val,y_val)],eval_metric=eval_metric,verbose=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1]\tvalid_0's auc: 0.778931\tvalid_0's binary_logloss: 0.16587\n",
            "[2]\tvalid_0's auc: 0.807186\tvalid_0's binary_logloss: 0.165194\n",
            "[3]\tvalid_0's auc: 0.820545\tvalid_0's binary_logloss: 0.164477\n",
            "[4]\tvalid_0's auc: 0.824886\tvalid_0's binary_logloss: 0.163797\n",
            "[5]\tvalid_0's auc: 0.822005\tvalid_0's binary_logloss: 0.163291\n",
            "[6]\tvalid_0's auc: 0.825354\tvalid_0's binary_logloss: 0.162639\n",
            "[7]\tvalid_0's auc: 0.828435\tvalid_0's binary_logloss: 0.162055\n",
            "[8]\tvalid_0's auc: 0.829837\tvalid_0's binary_logloss: 0.16149\n",
            "[9]\tvalid_0's auc: 0.832309\tvalid_0's binary_logloss: 0.160885\n",
            "[10]\tvalid_0's auc: 0.831846\tvalid_0's binary_logloss: 0.160385\n",
            "[11]\tvalid_0's auc: 0.83124\tvalid_0's binary_logloss: 0.159887\n",
            "[12]\tvalid_0's auc: 0.832192\tvalid_0's binary_logloss: 0.159423\n",
            "[13]\tvalid_0's auc: 0.832026\tvalid_0's binary_logloss: 0.159011\n",
            "[14]\tvalid_0's auc: 0.832345\tvalid_0's binary_logloss: 0.158563\n",
            "[15]\tvalid_0's auc: 0.834205\tvalid_0's binary_logloss: 0.158048\n",
            "[16]\tvalid_0's auc: 0.833866\tvalid_0's binary_logloss: 0.157677\n",
            "[17]\tvalid_0's auc: 0.833776\tvalid_0's binary_logloss: 0.157274\n",
            "[18]\tvalid_0's auc: 0.83468\tvalid_0's binary_logloss: 0.156797\n",
            "[19]\tvalid_0's auc: 0.835396\tvalid_0's binary_logloss: 0.156377\n",
            "[20]\tvalid_0's auc: 0.835977\tvalid_0's binary_logloss: 0.155954\n",
            "[21]\tvalid_0's auc: 0.836253\tvalid_0's binary_logloss: 0.15552\n",
            "[22]\tvalid_0's auc: 0.836007\tvalid_0's binary_logloss: 0.15517\n",
            "[23]\tvalid_0's auc: 0.836099\tvalid_0's binary_logloss: 0.15477\n",
            "[24]\tvalid_0's auc: 0.836764\tvalid_0's binary_logloss: 0.154385\n",
            "[25]\tvalid_0's auc: 0.836732\tvalid_0's binary_logloss: 0.154027\n",
            "[26]\tvalid_0's auc: 0.836701\tvalid_0's binary_logloss: 0.153707\n",
            "[27]\tvalid_0's auc: 0.836631\tvalid_0's binary_logloss: 0.153358\n",
            "[28]\tvalid_0's auc: 0.836454\tvalid_0's binary_logloss: 0.153067\n",
            "[29]\tvalid_0's auc: 0.837259\tvalid_0's binary_logloss: 0.15273\n",
            "[30]\tvalid_0's auc: 0.837161\tvalid_0's binary_logloss: 0.152385\n",
            "[31]\tvalid_0's auc: 0.837247\tvalid_0's binary_logloss: 0.152068\n",
            "[32]\tvalid_0's auc: 0.837371\tvalid_0's binary_logloss: 0.151807\n",
            "[33]\tvalid_0's auc: 0.837383\tvalid_0's binary_logloss: 0.151517\n",
            "[34]\tvalid_0's auc: 0.837346\tvalid_0's binary_logloss: 0.151272\n",
            "[35]\tvalid_0's auc: 0.837695\tvalid_0's binary_logloss: 0.150989\n",
            "[36]\tvalid_0's auc: 0.83757\tvalid_0's binary_logloss: 0.150753\n",
            "[37]\tvalid_0's auc: 0.837705\tvalid_0's binary_logloss: 0.150463\n",
            "[38]\tvalid_0's auc: 0.838209\tvalid_0's binary_logloss: 0.150166\n",
            "[39]\tvalid_0's auc: 0.83836\tvalid_0's binary_logloss: 0.149895\n",
            "[40]\tvalid_0's auc: 0.838351\tvalid_0's binary_logloss: 0.149638\n",
            "[41]\tvalid_0's auc: 0.838552\tvalid_0's binary_logloss: 0.149363\n",
            "[42]\tvalid_0's auc: 0.838374\tvalid_0's binary_logloss: 0.14916\n",
            "[43]\tvalid_0's auc: 0.838075\tvalid_0's binary_logloss: 0.148923\n",
            "[44]\tvalid_0's auc: 0.83803\tvalid_0's binary_logloss: 0.148725\n",
            "[45]\tvalid_0's auc: 0.83821\tvalid_0's binary_logloss: 0.148494\n",
            "[46]\tvalid_0's auc: 0.83813\tvalid_0's binary_logloss: 0.148259\n",
            "[47]\tvalid_0's auc: 0.837923\tvalid_0's binary_logloss: 0.14809\n",
            "[48]\tvalid_0's auc: 0.83778\tvalid_0's binary_logloss: 0.147897\n",
            "[49]\tvalid_0's auc: 0.837816\tvalid_0's binary_logloss: 0.147673\n",
            "[50]\tvalid_0's auc: 0.837968\tvalid_0's binary_logloss: 0.147462\n",
            "[51]\tvalid_0's auc: 0.837887\tvalid_0's binary_logloss: 0.147254\n",
            "[52]\tvalid_0's auc: 0.837911\tvalid_0's binary_logloss: 0.147057\n",
            "[53]\tvalid_0's auc: 0.838079\tvalid_0's binary_logloss: 0.146855\n",
            "[54]\tvalid_0's auc: 0.83783\tvalid_0's binary_logloss: 0.146708\n",
            "[55]\tvalid_0's auc: 0.837946\tvalid_0's binary_logloss: 0.146509\n",
            "[56]\tvalid_0's auc: 0.838058\tvalid_0's binary_logloss: 0.146323\n",
            "[57]\tvalid_0's auc: 0.83824\tvalid_0's binary_logloss: 0.146129\n",
            "[58]\tvalid_0's auc: 0.838606\tvalid_0's binary_logloss: 0.145935\n",
            "[59]\tvalid_0's auc: 0.838728\tvalid_0's binary_logloss: 0.145742\n",
            "[60]\tvalid_0's auc: 0.838584\tvalid_0's binary_logloss: 0.14561\n",
            "[61]\tvalid_0's auc: 0.838435\tvalid_0's binary_logloss: 0.145408\n",
            "[62]\tvalid_0's auc: 0.838726\tvalid_0's binary_logloss: 0.145218\n",
            "[63]\tvalid_0's auc: 0.838893\tvalid_0's binary_logloss: 0.14503\n",
            "[64]\tvalid_0's auc: 0.83904\tvalid_0's binary_logloss: 0.14487\n",
            "[65]\tvalid_0's auc: 0.839098\tvalid_0's binary_logloss: 0.144735\n",
            "[66]\tvalid_0's auc: 0.839238\tvalid_0's binary_logloss: 0.144568\n",
            "[67]\tvalid_0's auc: 0.839163\tvalid_0's binary_logloss: 0.14442\n",
            "[68]\tvalid_0's auc: 0.839239\tvalid_0's binary_logloss: 0.144267\n",
            "[69]\tvalid_0's auc: 0.839227\tvalid_0's binary_logloss: 0.144097\n",
            "[70]\tvalid_0's auc: 0.839235\tvalid_0's binary_logloss: 0.143943\n",
            "[71]\tvalid_0's auc: 0.839268\tvalid_0's binary_logloss: 0.143812\n",
            "[72]\tvalid_0's auc: 0.839369\tvalid_0's binary_logloss: 0.143658\n",
            "[73]\tvalid_0's auc: 0.839686\tvalid_0's binary_logloss: 0.143487\n",
            "[74]\tvalid_0's auc: 0.839842\tvalid_0's binary_logloss: 0.143331\n",
            "[75]\tvalid_0's auc: 0.839712\tvalid_0's binary_logloss: 0.143206\n",
            "[76]\tvalid_0's auc: 0.839831\tvalid_0's binary_logloss: 0.143071\n",
            "[77]\tvalid_0's auc: 0.839898\tvalid_0's binary_logloss: 0.142929\n",
            "[78]\tvalid_0's auc: 0.839885\tvalid_0's binary_logloss: 0.14278\n",
            "[79]\tvalid_0's auc: 0.839944\tvalid_0's binary_logloss: 0.142651\n",
            "[80]\tvalid_0's auc: 0.840137\tvalid_0's binary_logloss: 0.142518\n",
            "[81]\tvalid_0's auc: 0.840207\tvalid_0's binary_logloss: 0.142389\n",
            "[82]\tvalid_0's auc: 0.840187\tvalid_0's binary_logloss: 0.142274\n",
            "[83]\tvalid_0's auc: 0.840377\tvalid_0's binary_logloss: 0.142174\n",
            "[84]\tvalid_0's auc: 0.840494\tvalid_0's binary_logloss: 0.142054\n",
            "[85]\tvalid_0's auc: 0.840182\tvalid_0's binary_logloss: 0.141973\n",
            "[86]\tvalid_0's auc: 0.840158\tvalid_0's binary_logloss: 0.141886\n",
            "[87]\tvalid_0's auc: 0.84059\tvalid_0's binary_logloss: 0.141763\n",
            "[88]\tvalid_0's auc: 0.840744\tvalid_0's binary_logloss: 0.141635\n",
            "[89]\tvalid_0's auc: 0.840755\tvalid_0's binary_logloss: 0.141509\n",
            "[90]\tvalid_0's auc: 0.840735\tvalid_0's binary_logloss: 0.141434\n",
            "[91]\tvalid_0's auc: 0.840738\tvalid_0's binary_logloss: 0.141327\n",
            "[92]\tvalid_0's auc: 0.840778\tvalid_0's binary_logloss: 0.141204\n",
            "[93]\tvalid_0's auc: 0.840733\tvalid_0's binary_logloss: 0.141093\n",
            "[94]\tvalid_0's auc: 0.840829\tvalid_0's binary_logloss: 0.140979\n",
            "[95]\tvalid_0's auc: 0.840896\tvalid_0's binary_logloss: 0.140894\n",
            "[96]\tvalid_0's auc: 0.840881\tvalid_0's binary_logloss: 0.140788\n",
            "[97]\tvalid_0's auc: 0.840991\tvalid_0's binary_logloss: 0.140674\n",
            "[98]\tvalid_0's auc: 0.840945\tvalid_0's binary_logloss: 0.140595\n",
            "[99]\tvalid_0's auc: 0.841027\tvalid_0's binary_logloss: 0.140501\n",
            "[100]\tvalid_0's auc: 0.840924\tvalid_0's binary_logloss: 0.140414\n",
            "[101]\tvalid_0's auc: 0.841022\tvalid_0's binary_logloss: 0.140324\n",
            "[102]\tvalid_0's auc: 0.841096\tvalid_0's binary_logloss: 0.140218\n",
            "[103]\tvalid_0's auc: 0.841197\tvalid_0's binary_logloss: 0.140112\n",
            "[104]\tvalid_0's auc: 0.841388\tvalid_0's binary_logloss: 0.140003\n",
            "[105]\tvalid_0's auc: 0.841587\tvalid_0's binary_logloss: 0.139899\n",
            "[106]\tvalid_0's auc: 0.841744\tvalid_0's binary_logloss: 0.139824\n",
            "[107]\tvalid_0's auc: 0.841758\tvalid_0's binary_logloss: 0.139726\n",
            "[108]\tvalid_0's auc: 0.841739\tvalid_0's binary_logloss: 0.139633\n",
            "[109]\tvalid_0's auc: 0.841826\tvalid_0's binary_logloss: 0.139537\n",
            "[110]\tvalid_0's auc: 0.842029\tvalid_0's binary_logloss: 0.13944\n",
            "[111]\tvalid_0's auc: 0.841972\tvalid_0's binary_logloss: 0.139373\n",
            "[112]\tvalid_0's auc: 0.842029\tvalid_0's binary_logloss: 0.139294\n",
            "[113]\tvalid_0's auc: 0.842178\tvalid_0's binary_logloss: 0.139234\n",
            "[114]\tvalid_0's auc: 0.842278\tvalid_0's binary_logloss: 0.139158\n",
            "[115]\tvalid_0's auc: 0.842227\tvalid_0's binary_logloss: 0.139078\n",
            "[116]\tvalid_0's auc: 0.842465\tvalid_0's binary_logloss: 0.138983\n",
            "[117]\tvalid_0's auc: 0.842469\tvalid_0's binary_logloss: 0.138923\n",
            "[118]\tvalid_0's auc: 0.842398\tvalid_0's binary_logloss: 0.138874\n",
            "[119]\tvalid_0's auc: 0.842468\tvalid_0's binary_logloss: 0.138795\n",
            "[120]\tvalid_0's auc: 0.842572\tvalid_0's binary_logloss: 0.138723\n",
            "[121]\tvalid_0's auc: 0.842601\tvalid_0's binary_logloss: 0.138637\n",
            "[122]\tvalid_0's auc: 0.842645\tvalid_0's binary_logloss: 0.138575\n",
            "[123]\tvalid_0's auc: 0.842613\tvalid_0's binary_logloss: 0.13851\n",
            "[124]\tvalid_0's auc: 0.842621\tvalid_0's binary_logloss: 0.138438\n",
            "[125]\tvalid_0's auc: 0.842647\tvalid_0's binary_logloss: 0.138369\n",
            "[126]\tvalid_0's auc: 0.842718\tvalid_0's binary_logloss: 0.138321\n",
            "[127]\tvalid_0's auc: 0.842831\tvalid_0's binary_logloss: 0.138244\n",
            "[128]\tvalid_0's auc: 0.842768\tvalid_0's binary_logloss: 0.138178\n",
            "[129]\tvalid_0's auc: 0.8427\tvalid_0's binary_logloss: 0.138113\n",
            "[130]\tvalid_0's auc: 0.842671\tvalid_0's binary_logloss: 0.138074\n",
            "[131]\tvalid_0's auc: 0.842699\tvalid_0's binary_logloss: 0.138009\n",
            "[132]\tvalid_0's auc: 0.842808\tvalid_0's binary_logloss: 0.137945\n",
            "[133]\tvalid_0's auc: 0.842899\tvalid_0's binary_logloss: 0.137891\n",
            "[134]\tvalid_0's auc: 0.843057\tvalid_0's binary_logloss: 0.137833\n",
            "[135]\tvalid_0's auc: 0.843133\tvalid_0's binary_logloss: 0.137779\n",
            "[136]\tvalid_0's auc: 0.843189\tvalid_0's binary_logloss: 0.137712\n",
            "[137]\tvalid_0's auc: 0.843236\tvalid_0's binary_logloss: 0.137649\n",
            "[138]\tvalid_0's auc: 0.843253\tvalid_0's binary_logloss: 0.137592\n",
            "[139]\tvalid_0's auc: 0.843279\tvalid_0's binary_logloss: 0.137531\n",
            "[140]\tvalid_0's auc: 0.843313\tvalid_0's binary_logloss: 0.137475\n",
            "[141]\tvalid_0's auc: 0.843321\tvalid_0's binary_logloss: 0.137419\n",
            "[142]\tvalid_0's auc: 0.843403\tvalid_0's binary_logloss: 0.137371\n",
            "[143]\tvalid_0's auc: 0.843369\tvalid_0's binary_logloss: 0.137326\n",
            "[144]\tvalid_0's auc: 0.843335\tvalid_0's binary_logloss: 0.137287\n",
            "[145]\tvalid_0's auc: 0.843424\tvalid_0's binary_logloss: 0.137221\n",
            "[146]\tvalid_0's auc: 0.843428\tvalid_0's binary_logloss: 0.13718\n",
            "[147]\tvalid_0's auc: 0.84346\tvalid_0's binary_logloss: 0.137133\n",
            "[148]\tvalid_0's auc: 0.843607\tvalid_0's binary_logloss: 0.137062\n",
            "[149]\tvalid_0's auc: 0.843663\tvalid_0's binary_logloss: 0.137011\n",
            "[150]\tvalid_0's auc: 0.843638\tvalid_0's binary_logloss: 0.136974\n",
            "[151]\tvalid_0's auc: 0.843648\tvalid_0's binary_logloss: 0.136927\n",
            "[152]\tvalid_0's auc: 0.843739\tvalid_0's binary_logloss: 0.136889\n",
            "[153]\tvalid_0's auc: 0.843689\tvalid_0's binary_logloss: 0.136862\n",
            "[154]\tvalid_0's auc: 0.843739\tvalid_0's binary_logloss: 0.136819\n",
            "[155]\tvalid_0's auc: 0.843708\tvalid_0's binary_logloss: 0.136771\n",
            "[156]\tvalid_0's auc: 0.843791\tvalid_0's binary_logloss: 0.136722\n",
            "[157]\tvalid_0's auc: 0.843832\tvalid_0's binary_logloss: 0.136668\n",
            "[158]\tvalid_0's auc: 0.843932\tvalid_0's binary_logloss: 0.136608\n",
            "[159]\tvalid_0's auc: 0.843963\tvalid_0's binary_logloss: 0.136561\n",
            "[160]\tvalid_0's auc: 0.844024\tvalid_0's binary_logloss: 0.136517\n",
            "[161]\tvalid_0's auc: 0.844064\tvalid_0's binary_logloss: 0.13647\n",
            "[162]\tvalid_0's auc: 0.844359\tvalid_0's binary_logloss: 0.136409\n",
            "[163]\tvalid_0's auc: 0.844374\tvalid_0's binary_logloss: 0.136371\n",
            "[164]\tvalid_0's auc: 0.844498\tvalid_0's binary_logloss: 0.136317\n",
            "[165]\tvalid_0's auc: 0.844505\tvalid_0's binary_logloss: 0.136286\n",
            "[166]\tvalid_0's auc: 0.844469\tvalid_0's binary_logloss: 0.13624\n",
            "[167]\tvalid_0's auc: 0.844468\tvalid_0's binary_logloss: 0.136209\n",
            "[168]\tvalid_0's auc: 0.844577\tvalid_0's binary_logloss: 0.136163\n",
            "[169]\tvalid_0's auc: 0.844561\tvalid_0's binary_logloss: 0.136129\n",
            "[170]\tvalid_0's auc: 0.844504\tvalid_0's binary_logloss: 0.136092\n",
            "[171]\tvalid_0's auc: 0.844676\tvalid_0's binary_logloss: 0.13606\n",
            "[172]\tvalid_0's auc: 0.84479\tvalid_0's binary_logloss: 0.136008\n",
            "[173]\tvalid_0's auc: 0.844833\tvalid_0's binary_logloss: 0.135965\n",
            "[174]\tvalid_0's auc: 0.844798\tvalid_0's binary_logloss: 0.135936\n",
            "[175]\tvalid_0's auc: 0.844878\tvalid_0's binary_logloss: 0.1359\n",
            "[176]\tvalid_0's auc: 0.844806\tvalid_0's binary_logloss: 0.135876\n",
            "[177]\tvalid_0's auc: 0.844751\tvalid_0's binary_logloss: 0.135844\n",
            "[178]\tvalid_0's auc: 0.844676\tvalid_0's binary_logloss: 0.135811\n",
            "[179]\tvalid_0's auc: 0.844623\tvalid_0's binary_logloss: 0.135789\n",
            "[180]\tvalid_0's auc: 0.844645\tvalid_0's binary_logloss: 0.135764\n",
            "[181]\tvalid_0's auc: 0.844676\tvalid_0's binary_logloss: 0.135726\n",
            "[182]\tvalid_0's auc: 0.844724\tvalid_0's binary_logloss: 0.135688\n",
            "[183]\tvalid_0's auc: 0.844706\tvalid_0's binary_logloss: 0.135658\n",
            "[184]\tvalid_0's auc: 0.844612\tvalid_0's binary_logloss: 0.135634\n",
            "[185]\tvalid_0's auc: 0.844613\tvalid_0's binary_logloss: 0.135615\n",
            "[186]\tvalid_0's auc: 0.844605\tvalid_0's binary_logloss: 0.135583\n",
            "[187]\tvalid_0's auc: 0.844599\tvalid_0's binary_logloss: 0.135556\n",
            "[188]\tvalid_0's auc: 0.844576\tvalid_0's binary_logloss: 0.135524\n",
            "[189]\tvalid_0's auc: 0.844572\tvalid_0's binary_logloss: 0.135497\n",
            "[190]\tvalid_0's auc: 0.844609\tvalid_0's binary_logloss: 0.135464\n",
            "[191]\tvalid_0's auc: 0.844707\tvalid_0's binary_logloss: 0.135432\n",
            "[192]\tvalid_0's auc: 0.844728\tvalid_0's binary_logloss: 0.135412\n",
            "[193]\tvalid_0's auc: 0.84477\tvalid_0's binary_logloss: 0.135373\n",
            "[194]\tvalid_0's auc: 0.844757\tvalid_0's binary_logloss: 0.135337\n",
            "[195]\tvalid_0's auc: 0.844738\tvalid_0's binary_logloss: 0.135312\n",
            "[196]\tvalid_0's auc: 0.844883\tvalid_0's binary_logloss: 0.135271\n",
            "[197]\tvalid_0's auc: 0.844888\tvalid_0's binary_logloss: 0.135243\n",
            "[198]\tvalid_0's auc: 0.844924\tvalid_0's binary_logloss: 0.135208\n",
            "[199]\tvalid_0's auc: 0.845035\tvalid_0's binary_logloss: 0.135175\n",
            "[200]\tvalid_0's auc: 0.845095\tvalid_0's binary_logloss: 0.135149\n",
            "[201]\tvalid_0's auc: 0.845058\tvalid_0's binary_logloss: 0.135119\n",
            "[202]\tvalid_0's auc: 0.845091\tvalid_0's binary_logloss: 0.135101\n",
            "[203]\tvalid_0's auc: 0.845216\tvalid_0's binary_logloss: 0.135063\n",
            "[204]\tvalid_0's auc: 0.845214\tvalid_0's binary_logloss: 0.135034\n",
            "[205]\tvalid_0's auc: 0.845293\tvalid_0's binary_logloss: 0.135006\n",
            "[206]\tvalid_0's auc: 0.845314\tvalid_0's binary_logloss: 0.134977\n",
            "[207]\tvalid_0's auc: 0.845275\tvalid_0's binary_logloss: 0.134955\n",
            "[208]\tvalid_0's auc: 0.845287\tvalid_0's binary_logloss: 0.134928\n",
            "[209]\tvalid_0's auc: 0.845342\tvalid_0's binary_logloss: 0.134898\n",
            "[210]\tvalid_0's auc: 0.845336\tvalid_0's binary_logloss: 0.134877\n",
            "[211]\tvalid_0's auc: 0.84531\tvalid_0's binary_logloss: 0.134854\n",
            "[212]\tvalid_0's auc: 0.845335\tvalid_0's binary_logloss: 0.134837\n",
            "[213]\tvalid_0's auc: 0.845338\tvalid_0's binary_logloss: 0.134819\n",
            "[214]\tvalid_0's auc: 0.845332\tvalid_0's binary_logloss: 0.134798\n",
            "[215]\tvalid_0's auc: 0.845366\tvalid_0's binary_logloss: 0.134774\n",
            "[216]\tvalid_0's auc: 0.84535\tvalid_0's binary_logloss: 0.134751\n",
            "[217]\tvalid_0's auc: 0.845352\tvalid_0's binary_logloss: 0.134724\n",
            "[218]\tvalid_0's auc: 0.845344\tvalid_0's binary_logloss: 0.134702\n",
            "[219]\tvalid_0's auc: 0.845343\tvalid_0's binary_logloss: 0.134685\n",
            "[220]\tvalid_0's auc: 0.845297\tvalid_0's binary_logloss: 0.134669\n",
            "[221]\tvalid_0's auc: 0.845297\tvalid_0's binary_logloss: 0.134647\n",
            "[222]\tvalid_0's auc: 0.845401\tvalid_0's binary_logloss: 0.13462\n",
            "[223]\tvalid_0's auc: 0.845345\tvalid_0's binary_logloss: 0.134604\n",
            "[224]\tvalid_0's auc: 0.845363\tvalid_0's binary_logloss: 0.134581\n",
            "[225]\tvalid_0's auc: 0.845387\tvalid_0's binary_logloss: 0.134564\n",
            "[226]\tvalid_0's auc: 0.845392\tvalid_0's binary_logloss: 0.134545\n",
            "[227]\tvalid_0's auc: 0.845469\tvalid_0's binary_logloss: 0.13452\n",
            "CPU times: user 37 s, sys: 318 ms, total: 37.3 s\n",
            "Wall time: 20 s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=0.5,\n",
              "               gamma=5, importance_type='split', learning_rate=0.01,\n",
              "               max_depth=7, min_child_samples=20, min_child_weight=0.001,\n",
              "               min_split_gain=0.0, n_estimators=227, n_jobs=-1, nthread=-1,\n",
              "               num_leaves=31, objective='binary', random_state=None,\n",
              "               reg_alpha=0.3, reg_lambda=0.0, scale_pos_weight=1, silent=True,\n",
              "               subsample=0.8, subsample_for_bin=200000, subsample_freq=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "k_CEadVqLzvy",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "b40ab0ad-aec5-468d-941a-cf0072237e40"
      },
      "source": [
        "y_pred = model.predict_proba(X_val)[:,1]\n",
        "label=\"LightGBM Classifer\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "i+=1\n",
        "result = plot_auc(y_val,y_pred,label)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t LightGBM Classifer on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.135 and 0.845 for LightGBM Classifer on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "BF22prvpLzv1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "1c853c5a-250a-4d52-800d-135a414e5c67"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133739</td>\n",
              "      <td>0.845052</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>LightGBM Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.134520</td>\n",
              "      <td>0.845469</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                      Model                        Dataset  log_loss       auc\n",
              "0        Logisic Regression  Normal with Response Encoding  0.523996  0.823515\n",
              "1  Decision Tree Classifier  Normal with Response Encoding  0.149110  0.769599\n",
              "2  Random Forest Classifier  Normal with Response Encoding  0.136829  0.836315\n",
              "3         XGBoost Classifer  Normal with Response Encoding  0.133739  0.845052\n",
              "4        LightGBM Classifer  Normal with Response Encoding  0.134520  0.845469"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "LECmbx42-V2m"
      },
      "source": [
        "# Random Forest (top 250)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "CUgv74lV-V2n",
        "colab": {}
      },
      "source": [
        "model = ensemble.RandomForestClassifier(class_weight='balanced')\n",
        "params = { 'n_estimators':[1000,2000],\n",
        "          'max_depth':[500,1000], \n",
        "         'min_samples_split':[100,500],\n",
        "          'min_samples_leaf':[3,5],\n",
        "          'max_leaf_nodes':[100,250] \n",
        "          }\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "eCt8L0AV-V2q",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "2e7adac8-5600-469f-c3f2-f5285281f4ed"
      },
      "source": [
        "%%time\n",
        "find_best_params(model,params,cv=3,X_train=X_train_200)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 10 candidates, totalling 30 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  30 out of  30 | elapsed: 66.0min finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "The best auc score was 0.822\n",
            "The best params were: {'n_estimators': 2000, 'min_samples_split': 500, 'min_samples_leaf': 3, 'max_leaf_nodes': 250, 'max_depth': 500}\n",
            "CPU times: user 6min 10s, sys: 915 ms, total: 6min 11s\n",
            "Wall time: 1h 12min 13s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
              "                       criterion='gini', max_depth=500, max_features='auto',\n",
              "                       max_leaf_nodes=250, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=3, min_samples_split=500,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
              "                       n_jobs=None, oob_score=False, random_state=None,\n",
              "                       verbose=0, warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "DVdFANOy-V2u",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "outputId": "917d1c3b-9dae-446c-c05b-29ad3b3a8fb9"
      },
      "source": [
        "model = ensemble.RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
        "                       criterion='gini', max_depth=500, max_features='auto',\n",
        "                       max_leaf_nodes=250, max_samples=None,\n",
        "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
        "                       min_samples_leaf=3, min_samples_split=500,\n",
        "                       min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
        "                       n_jobs=-1, oob_score=False,\n",
        "                       verbose=0, warm_start=False)\n",
        "\n",
        "model.fit(X_train_200,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight='balanced',\n",
              "                       criterion='gini', max_depth=500, max_features='auto',\n",
              "                       max_leaf_nodes=250, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=3, min_samples_split=500,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
              "                       n_jobs=-1, oob_score=False, random_state=None, verbose=0,\n",
              "                       warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "okg4Xksh-V2y",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "outputId": "000ca9b0-7596-4227-a14a-73934b624c41"
      },
      "source": [
        "cc_model = calibration.CalibratedClassifierCV(model,cv='prefit')\n",
        "cc_model.fit(X_train_200,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CalibratedClassifierCV(base_estimator=RandomForestClassifier(bootstrap=True,\n",
              "                                                             ccp_alpha=0.0,\n",
              "                                                             class_weight='balanced',\n",
              "                                                             criterion='gini',\n",
              "                                                             max_depth=500,\n",
              "                                                             max_features='auto',\n",
              "                                                             max_leaf_nodes=250,\n",
              "                                                             max_samples=None,\n",
              "                                                             min_impurity_decrease=0.0,\n",
              "                                                             min_impurity_split=None,\n",
              "                                                             min_samples_leaf=3,\n",
              "                                                             min_samples_split=500,\n",
              "                                                             min_weight_fraction_leaf=0.0,\n",
              "                                                             n_estimators=2000,\n",
              "                                                             n_jobs=-1,\n",
              "                                                             oob_score=False,\n",
              "                                                             random_state=None,\n",
              "                                                             verbose=0,\n",
              "                                                             warm_start=False),\n",
              "                       cv='prefit', method='sigmoid')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "5eSsmef9-V21",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "9c908579-ebd1-4fae-821e-3685f082662c"
      },
      "source": [
        "i+=1\n",
        "y_pred = cc_model.predict_proba(X_val_200)[:,1]\n",
        "label=\"Random Forest Classifier (top 250)\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "result = plot_auc(y_val,y_pred,label,dataset)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t Random Forest Classifier (top 250) on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAEcCAYAAADA0KJcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3hUZfbA8e+UTDqBJIQeQMpLCSACAhZ6R2SVIgr627XirqAuNlRcO1hYBcvaFSsgCog0ARVWlAWDogi8tFBTIIWQZJLJlPv7404whAABkgwk5/M8PMzcueXcuZPkzLlvsRiGgRBCCCGEENWVNdABCCGEEEIIEUiSEAshhBBCiGpNEmIhhBBCCFGtSUIshBBCCCGqNUmIhRBCCCFEtSYJsRBCCCGEqNYkIRYBp5RqopQylFL2c9xPrlLqolO8vkcp1e9cjiEqnlLqA6XU0/7HVyqldCUd93Gl1MeVcSwhhBDnl3NKQKoLpdQeoA7gBXKBZcBdWuvcYutcBjwNdAF8wBrgQa31lmLr1ACeBK4FooE0YBHwtNY6vQLj7wV8B/xHa/33Yst/AN7RWn9QUceuTFrriKLHSqkPgANa60fPZl9Kqb8C7wL5mNczCXhEa/31uUd6/lJKfQ90AzzFFn+ntR4WiHi01v8FVCCOfSrn+vk6344jhBDVnVSIy26YP+G6GOgITC56QSnVHfgGWAjUB5oCm4C1RRVLpZQDWAW0BQYBNYDuQAZwaSXEnwfcqJRqcq47OtdK7gXkJ/81rwm8DsxWStUMcEyV4S6tdUSxfwFJhoUQQojKUl0Sm3KjtU5VSi3HTIyLPA98qLWeUWzZo0qpTsDjwE3+f/FA72KV5UPAUyc7lr/qPANoCWwH7tZa/+h/7Xvgv0AfoD3wE3DDKSrNR4D5wL+Av5VyLCvwMHAbEIpZBZ+gtc72J9FJwK3+7fcopd7zr7vev79MYJw/1qeAYOB+rfUs//6HYlbQmwHZwLta68dPdu7F4vobcG1RUqaU2gH8qrUe5X++H/PLyq9KKQNo4X9PxgKGUuoejq9wXqyU+jfQ2H+O/6e1LjhVDFprn1LqI+AN//43KKWCgWeA0f5znQ/cq7XOV0rFAh8AV2BWl/8Aevr3swd4E7gRqAcsAO4sikEpdRvwIOYdhB+A8VrrZP9rBnAnMAmoDXyCmbwaSqnmmBXtiwE3sEprfZ1/u1bAK0An4DAwRWs993TvfUn+Ow0fAy/5Y/QCD2ut3/e/Hop5jUdifon4Hejvf0+uBqYCDYBf/ee81b9dR3/sLYAlgFHymFrrhv7ne4BXMX+eTriGSqkHgHv9+3gMeBtoobXeWcr5NMW8TpcA6wBd4vXPgSsxfx42+WP+Qyl1O6V8vpRSD2H+TMQB+zHvKMz37+uMr8/JjnOqaySEEOLsSIX4DCmlGgKDgZ3+52HAZcDnpaw+F+jvf9wPWFa8mcVpjhMNLAZmAjHAv4HFSqmYYqvdgJmMxgEO4L7T7PYZYIRSqrRb0H/1/+sNXAREYCYexfUEWgMD/c+7Ar/54/sUmI3ZZKQ5ZnL8qlKqqBlDHmYSUxMYCtyplPrLaeIFWA1cqZSyKqXq+8+zO4C/+h7hj+EYrfVbmMni86VUOEdjVuibYn6R+OvpAlBK2TDfZzew1794Gmbyf7H/fBtgJmBgJqwHMJPWOphfNIrPkT4W8z1s5t/Ho/7j9MFMGkdjJst7Md/T4q7CfI/b+9cruhZPYd6lqAU0xEywUEqFAyswr08cMAZ4XSnV5nTnfRJ1gSj/+d4CvKaUquV/7UXMpO4yzIT+AcCnlGoJfAbc439PlgCLlFIO/52TBcBH/m0+B0acJoZSr6FSahDwT8yfteZAr9Ps51MgEYjFfP/+r8TrSzGT9DhgI+Zn6lSfr12YCXQU8ATwsVKqnv+1M74+p/kcCyGEKEdSIS67Bf4KXQTwLWalFMw/4lYgpZRtUjD/2IKZNCaewfGGAju01h/5n3+mlJoIDMOsagG8r7XeDqCUmgtcfaod+qvbb2C2Y76uxMtjgX9rrXf79zcZ2Oyv0BZ5XGud538dIKlYdXAO8AjwpNbaBXyjlCrETEx+1Vp/X2w/vymlPsNMsBecJubdSqkczMSzJbAcs8rbCjMx/q/W2neqfZQws1jFdRHHV/pL6qaUOgKEY7apHae1PqSUsgC3A+211pn+fT2LmdRMxkyc6wGN/ZXJ/5bY76ta6/3+7Z7BTI4exbwG72mtN/pfmwxkKaWaaK33+LedprU+AhxRSn3nj3+Z/5iNgfpa6wOY1WUwE+g9RdcJ+EUp9QUwCjNpK/U9Ukq9WOz5K1rrKf7Hbsxr7AGWKKVyzVDVeuBmoJvW+qB/3aK7GdcBi7XWK/zPXwTuxkycfUAQ8LLW2gDmKaX+eZK4jsV3kms4GvNn4g//a4/739MTKKXiMb9Y9PN/Xtf493WM1vq9Yus/jnktorTW2aXtU2td/EvxHP/1uxSzKVV5Xh8hhBDlTBLisvuL1nqlUqonZuITi9kMIQvzj3o9YFuJbeoBRU0YMvzPy6o+f1Yji+zFrMwVSS322ImZrJ/Oc8AupVSH0xxvL+bno06xZftLbJNW7HE+gNa65LIIAKVUV8yqagJmlTeY0qvqpVmNWe1r7n98BDOZ7u5/fiZKvmf1T7HuOq31Ff4q97uY1b+5mFXOMCCxWLHdAtj8j1/AbCrzjf/1t7TW04rtt/j7uLdYDPUxK5EAaK1zlVIZmNd8z0niL7rmD2BWIdcrpbKA6f6ErjHQ1Z/YF7FjVmRPZqLW+p2TvJbhT4ZLxhALhGBWSUs67rPlbzqy339eXuCgPxkuUvJzX9LJrmF94Odir5X8vJaMKavoC16x4zaCY3cFnsFMTGtj/oyDeZ6lJsRKqZswK9RN/IuK3hco3+sjhBCinElCfIa01qv9Pb9fxEyS85RSP2H+4fyuxOqjMTvSAawEnlZKhZf4I3wyyZh/LIuLx6wGnjWtdYZS6mVObLtc8njxmFXRNMxbvHD8bf8z9SlmE4zBWusCfwyxp9mmyGrMynhT4FnMhHgsZkJcsllHkXOJ9Tj+xPROYLe/7fQmzGS/bbFqaPH1czCbTUxSSiUA3yqlNmitiz4LjYqtHo/53kOJa+C/nR4DnHCMUo6Zitl+FaXUFcBKpdQazKRwtda6/6m2LwfpQAFmM5BNJV5LBtoVPfFX2BthnpcBNFBKWYolxfGUnlifTgp/flbh+Pe5tHVrlfh5jOfPz80NwHDM5hd7MJtBZGF+8YESny+lVGPM9sp9MTtjepVSvxatfw7Xp9w+x0IIIU5OEuKz8zJmx7IOWutNwEPAcqXUNuB9zPd1EmbC1sW/zUfAHcAX/g4y2zHbE96B2aRgSYljLAFeUUrdgFmVHAG0Acpj2K9/A7v58487mG08H1RKLcXs2PMsMEdr7Sm9yfEZiwQy/cnwpZgJxzdl3Ha1P+Y0rfUBpdRRzPfTDvxykm3SMNtClwutdaZS6h3gMa31NUqpt4GXlFJ3+ZtRNAAStNbLlVJXYd4t2IVZTfTyZ4UR4B9Kqa8xq5uPAHP8yz/DbBrzKbAV8xr8r1hziZNSSo3CTMQOYCZuhv+YXwPTlFI38md75IuB3KJObeXBX/V9D/i3/1hpmM0FNmJ+fh9SSvXFHI7wbsCFv0kF5heviUqp1zG/+FzKiV8uy2Iu8J4yO0DuBaacbEWt9V6l1M/AE0qph/3HHAZ85V8l0h9jBubdgGdL7KLk5ysc8z0/DMc6gyYUvXgO16dcP8dCCCFKJ53qzoLW+jDwIf5OVFrrHzA7N12LWXnaizk02xVa6x3+dVyY1aZtmJ1ojmKO0BAL/K+UY2Rgti+chPlH+QHgqlOMInEm8R/FHBkjutji9zCTzDWYI0oUABPO9VjF/B140t8e+DHM5KVM/O2kc/G3xfXHvxtYq7X2nmSzd4E2SqkjSqlTtlM+Ay8DQ5RS7TFHWdgJrPMn6Cv5c7zcFv7nuZijf7yutS6e4H2K+WVgN2bS/LT/vFZiJnFfYH6OmmF2siqLLsD//G16v8IckWS3v1o9wL+fZMzmBs9hNlk5mVeVOclJ0b+ytn2/D3NkiQ2Yo448B1i11hqzk+UrmJXkYZgjgxRqrQsxf27+6t/mOuDLMh7vOFrrpZidUL/Df238L7lOsskNmB1DMzH7BHxY7LUPMX+ODwJbiu2ryHGfL22ONz4d83qnYVbE1xZb/2yvT0V8joUQQpRgMQy5IydEZVHmsGG3+pNfUYGUUq2BzUBwiXbPQgghxHGkyYQQospQSl2D2dwoDLPSukiSYSGEEKcjTSaEEFXJHZgT3uzCbLt9Z2DDEUIIcSGQJhNCCCGEEKJakwqxEEIIIYSo1qpMG+LExMRgzJ7cKZi3SoUQQpyeDXPSoA2dOnU62YgcQghRpVWZhBgzGS45Ra4QQoiyuZI/p5Q+Y4mJiXF2u/0dzPGX5e6jEOJ8Ylgslmyv1/u+z+f7T6dOnQpLrlCVEuIUgJYtW+JwOM54482bN5OQkHD6FasQOefqQc65ejjbcy4sLGT79u3g/x16tux2+zt169ZtXbt27Syr1SqdU4QQ5w3DMCgsLAxKTk6ecPTo0UuA/yu5TlVKiL0ADoeD4OBTzTlwcme73YVMzrl6kHOuHs7xnM+1qVmCJMNCiPORxWIhODjY3bhx4+zff//9itLWkdtaQgghyoNVkmEhxPnM/zvKVuprlRyLEEIIIYQQ5xVJiIUQQlQ5V155Zbs+ffq0HThwYJu+ffu2ff/992Mr4hi///57SHnvt8jIkSNVjx492g0aNKjNoEGD2tx+++0XVdSximRlZdleeumlOqdaJyUlxf6Pf/yjSY8ePdoNGTKk9bBhw1q/9957sQATJkxo8sYbb9Quz5g2bNgQdueddzYtev7WW2/V7tOnT9vBgwe32bNnj2PUqFEty+tYt9xyS7P169eHASxcuLDm//73v7Dy2jdAenq6bdy4cc379OmTMGDAgDY333xzs0OHDh1rvqqU6jRw4MA2Rdf8t99+Cy16bdGiRVF9+vRp26tXr4Tbb7/9ory8PCvA5s2bQ8aOHdvidMf+17/+1aBt27aXpKWlHddcVinVKScn57h8sHPnzh2SkpIcAD6fj9deey2uf//+bfv379928ODBbe69997GWVlZpVZaS6O1Dh4+fHir3r17JwwfPrzV9u3bS23flZaWZh83blzzop/b+++/P97tdgMwa9asmKL3ZsCAAW1ef/31uKLt/vGPfzQpes8GDRrUplWrVp0WLVoUVdb4oJLaECulXgRGAE2AdlrrzaWsYwNmAoMAA5imtX6nMuITQghR9cyYMWNXu3btCn7//feQMWPGtBkwYEB2gwYN3IGO60w88MAD+6666qrss9nW7XYTFBR0RtscOXLE9sknn9S9995700p7PS8vzzp27NhWQ4cOTZ85c+Yem83GkSNHbPPmzat1NjGWRZcuXZxdunRJKno+Z86cuGnTpiVdeumlToDPP/98+5nus7T3Zt26deFOp9NatN+VK1fWbNu2bV7Xrl2d53gKx1gsFm699da0Xr165QA89thjDZ955pkGM2bM2Fu0zrx587ZFRkb6im+Xk5Njffrpp5t89NFH21q2bOm65557Gr/yyit1HnrooZSEhISCoKAg33fffRfZu3fvnNKO6/F4WL58eUzbtm1z58yZEzNx4sRSr29pnnnmmfq//PJL5EcffaTr1q3r8fl8LFy4sGZmZqatVq1aZep7MGXKlMbXXXfdoRtuuCHz008/jX700Ucbz50794TrNmPGjHpNmzYt+Pjjj3cWFhZaRowYoRYsWFBr1KhRWX/5y1+ybrzxxgyr1crRo0etQ4YMaXv55ZfndOjQIf+1117bU7SPX3/9NfTWW29V/fv3P1rWc4TK61S3AJjBqYdFGws0B1oAMcAvSqmVWus9FR+eEEKI8vLVT+kxqzZmlntFFqDvJdHpV3ePzTiTbdq1a1cQHh7uPXjwYFCDBg3cs2fPjv7kk0/iPB6PFeC+++7b37dv3xwwq76DBg3KWL9+fY3MzMygsWPHpo4fP/4wwJo1ayKefvrpeICLL744t/hMr+vXrw975pln4gsKCqwhISG+Rx55ZN+ll17qTEpKcowaNar1sGHD0tetWxflcrkszz//fNInn3xSe8uWLeEOh8P31ltv7axXr56nrOdzumMNHjw4IzExMfKaa65JHzp06JEpU6Y0OnToULDL5bIMGDAg87777kv1er089NBD8YmJiZFBQUFGaGiob8GCBdumTJkSn5eXZx80aFCbkJAQ34IFC7YVP/bcuXOjIyMjPZMmTUotWlazZk3vrbfeml4yzlWrVkXOnDmzQWFhodXr9XLLLbekXHfddVkA06ZNq/fNN99EOxwOw2Kx8Omnn2qHw2HcfffdTfbs2RNqs9mM+Pj4grfffnv3999/H/nCCy80XLx48dbbb7/9otTU1OCHHnqoqVLK+cADDxwcNWpU659//nkTmEnt9OnTG+Tl5dkA/vGPfyQPHTo0u7T35rbbbjtcPN7PPvssdsiQIZkAy5cvr/Hjjz/WTExMrLFgwYLa48aNSxs3blzGv//977pLly6NAWjVqlXes88+uy8yMtI3derU+rt37w7Jzs62p6enBzVt2rRg+vTpe2rWrHlcwhgTE+MtSoaLPkdz5syJ4zSWL18e1bJly7yWLVu6AMaOHXv44YcfbvrQQw+lAAwZMiRz7ty5sSdLiJcuXRpVv35914QJE5KffPLJxmVNiHNycqxz586t+/nnn/9Rt25dD4DVauWaa645Upbtwaz67tixI2z06NGZAKNHj8584YUX4g8dOmSPi4s74XOfl5dn83q9uFwui8fjsdarV68QICoq6tiXBKfTafV4PBaLxXLC8WbPnh3br1+/jJCQkDPq01ApCbHW+gcApdSpVrsOeFtr7QMOK6UWAKOAFyo+QiGEqHg+n/Q5C4QffvghPCoqytOhQ4d8gH79+mWPHj0602q1sm3btuCbb75Z9e3b97ei9QsKCqwLFy7clpSU5PjLX/7SduzYsRlBQUHGgw8+eNHUqVOTevXqlfP555/XWrhwYW0Al8tluffee5s9+eSTe/r27ZuzcuXKyHvvvbfZt99+uxkgJyfH3rlz59x//etfB19++eU6t912W8v33ntPd+zYMf/++++Pf+edd+KmTJmSXFrszz//fPyrr77qBbj++uvTxowZk3m6Y7Vr1y7vqaeeOgAwZsyYFnfeeWdKz549c10ul+X6669vuWLFCmdMTIx748aNkStWrPjDZrORmZlpA3jqqaf2jRo1qvWyZcu2lBbPli1bwtq2bZtXlve9Y8eOzi+++GKb3W4nNTXVfu2117bp37//UcMwmDNnTp21a9duCgsLM44ePWoNDQ31LV68uGZeXp5t5cqVfwDHYirurbfe2n3llVe2K6r+F93WB7O5xxNPPNH43Xff3VG/fn13cnJy0MiRI1tfdtllf5T23pT066+/Ro4fPz4NYODAgUe//vrrI23bts0r+kK0dOnSGkuXLo2ZN2/e1sjISN+ECROavPjii/WeeOKJgwC///57xIIFC7bUrVvXM3HixCbTp0+vd7JjAXi9XubMmRPXs2fP45LLMWPGKK/Xa+nevXv2gw8+mBwSEmIkJyc76tate2zs3Pj4+MLDhw8fO/euXbvmTZ8+Pf5kx/ryyy9jhw8fnn7llVfmejwey7p168K7det22uu4ZcuWELvd7mvVqlWpk/ZkZWXZrr/++lKTu6ZNm+b/5z//Sdq/f78jNjbWbbebKafdbicmJsa9f/9+R8mEeNKkScnjx49v1r179w4ul8t6zTXXHLriiiuOxblo0aKomTNnNkxJSQm+4447DrRv3z6/+PYul8uycuXK6LfffvuM7xqcT8OuxQN7iz3fBzQKUCxCiGou3+Xlyx8Ok+Msn4kvN+ijJGecMBb8ObHgo3fQBq5xfM+RNqOgU6dy3f/Zurp7bMaZVnErwt13390MsKSkpAQ/++yzu4KDgw2AXbt2BU+YMKFBenq6w2azGVlZWUEpKSn2oirt1VdfnQnQtGnTwoiICO+BAwcchYWFluDgYF9RdW/UqFFZzz77rBdAax0SFBRkFFWZ+/Xrl/P0008bWuuQyMhIb0hIiG/o0KHZAO3atXPGxsa6O3bsmA/Qtm1b548//ljjZOdQssnEb7/9FnqqYzkcDmPkyJFZALm5udbffvstcurUqUFTp04FID8/37pjx46QLl265Hm9Xss999zTpFu3bkcHDx58Vs0yTuXw4cP2SZMmNTlw4ECIzWYzcnJybFrrkC5duuTVr1/fNXHixKaXXXbZ0UGDBmXXqFHD165dO+cLL7wQcv/998d37do150xj+umnnyJSU1MdN99887H2tBaLhZ07dwbHxsZ6ir83pUlPT3fUqVPnpE1q1q5dW6N///6ZRZXKMWPGpE+dOrURcBCge/fu2UVV1NGjRx9+9tlnT5qgAjz00EPxoaGh3jvuuONQ0bIVK1b8Fh8f787OzrZOnDix6QsvvFDvZF+Wiqtbt677yJEj9sLCQovD4Tjum3daWpp906ZNkTNmzEgCGDJkSPrcuXNjT5cQl1Z9LalWrVrek315Ohvz58+v1bx58/zZs2dvz8nJsf31r39t8cUXX9QaMWJEFsCwYcOyhw0blr13717HnXfe2ax///7ZxZP1r776qmZcXFxh0c/XmTifEuJysXnzCc2TyywxMbEcI7kwyDlXD1XxnAvckJ4DhZ7Sfmlb2L10I4dzYPchC15fKaucxqGjFjLzLIQElV9Vt04NaNvwLIIpRQ13BldmLaCeaw8pIRcRGlu7Sl7nc1FURZw3b16tJ598skn37t03161b1/PAAw9c9M9//vPA8OHDj3i9Xjp06HBJQUHBsU5FISEhxy6S1Wo1PJ7SWzOUJWEACAoKOrY/m82Gw+E4bv9eb/l86QIIDg72Wq3mqfh8PiwWC1999dXWkkkSwLJly/5YvXp15Nq1a2u88sorDefPn3/axKZNmzbOhQsXlqk5zJQpUxr37NnzyLvvvrvLarXSu3fvBJfLZbXb7SxYsGDr2rVrI3788cca1157bes333xzR4cOHfKXLl36x3fffVdj9erVUa+++mqDpUuX/lHWczcMg4suuij/iy++0CVfS0pKchR/b0rjcDh8/s9B+V2Qk5gyZUrDAwcOBL///vs7bbY/C+Hx8fFuMJsHjBw5Mv3DDz+sA1C/fv3CDRs2RBatt2/fPkft2rWPfcMuKCiw2O12o7TrPGfOnBiv12sZOnRoWwCv12spKCiwOp3OfWFhYUZUVJQnIyPDHhkZWQhm+2qn02mrXbu2JyYmxuN2u61a62Cl1AlV4rJUiBs1alSYnp4e5PF4sNvteDweMjIygho1anRChWDOnDl1nn766SSbzUbNmjW9PXr0OLJu3brIooS4SOPGjQtbt26d980339Rs1arVseYf8+fPj7366qtPaL5TFudTQrwPaAxs8D8vWTEuk4SEhLManD4xMZFO50l1pbLIOVcPVeGcDcNg0+5cDh8xizcHDhfwxX8PlynRrVPLQXjImQ+oEx1lYUTPmlx75Wmb91Uqw+fFtX4B+Ws+xmILInTIRFp3GMDGjRvP6jq7XK5zKiRcCEaOHJm1bNmyWjNnzqz37LPP7s/Ly7M3btzYBfDhhx/Gut3u02a2SqmCwsJC65o1ayJ69OiR+8UXX9QqaqeqlCpwu92Wok5N3377baTH47EopQqSk5PPrFdbGeIo67Fq1Kjha9u2be5LL71U98EHH0wB2LdvX1BQUJBhs9mw2+3GwIEDj/bt2/fo2rVro3bv3h3cqlWrApfLZT1Zh7xRo0Zlzpo1q+7LL79cZ+LEiWlWq5Xs7GzrvHnzom+55ZbjEpHc3Fxbo0aNCq1WKytWrKiRkpISDHD06FFrbm6urVevXrm9evXK/f3338O3bt0aGh0d7YmOjvYOHz78SL9+/Y5eccUV7TMzM8ucp3Tv3j33qaeeCi7euWz9+vVhnTt3LlOnuKZNm+Zv3749pH79+m6A8PBwb05OzrFs9fLLLz/60ksvNRw/fnxaRESEb+7cubGXXnrpsY5bP/30U1RRu9h58+bFdurUqdT2vE8++WSDbdu2hX3wwQc7i7dzzczMtIWEhPjCwsIMt9vN8uXLa7Vo0SIfYMCAAdnPP/98/Pbt24Nbtmzp+uSTT2r37ds3s2jbrVu3hjZp0qTUquiiRYtip0+fvrNfv37H4rn++utbzJ8/v9bYsWMzO3XqdPTDDz+s/dhjjx0EeP/992u3atUqLzw83AcwatSotEceeaTJa6+9tqtOnToen8/HokWLaiYkJDibNWtWeLoKcZ06dTzNmjXLnzt3bvQNN9yQOXfu3OjmzZs7S2s/XK9ePdd3330X1bVrV6fL5bKsX78+sk+fPkfAbL7Rpk2bAjDvPvzyyy81Bg4cuK9o2/379wdt3rw54tVXX919qnhO5nxKiD8HblNKfYnZqe4vwJWBDUkIURm8PoO1m4+w/7ALwwCX20dyRiH7DxWQmePG64OCwuOz32b1Qhjbry7hISeO/LN9u6ZlS0V0pJ2GtStsVKxK5zmUhHPxDLwpOwhq2Y2wgX/HGhkT6LAuCA8++ODB0aNHt77rrrtS77333n0TJkxoHhkZ6enWrdvRyMjI03ZoCwkJMaZNm7b76aefjrdYLHTo0CG3qEIXHBxsvPTSS7ueeeaZ+GnTpllDQkJ8//73v4810ShPZ3qsGTNm7H788ccbDRgwoA1AWFiYb+rUqXucTqd1ypQpjb1er8Xn81m6deuW3a1btzybzUa/fv0yBw0a1DYyMtJbslNdRESE75NPPtHPPPNMw549e7YLDQ312u12Y/To0YdLHvvee+89+Mwzz8S//vrr9Vu3bp1XlLBlZ2fb/v73vzcrLCy0+nw+S8uWLfOGDx+e9f3330e+/PLLDQF8Pp/lxhtvTG3QoIF7x44dZfohjo6O9s6cOXPn888/32jatGk2j8djqVevnmvWrFk7y7J97969s1avXl2jqNNtV18AACAASURBVFnMtddem/HII480XbVqVXRRp7qtW7dmjhgxojWYneomTZp0bMrz9u3b5951110XpaenBzVp0qTg8ccfP6H98ObNm0M++eSTug0aNHCNGDGiFZhJ4Pvvv79r27ZtIU888URjMKu4CQkJuZMnTz4I5pebRx99dO/48eNb+Hw+mjdv7rzrrruOVUa///77Gn369DmhOci6devCc3JybCU72w0dOjRzwYIFsWPHjs184okn9j/22GONBg4c2MZisRAXF1c4ffr0Y6N6TJky5eBrr71WZ+zYsQrMAkX79u1ze/ToUWrCX5qnnnpq7wMPPNDk7bffrh8REeF54YUXju1/3Lhxze++++7kLl26OB977LH9jz76aPyAAQPa+Hw+S8eOHXP+9re/HQb4+OOPa2/YsKGGzWYzDMNg1KhRhwYMGHDsC8ns2bNjunXrlh0dHX1WFX5L8V6yFUUpNRO4FqgLpAMZWuu2SqklwGNa65/9w669Cgzwb/ac1vqtsh4jMTGxCZAkFeKyk3OuHsr7nLPzPGQcPbORq/alFbBxZw6l/brJd/nYui+PzJw/cxKbFerHBNOwdjC1oxxYLFAv2kHX1jWwWCxYLBBbIwirtfTCXlW7zobHTcHa2RT89DmWkEjCBo4nqNUVx92yP9tzLlYhbtqpU6c9Zxvjpk2b9nTo0OGsblUKcT7Izs62jh49utX8+fO3hoWFnVFyNHXq1PpOp9N6qk50FcXlclmGDx/e+uOPP9axsbEV3tzjQrdp06bYDh06NCm5vLJGmZgITCxl+ZBij73AnZURjxBVUb7LS1aOB6fLi9Plw+nyYhiwK9nCHuchsnI8eLwGPp+Bx2f4H4PHaz73+ZcVLTc48e+Bz4Bt+5x4vGf+RdpqgdioE2/BBjusNIoLYcil4YzqFYfDLvMFFec5uI28xTPwpe/DkdCH0H63YQ07aR8sIcRZioqK8t13330HkpKSgtu2bVsQ6HjKau/evY6JEycelGT43JxPTSaEEGcpI7uQW6Zvw+UuLVG1ASlYLRDisGK3WbBZLdhsFuxWCzYr5uOi5f7XTtZfqFvrGlyeEEXQGSSuVgs0rB1MoyrUfKGiGYUF5K/+ENeGr7DUiCXiuicIatY50GEJUaWd6WQORSZPnnzakSAqSsuWLV1F4xOLsycJsRAXoEKPj6wcD7n5XnYl5zPn+zRcboPRPeNoFR9GWLCNEIcVmw22btnKxe3b0rB2cJl7xYvAcif9inPpTHxH0gi+ZCihvf+KJbhcZ5AVQghRjCTEQlxAftqSzcqNWWzckXNcJ7OocDuTr4+nR/sTZ0/NToFGcVKZvRD4CnLJX/UuhZu+wRpdn4hxzxEUnxDosIQQosqThFiI81Ruvpf127IxDDCAzUl5rEjMpEa4ncZxIXRWkTSuE0KdmkE0qx+GzSbV3wtZ4fafcC57HSPvCMHdRxJ6xQ1Ygs68g7AQQogzJwmxEAFWUOjDZxjsOJDPH3v+nDho0bp0juQeP/JC19Y1mDSyEeGh8qNbVfhys3CueBP31v9ii2tK2KjHsNdrcfoNhRBClBv5qypEJcnMcbM5KZfk9ELSjhSSmlnIgXQX6dknH8KsWf1QHr6hMQA1I+yEBZ845q64MBmGQeHm78hf+RZGYT4hPW8ipNsILDb5tVxeMjMzbT179uxw1VVXHZ46dep+KH14rDfeeKP2H3/8Ef7KK6/sAdBaB0+bNq3hrl27wiIjIz1BQUHG3/72t9Thw4cfKeuxn3vuuXpLliyJARgyZEhG0cQYJSmlOjVp0iS/qH3/888/n9S+ffsznnZWCHFu5DevEJXgyY+S+GnLn52Xo8Lt1I120L5pBI3iggmym38ME5pE0LxB6LH1rJayTw8rLhy+7EPkLXsNz66fsTVoRfjQu7HFxgc6rCrn888/j27VqlXuqlWrol0u14GyTJSRnJwcdNNNN7WaOHHi/vfff38XQEpKin3VqlVRZT3u6tWrI7799ttaRdMOX3PNNa27deuW07Nnz9zS1p83b962yMjI8pnTWwhxViQhFqICGIbBe8tS+PGPbPIKfGTneejcMpJrrqiNahRKeIj86FVHhuHDtXEJ+d99AIZBaP87CO40FItVKv8VYeHChbGTJk068NZbb9VbtGhRzZEjR54wk1dJ7777bu2LL7746NixY49Ni1uvXj3PuHHjMsp63K+//jp6yJAhGUWTOwwZMiTj66+/jj5ZQiyECDz5qyxEOUrJdDFreSp6v5PUrEI6No8gKtyOzQo3D6pPdI0TJ6YQ1YM34yDOJTPw7P8De9OOhA2egK1mnUCHVSEKfl4UU/j7qtiK2LejXd/0kM7DTpucbtq0KTQnJ8feu3fvnEOHDgXNnz8/tiwJsdY6rFu3bicdi3b69Ol1V6xYEV3aaw8//PC+Hj165KalpTm6du16bFrb+vXrF27cuDHyZPscM2aM8nq9lu7du2c/+OCDySEhIRU/hawQ4jiSEAtxhvak5uN0/Xl30zBgRWImmTluNmjzb2BCk3D+cnksw7rHnnR6YVE9GD4vrv99Sf6aT7AEOQi76h4c7fpJU5gKNnv27NhBgwZlWK1Whg8fnvXiiy/GHzhwIMhisZxTsjlp0qTUSZMmpZZXnCtWrPgtPj7enZ2dbZ04cWLTF154od6UKVMCNsmDENWVJMRClNHbi5P5aUs2KZmFpb4eHmKlV4ea9OpQk66ty9zcUFRhnrTdOBe/jDd1F0HqMsIG3ok1otTiYpUS0nlYRlmquBXF5XJZVq5cGR0UFGQsW7YsBsDj8Vg+++yzmOjoaM/BgwePG88uKysrKDo62g2glHJu3rw5/GT7LkuFuE6dOoUHDx50FC1PTk52xMXFlfqLIz4+3g3mtMEjR45M//DDD6vmbQMhznOSEAtRTFaOm/8sOkhSSgE5+X9OC5+dZw5/ZrXC+GH1aRh7/EQXjiALCU3CpeonADA8hRT8MJuCdfOwhEYSfu3DOFpdHuiwqo1FixbVbNiwYcH8+fN10bIff/wx/OGHH276+uuv73zvvffq7d+/P6hRo0bujIwM28qVK2s9/PDD+wBuvvnmw9dcc02b2bNnR48ZMyYTIC0tzb5ixYqocePGZZSlQjx06NCsqVOnNrrtttsOASxZsiRm8uTJ+0qul5mZaQsJCfGFhYUZbreb5cuX12rRooWMMCFEAEhCLATg8xn8b9tR3luWwoHDLrq1rkF0jSCKp7d2m4WbB9fDYbcGLE5x/vMc2ELe4hn4Mg7gaNeX0H63YQ09afNRUQHmz58fO3jw4Mziyy677LI8wzA4dOiQ/Z///Of+O+64o7nP57MYhsF11113qG/fvjkADRo0cM+aNUtPmzatweuvv94gJCTEGxoa6rvlllvK3EyiV69eOT/99NORQYMGJQAMHjw4o1evXrkAixYtilq1alXNl19+ee+2bdtCnnjiicYAXq/XkpCQkDt58uSD5fdOCCHKShJiUS05XV5+25VLboGXvHwvS9ZnsO+QC5sVbugbx4396gU6RHGBMQrzyf/+Q1w/L8JaI5aIMU8SdFGnQIdVLX3yySc7Slu+evXqzUWPR48efdIOdq1bty6YNWvWrnOJYfLkycmTJ08+oS3wsGHDsocNG5YNZpK+fPnyLedyHCFE+ZCEWFQ7c75L44Nvji/2WC0wtGsM115Zm/oxMl2uODPu3RtxLn0FX/ZhgjsNJbTX/2EJDgt0WEIIIcpIEmJRLRiGwY9bjvLmchuHc1IJcVgZ2Dma4ZfHEh5sIzzEhs0m7X/FmfHl55C/6h0Kf1uJNbohkTc+h71R20CHJYQQ4gxJQiyqvN0p+bz1dTKbducSVwPG9I5jWLcYoms4Tr+xECdRuG0tzuX/wXBmE3LZaEKuuB6LXT5TQghxIZKEWFRJBYVeXv7iACmZLnYczCcixMbfr25AHfteLu0i7YPF2fPlZuJc/gZuvRZbnWaEjXkCe51mgQ7rfODz+XwWq9Uqk0oIIc5LPp/PAnhLe00SYnFB8/kMUrMK+XVXLnnFhklb9UsW+w4VcHGzCEZcWZvRveKIDLWTmLg3gNGKC5lhGBT+vpL8le9guF2E9vo/grtei8Umv0b9Nh8+fLhN7dq1syUpFkKcTwzDoLCwMCg5OTkC+KG0deQ3ubhg5Tg9PPLebnYcLH3Yzn6X1GLSqPhKjkpURd4jaTiXvoonaSP2hm0JGzoRW0zDQId1XvF4PLempqa+k5qamgDI2IRCiPOJz2KxZHu93pk+n+8/pa0gCbG4ICVnuPjnf3Zw1OnlrwProhqF0arR8ZNLBQdJJzlxbgzDh+vnr8n/fhZYLIQOuJPgTkOwWCTfK6lTp06HgKsDHYcQQpwNSYjFBWXjjqN8vDKNA4ddeA2DqbdcRIdmMumBKH/e9P3kLZmJ98AW7Bd1ImzwXdii4gIdlhBCiAogCbG4oLy7NIXdKQX0bF+T3hfXlGRYlDvD66Fg3RcU/PApFkcoYcMm4UjoLdNyCyFEFSYJsbgg5Dg9vPzlfnanFDC0awx3/UXab4ry50ndiXPxDLxpuwlqdQVhA8ZjjagV6LCEEEJUMEmIxXlvb1oBD7+7iyO5Hq65PJa/DZJh00T5MtwuCn74jIJ1X2AJiyJ8xCM41GWBDksIIUQlkYRYnJe8PnPUpsyjhdzz+g4swLPSXlhUAM/+P8hbPANf5kEcHfoT2ucWrKHyORNCiOpEEmJx3sjMcfPNz5l8+0sW+w+7ji23WuCVCS25qF5oAKMTVY3hcpL//SxciV9jjapDxPVPE9S0Y6DDEkIIEQCSEIuA8voMftudy8K16fy8/SheHzSIdXBdrzgcditg0K5puCTDoly5d/1M3tJXMY6mE9xlOKE9b8LiCAl0WEIIIQJEEmIRMF6vwcPv7eK33Xk47BZ6X1yLYd1jaNkw/PQbC3EWfM6j5K98m8LN32KNaUTETS9gb9g60GEJIYQIMEmIRaVzurwkpeQz9bO9ZBz1cFG9EJ67rTkRobZAhyaqKMMwcG/7AefyNzAKcgi5fAwhl4/BYg8KdGhCCCHOA5IQiwphGAY7DuazQR8lJaPw2PLMHDe/7MwFwGaFO4fVZ0jXWOw2GeNVVAxfbibOZa/j3v4TtrrNCbvhaexxTQMdlhBCiPNIpSXESqmWwCwgBsgAbtJa7yixThzwPtAICAK+AyZqrT2VFac4dysSM/lgeQqZOeZlqx0VhNVqJrwWCwzrHkO96GCa1gvhYhk1QlQQwzAo/G0F+SvfwfC6Ce1zM8GX/gWLVe5ECCGEOF5lVojfAF7TWn+slBoHvAn0KbHOw8BWrfVQpVQQ8ANwLTC3EuMUZ2Hrvjy27XNS4Pbx4TepxMcFM6pnHJe1jSKupiPQ4Ylqxu7MJPezR/Ds2YS9UQJhQydii24Q6LCEEEKcpyolIfZXfi8B+vsXfQa8qpSqrbU+XGxVA4hUSlmBYMABHKyMGMXZ+z0pl3/NSiLf5QMgOMjCfaPiadEwLMCRierG8Hlx/byIhj9+gMdmJ2zQP3B0HITFYg10aEIIIc5jFsMwKvwgSqlOwIda67bFlm0BxmmtNxZbFg18AbQBwoFXtdYPleUYiYmJTYCk8oxbnFyhBxb9YiUz18LeDAs2q8HfrvRSJwrsNgiSu9KikgXlHqL25gWEZB/AGduCw22uxhsaFeiwLiRNO3XqtCfQQQghRCCcb53qRgG/AX2BSGCpUmqk1npeWXeQkJBAcHDwGR84MTGRTp06nfF2F7KzPecjuR7eX5bML3uziAq3Uz/GyvhhDeiialRAlOVLrnPVY3jdFPw0j4J1s7E4wggbfj+78yPo1LlzoEOrVGd7nV0uF5s3b66AiIQQ4sJRWQnxfqCBUsqmtfYqpWxAff/y4iYAN2utfUC2Umoh0Bsoc0IsKo7XZzD/h8PM/+EwmTkeWseHMX18cywWGSFCBIYnZQfOxTPwHkoiqE0PwvqPxxoeBYmJgQ5NCCHEBaRSEmKt9SGl1K/A9cDH/v9/KdF+GMwmD4OA9UopB9AP+LIyYhSnlpSaz78+SOJwtptGccHcPKgefTrWkmRYBIThLiD/v5/i+t98LOG1CB85BUfLboEOSwghxAWqMptMjAdmKaUeA7KAmwCUUkuAx7TWPwP3AG8opX4HbJjDrr1diTGKk3h1/gFy873cMrgeI66sLYmwCBj33t9xLpmJLysZx8UDCetzC5YQmd1QCCHE2au0hFhrvQ3oWsryIcUe7+LPkSjEeWLJ+gy27HMypnccI3vEBTocUU0ZLifOb9+n8JclWGvWJeKGZwlq0iHQYQkhhKgCzrdOdeI8syIxk1cXHKBFg1Cu7y3JsAgM98715C19DSM3k+BLryG05zgsQSGBDksIIUQVIQmxOEGh28fbS5JJyypkg84hPi6Y525rhkPGUhOVzOfMJn/FWxT+8T3W2MZEXDsZe4NWgQ5LCCFEFVPmhFgp1R8YA8RprYcppToDNbTW31ZYdKJS7UvLZ1dKAWt+O8K6rUdpWDuY7m1qMLZvXUKDJRkWlccwDNxb1uD85g0Ml5OQK24g5PLRWGxBgQ5NCCFEFVSmhFgpNQG4G3gHGOlfnA/MBC6rmNBEZVqRmMnLX+zH55+n5eruMdx5dcPABiWqJV9OOs5lr+Pe8T9s9VoSPvRubHFNAh2WEEKIKqysFeJ7gL5a6z1KqQf9y7YBqmLCEpUpPbuQGV/up1aknSnjmlIr0k5cTUegwxLVjGEYFP66HOe374LXS2jfWwjuMhyLVe5OCCGEqFhlTYgj+XMSjaK5noOAwnKPSFSqzBw30z7bi88HU29pRqM46agkKp83KwXnkpl49v6GPb4dYUMmYouuH+iwhBBCVBNlTYjXAA8BzxRbNhFznGBxAXvywyT0gXxGXBkrybCodIbPi2vDV+Sv/ghsNsIGT8Bx8QAsFmugQxNCCFGNlDUhngAsUkrdBkQqpTSQA1xVYZGJCnfoKOgD+fS7pBa3DJZqnKhc3kN7yFsyA2/ydoJaXErYoH9gjYwNdFhCCCGqoTIlxFrrFKVUF6AL0Biz+cR6rbWvIoMTFcfrM/j0J7Nt5sgeMvOcqDyG103Bj59TsHYOluAwwoc/QFCbHvIZFEIIETBlHWViodZ6OLDe/69o+Zda62srKjhRcZ79dA/pORaGdY+hcZ3QQIcjqglPsiZv8Qx8h/fiaNuL0P63Yw2LCnRYQgghqrmyNpnofZLlvcopDlGJ1m3J5sc/jtKmvo/retUJdDiiGjDcBeSv+RjX+oVYImoRPupfOFpcGuiwhBBCCOA0CbFS6kn/Q0exx0UuAvZWSFSiwhzKcjFj/gHCgq1c29lDTA2Z6EBULPeeTTiXzMR3JBVHx8GE9f4blpDwQIclhBBCHHO6CnEj///WYo/BHHptP/B4BcQkKsjc79NY9FMGR3I93Dq4HiGO/affSIizZBTk4fz2XQp/XY61Vj0ixk4lqHH7QIclhBBCnOCUCbHW+m8ASqkftdZvV05IoiJkHHXz8ao03B6DSSMb0a9TNImJkhCLilG44384l76GkZdFcNdrCe0xFkuQDOsnhBDi/FTWUSbeBlBKRQKxgKXYa7srJjRRHgzD4Eiuhynv7wYD3rynJfHSiU5UEF9eNs4Vb+DesgZr7cZEjHwUe/2WgQ5LCCGEOKWyjjLRGvgU6IDZXMLCnzPWybyq5ymfz2DCq9vZnVIAwO1D60kyLCqEYRi4t6zG+c2bGC4nIT3GEdJ9JBabtFEXQghx/ivrKBP/wZyVrjeQBDQBpgI/VkxYojwcSHexO6WAdk3DGXFlbbq2luGtRPnzHT2Mc9lruHduwFZfET70bmy1Gwc6LCGEEKLMypoQdwD6a63dSimL1jpbKXU/sBn4uOLCE2crM8fN5Hd2AXDTgLokNIkIcESiqjEMH4W/LMP57Xtg+AjtdxvBnYdhscpNIyGEEBeWsibEBUAQ4AbSlVLxQBYQU1GBibPn9Rk8/O5usvM8DOoSjWoYFuiQRBXjzTyIc8krePb9jr1JB8IGT8BWq16gwxJCCCHOSlkT4v8Co4EPgHnAUsAFfFsxYYmzZRgGE17Zzt60Am7sV4cb+tYNdEiiCjF8XlzrF5C/5mMstiDChkzE0WGATLsshBDiglbWUSZGF3v6MPAHEAHMqoigxNk7dMRNUmoBAztHM6a3zEInyo/nUBLOxTPwpuwgqGU3wgb+HWuk3CQSQghx4StrhfgYrbUP+Egp5QBuA14r96jEWVvzWxYAfTrWxGqVqp04d4bHTcHa2RT89DmWkEjCr3mIoFZXSFVYCCFElXHahFgp1Re4GNiptV6olLIDfwceBDKRhPi84fUafL7mMM3qh9D+oshAhyOqAM/BbeQtnoEvfR+OhD6E9rsNa1iNQIclhBBClKtTJsRKqQeBKZhNJNoqpV4HemG2H75da724wiMUZXY4u5Acp5exfaWphDg3RmEB+as/xLXhKyw1Yom47gmCmnUOdFhCCCFEhThdhfgOoKfWOlEp1Q1YC0zSWr9c8aGJM5WW5QagYaxMkSvOnjvpF5xLXsGXnUbwJUMJ7f1XLMEyUokQQoiq63QJcazWOhFAa71OKeUCZlR8WOJs7DzoBKBBrCPAkYgLka8gl/xV71C4aQXW6PpEjHuOoPiEQIclhBBCVLiytCG2YE7VbMEcjxillLXodX8nOxFAeQVe7n9zJ0mpBdSKsFO7piTE4swU6p9wLn8dI+8IId1HEXLF9ViCggMdlhBCCFEpTpcQRwCeYs8txZ5bAAOQaakCxOcz2JWcz6ffppGUWsA1V8Ryedua2GR0CVFGvtwsnN+8gXvbD9jimhI26jHs9VoEOiwhhBCiUp0uIW5aKVGIs/LxylQ+++4QAK3jw7h9aIMARyQuFIZhULj5O/JXvoVRmE9Iz5sI6TYCi+2MR2IUQgghLnin/Ountd5bWYGIsnO5fby1OJlVGzOJCLUx9ZaLiK8jHelE2fiyD5G39FU8uxOxNWxN+JCJ2GLjAx2WEEIIETCVVg5SSrXEnNkuBsgAbtJa7yhlvdGYQ70VNcnop7VOq6w4LwQvzN3H2s3ZtG0cxv2j46kTLW09xekZhg9X4hLyv/8ADIPQ/ncQ3PkqLBbrabcVQgghqrLKvD/6BvCa1vpjpdQ44E2gT/EVlFKdgceBPlrrVKVUFOaYxwIo9Ph4+uM9bNA5dG9Tg8dulBYtomy8GQdwLp6J58Af2Jt2JGzwBGw1ZbxqIYQQAiopIVZKxQGXAP39iz4DXlVK1dZaHy626r3Ai1rrVACtdXZlxHchMAyDW1/cxuFsNx2bRzBplNziFqdn+LxE7V7D0ZWrsQQ5CLvqHhzt+sm0y0IIIUQxZ5QQK6UaAQ201uvO8DiNgINaay+A1tqrlEr2Ly+eELcBkpRSazBHuPgSeEZrbZzh8aqcz9cc4nC2m0tVJI//X1NJaMRpedJ24Vw8g5jUXQSpywgbeCfWiOhAhyWEEEKcd8qUECul4jGruhdjtuuNUEqNBAZprW8tx3hsQHvMSrIDWAbsAz4s6w42b9581gdPTEw8620rkmHA/DU26kTBsIQsNm7MKrd9n6/nXJGq+jlbvG5q7l5NzaQf8AaFkdHhOvLqtgWdBCQFOrxKU9Wvc2mq4zkLIUR5KGuF+E1gMXAlZoc4gBXA9DJuvx9ooJSy+avDNqC+f3lx+4B5WmsX4FJKLQQu5QwS4oSEBIKDz7yTWWJiIp06dTrj7SrDlr15HHHu5I6r6tO5c+1y2+/5fM4Vpaqfs2f/FvIWv4Uv8wCO9v0I7Xsr+7Zsr9LnXJqqfp1Lc7bn7HK5zqmQIIQQVUFZu5dfCkzzz0pnwLH2vVFl2VhrfQj4Fbjev+h64JcS7YcBPgUGKKUsSqkgoC+wqYwxVklen8F9b+4EoO8ltQIcjThfGYX5OL95g5yPHgBvIRFjniL8qnuxhkYGOjQhhBDivFfWCnEa0BzYXrRAKdUGs6JbVuOBWUqpx4As4Cb/fpYAj2mtfwZmA52BLYAPWA68ewbHqHJWbzqCYUCLBqFEhsqkCeJE7t0bcS59BV/2YYI7X0Vor//D4ggNdFhCCCHEBaOsGdaLwNdKqamAXSl1PfAwMK2sB9JabwO6lrJ8SLHHPuCf/n/Vns9nMHP+fmJqBPHSnTKdrjieLz+H/FXvUPjbSqzRDYm88XnsjdoEOiwhhBDiglOmhFhr/Z5SKgO4A7Pd703AFK31gooMrjrzeg0+XpWKy21w7RW1sNlkVAnxp8Jta3Eufx3DeZSQy0YTcsX1WOyOQIclhBBCXJDKOsqETWu9EFhYwfEIv0fe282m3blER9oZ00cmUBAmX24mzuX/wa1/xFanGWFjnsRep1mgwxJCCCEuaGVtMpGqlPoc+ERrvbYiAxKw/YCTTbtziY8LZsq4JjjsMrVudWcYBoW/ryR/5dsY7kJCe/+V4K7XYrHaAh2aEEIIccEra0I8AHNkiM+UUl7Mzm+faq1/r7DIqqncfC9TP90LwKRR8TSsHRLgiESgeY+k4Vz6Cp6kX7A3bEvY0InYYhoGOiwhhBCiyihrG+JfgF+AB5RSPTGT42+VUila6/YVGWB189GKVFKzCrl3RCNaNgwLdDgigAyfF1fiYvK/nwUWC6ED7yT4kiFYLHLHQAghhChPZzOO1zZgK+aQazL0QTkyDIOlGzJo0SCUAZ1lit3qzJu+j7wlM/Ee2Ir9ok6EDb4LW1RcoMMSQgghqqSydqqrCYwAbgC6Ad8AzwFfVVxo1c+hI27cHoMr29UMdCgiQAyvh4J18yj44TMsjlDChk3CkdAbi0VGGRFCCCEqSlkrxMnAj5gzyY3QWh+puJCqJ8MweHtxMgBdW8nsYtWRJ3Unzq9fxnsoiaBWVxA2YDzWCJmdUAghhKhoZU2Im2mtUyo0kmruo5VpI6qM1gAAIABJREFUrP0jm+GXxRBfR2YZq04Mt4v8Hz7Fte5LLGFRhI94BIe6LNBhCSGEENXGSRNipVQPrfUa/9PWSqnWpa2ntf62QiKrZtZuPkKTOiHccVWDQIciKpF732acS2biyzyIo8MAQvvcjDVU7hAIIYQQlelUFeLXgQT/43dPso4BXFSuEVVDhR4fB9NdDOgcLW1FqwnD5ST/uw9wbVyMNaoOEdc/TVDTjoEOSwghhKiWTpoQa60Tij1uWjnhVD/p2YXcOG0rAEqGWasW3Lt+Jm/pqxhH0wnuMpzQnjdhcch400IIIUSglGlAU6VUqVM2K6W+LN9wqp8PlqcC0LdjLTq1lFvlVZnPeZS8r6aTO+dfWIJCiLzpBcL63y7JsBBCCBFgZe1U1/sky3uVUxzVktdnsOqXLKIj7dw3Oj7Q4YgKYhgG7q0/4Pzm/9u78/A463rv4++ZyTpJS3e60FKg9ItYthYsBcpiWWpFBQHZBEUW4SAg6qMePCDq4Vz6PMdHKNBDQeBw2ETwiD62UNayLxLKUkq/LdAl3femzUyWmbmfP2Z6CDFNJu1kJsl8XtfF1cyde+7780ti/F6/fO/f7z8IGrZTcfQ5VBx9DqGS0kJHExERETooiM3sl5kPy1p8vMO+wLIuSVUkXno/vXrd0eP2KHAS6SqpbRuJzZlB86LXiQwdQ/Tcf6VkT7Xdi4iIdCcdzRCPzPwbbvExpB+mqwVu7IJMReP3s9PrDn/r5GEFTiK5FgQBTe8+RfzZuwmSzVSecBHlE08nFI4UOpqIiIi00m5B7O4XAZjZq+5+V34i9X5BEHD/M2vZWJdgn6EVVFWoSOpNkptXE5t9K4ll71IychzRL19NZICW0xMREemu2luHeLS7L828fNbM2vw7r7t/0hXBerOHnl3Lw8+tpaoizI/PVu9wbxGkkjS+9f+Iv/BfEAoTnXolZYdNJRTK6tlVERERKZD2ZojfB3Yse/AR6TaJ1ovkBoCmNzth7eYmHnxuLWOGV/LbK8ZQVqJiqTdIrl9G/axbSK5ySvY7nKovfY9w38GFjiUiIiJZaG8d4j4tPlbVlgPrtzRy9W2LCQK4/CvDVQz3AkGymYbXHqPh5T8QKo8S/eqPKPv88dpgRUREpAfJdtm1z8i0T6RatFRIFp6u2UxdLMlVp+3F50dXFzqO7KbEqkXEZt1Ccv1SSg88luhJ3yVc1a/QsURERKSTst2Y42EzOyrz8UXAB8AHZnZxV4brTZoSKZ6q2UxZSYhpEwcWOo7shqC5gdizd7Ptvh+Sim+j6szrqT7tJyqGRUREeqhsZ4inAN/KfPwD4ERgC/A4cHcX5Op1bv5TLWs3NzFhf80M92TNy94jNvtWUptXUXboKUS/eDGhiqpCxxIREZHdkG1BXObuTWY2Ahjg7q8AmNmeXRet90gkA156fysTD+jDzy/cp9BxZBcEDfXEnr+XpnlPEO43lOrz/o3S0YcUOpaIiIjkQLYF8Ttm9s/A3sAsgExxXNdVwXqTpWviJJIBR4/rp4eteqDmj96k/onbCLZvpnzi6VQe+01CpRWFjiUiIiI5km1BfDHwK6AZ+HHm2CTgwa4I1dssXhkDYP8RlQVOIp2Rqt9K/Jk7afpgLuHBe1N9xs8oGW6FjiUiIiI5llVB7O4fA+e1OvYY8FhXhOptFiyLURoJMXKIZhV7giAIaF7wIrGn7iBojFEx+XwqjjqLUKS00NFERESkC2S97FpmdYkLgBHASuB+d7+3q4L1FolkwJsL6xgzopJIWO0S3V2qbgOxObfTvPhNIsPHUjXtGiJDRhc6loiIiHShrApiM/sZcCHwW2AZ6V7iH5vZcHe/qQvz9XjvfLSNuliS7x83pNBRpB1BkKLpnTnEnrsHkkkqp1xC+RFfJRTWRowiIiK9XbYzxJcAx7v7sh0HzGwO8CKggrgdc9/dQnVFhAlj+3R8shREctMqYrOnk1j+PiV7H0x02tVE+g8rdCwRERHJk2wL4ipgfatjGwE9JdaOTXXNPDtvM6ccPkDbNHdDQSpJ49//QvyFByASITrtasoOOVkrgYiIiBSZbAviJ4EHzeynwHLSLRM3AXOyvZGZjQXuAwaSLqYvdPfFOznXgHnADHf/Ubb36E6CIOCmh5YCcOKE/oUNI/8guW4p9bNuIbl6EaX7f4Ho1CsJ9xlU6FgiIiJSANlOW34P2Aa8B2wH3gHqgas6ca87gNvdfSxwOzCzrZPMLJL53OOduHa3887H21mwLMZpRw9i3GjtTtdtpBLEX3yQunuuIbV1LVWn/YSqM29QMSwiIlLEOpwhNrM9gP2AK4FvA4OADe6eyvYmZjYEGA+clDn0MHCbmQ1299atGD8F/gZUZ/7rkWa/sZHSkhDnTdFmft1FYpWz12t30LB9HWWfP57Kky4jHN2j0LFERESkwNotiM3sy8AfSfcKbwNOc/fnd+E+I4GV7p4EcPekma3KHP+fgtjMDgFOAU4Art+F+zB//vxdeRsANTU1u/zeluob4eX5JYwbkWLRgndzcs2ukqsxd2ehRBP9P3qOPZa9RriiD6vHn098sMGHHxU6Wt4Uw/e5NY1ZRESy1dEM8a+AnwD3AJeS7hs+qiuCmFkpcCdwUaZg3qXrjBs3jvLy8k6/r6amhgkTJuzSPVub9cYGYCVfPW4fJhzULyfX7Aq5HHN31bz0XWKzZ5Dasoby8dNY2P9Qxk88utCx8qoYvs+taczZa2xs3K2JBBGR3qCjHuJ93f02d4+R7vsds4v3qQVGZPqDd/QJD88c32EY6daM2Wa2FPg+cKmZ3bmL9yyIuvoEtz2+kn7VJRx1oP4cXyiphu3Uz57O9oeug1CY6vN/TXTqlQQl2i1QREREPqujGeL/KZjdPWFmWe9s15K7rzOzd4BzgQcy/85r2T/s7stJ9ycDYGY3AtU9bZWJB55ZA8DJEwYQiWj5rkJoWvQ6sSdnENRvpvzIM6icfD6h0s7/1UBERESKQ0cFbtTMXmzxuk+r17j7sVne63LgPjO7AdhMeuc7zGw2cIO7v5XldbqtVCrgtQ/r+NyoKBdN1cYO+Zaq30LsqZk0f/gikcGjiZ51PSXD9i90LBEREenmOiqIL271+u5dvZG7LwQmtnF82k7Ov3FX71UoW+sTbNjazBcP7b59w71REAQ0fTCX+NN3EjTFqDj2AiomnUEoUlroaCIiItIDtFsQu/t9+QrSGyxZ3QDAwfv12NXiepxU3Xrqn7iNxMdvERlxAFXTriEyeFShY4mIiEgPsks9wdK2JWvjAIzeUztad7UgSNH09hPEnr8XghSVJ15G+eGnEgpHCh1NREREehgVxDmyPZ7k97NXU1YSYkAffVm7UnLTSmKzppOonU/J6EOJTruKSL+hhY4lIiIiPZQqtxxJrz0M4/apIhTS6hJdIUglaXzjz8RfepBQpJTol6+h7OCT9PUWERGR3aKCOEfe8m2Ul4b414v2LXSUXimx9hNis24hueYjSsdOInrKFYT7DCx0LBEREekFsiqIzawcuIH0+sED3X0PMzsZGOvut3VlwJ4gmQzwFTGOPaifZitzLEg00/DKH2h47VFCFX2oOv2fKT3gaH2dRUREJGeynSH+HTACOB94InPsg8zxoi+I31q0jeZEwGFjtLpELiVWfEj9rFtIbayl7KApVE65hHC0b6FjiYiISC+TbUF8OjDG3evNLAXg7ivNbETXRes5pv+5lqqKMJMP0vrDuRA0xYm/cD+Nf/8r4b6DqD77F5Tud3ihY4mIiEgvlW1B3NT6XDMbDGzMeaIepimRYtO2BMeM60tZabjjN0i7mpfMIzb7VlJb11I+4VQqj/8WofJooWOJiIhIL5ZtQfwo6W2XrwUws2HAzcAfuipYT7FkdXrt4SM/t0eBk/Rsqfg24s/dTdO7TxMeMILqb/6G0lHjCh1LREREikC2BfF1wG+A94EosBi4C/hFF+XqMWrXNwIwZoRmMXdVk79K7MkZBLGtVEw6i4rJ5xEqKSt0LBERESkSWRXE7t4EXAtcm2mV2ODuQZcm6yHijSkA+ka1Q1pnpbZvJvbUHTQvfJnInvsSPftGSoaOKXQsERERKTLZLrvWenHdPmYGgLt/kutQPcnbi7fRr7qEvlVa0jlbQRDQNP854k/fSdDcSMXx36Ji4tcJRfQ1FBERkfzLtgL5CAiAlou/7pghLtqp0SVr4rz+YR1fmTSQSFjr4mYjuXUdsSduI/FJDZG9PkfVtGuIDBpZ6FgiIiJSxLJtmfjM8glmNhT4OfBSV4TqKeYvqQfgkH21/nBHgiBFY81s4nP/E4KAypO+S/nhpxIKaWUOERERKaxd+hu1u68xs+8Di4CHchup55j77mYAjjxQK0y0J7lxBbFZ00ms+ICSfcYT/dL3iPTbs9CxRERERIBdLIgzjPSKE0UpCAJWbWxi6IAytUvsRJBM0PjGfxN/6SFCpeVET72WsoOmaNtlERER6VayfajuJT7tGYZ0Ifx54JddEaonWLelmS3bE1z5NW3W15bEmo+JzbqF5NqPKT3gaKInX064ekChY4mIiIj8g2xniH/f6nU98K67L85xnh7Da2MAHDCyaCfJ2xQkmmh4+WEaXnuMULQvVV+/jrIDji50LBEREZGd6rAgNrMI8EXgMndv7PpIPYPXxigrCTF6aGWho3QbidoF1M+6hdSmFZQdfBKVUy4mXNmn0LFERERE2tVhQezuSTM7GUjlIU+PsbC2njEjKimJqB82aIwRn3sfjTWzCO8xmOpzfkXpvuMLHUtEREQkK9muefU74BdmVtqVYXqKRDLgo5VxbC+1SzR/UkPdXf9EY80syg//Cn0vnaFiWERERHqUdmeIzexcd38YuAoYCvzAzNbT4gE7dx/VtRG7n6Vr4zQlAmxU8RbEqfg24s/cRdP7zxIeuBd9LvjflIw8sNCxRERERDqto5aJmcDDwDfzkKXH2PFAXbHOEDctfIXYnBkE8W1UHHU2FcecQ6ikrNCxRERERHZJRwVxCMDdX8hDlh7Da2PsUVXCnv2LqwhMbd9EbM5/0OyvEhm6H9FzfknJnvsVOpaIiIjIbumoII6Y2QlkCuO2uPtzuY3U/S2sjWEjo0WzwUQQBDS9/wzxZ+4iaG6i8oRvUz7x64TCkUJHExEREdltHRXE5cDd7LwgDoB9c5qom6tvSLJiXSPHHdSv0FHyIrllLbEnbiWxZB4lIz9PdNo1RAZqMxIRERHpPToqiOvdvagK3o6s3tRIAIwcUl7oKF0qSCVprJlFfO59EApRecoVlI+fRiiU7cIkIiIiIj1DtjvVSUZ9PAlAn2jv/dIlNyynfvZ0kis+pGS/w6maeiXhPYYUOpaIiIhIl8jqoTr51KIVcQD6VfW+gjhIJmh4/TEaXn6YUFkl0a/+kLLPn1A0vdIiIiJSnNqt6tw9Z/vumtlY4D5gILARuNDdF7c653rgHCAJNAPXufucXGXIhQ+W1We2bK4odJScSqz5iNjfbia5bgmln5tM9OTLCVcVR5+0iIiIFLd8NoTeAdzu7mOB20mvcdzam8AR7n4w8B3gETOrzGPGdjUnUtQs2saU8f17zaxp0NxI7Pl72XbvtaRiW6k641+oPv2nKoZFRESkaOTl7/5mNgQYD5yUOfQwcJuZDXb39TvOazUb/B7plo2BwIp85OzI6x/WkUgGHDiqqtBRcqJi81Lq7p5JatNKyg45mcopFxOuqC50LBEREZG8ylcj7EhgpbsnAdw9aWarMsfX7+Q9FwIfu3u3KIYB5ry1kYqyMIfs17OLxqAxRvz5/2T427Og355Un3sTpfscWuhYIiIiIgXRLZ8MM7PjgF/x6Yxy1ubPn7/L962pqdnp55oS8N7HEQZUw/KP32f5Lt+lsCrXL2Lwgv9HpKGOrXtPYvOYKQSbkrBp52Pvbdr7PvdWGnNxKMYxi4jkQr4K4lpghJlFMrPDEWB45vhnmNkk4AHga+7unb3RuHHjKC/v/BrBNTU1TJgwYaeff33BVpqTS7noS6OYcEj/Tl+/0FKxOuLP3EXT/OcIDxpF1dk3sGRNfbtj7o06+j73RhpzcdjVMTc2Nu7WRIKISG+Ql4fq3H0d8A5wbubQucC8lv3DAGZ2BPAIcKa7v52PbNlasLyeUAi+cEDfQkfplCAIaFrwEnV3Xk7TgheoOOZc+n5nOiUjDih0NBEREZFuIZ8tE5cD95nZDcBm0j3CmNls4AZ3fwuYAVQCM81sx/sucPf385izTR8ujzG0fxmV5ZFCR8laattGYnNm0LzodSLD9id63k2UDNmn0LFEREREupW8FcTuvhCY2MbxaS0+PiJfeTqjviGJ18aYdGDPmB0OgoCmd58i/uzdBMlmKr/4Hcq/cBqhcM8p5kVERETypVs+VNfd/PXVDTQnAk6eMKDQUTqU3Lya2OxbSSx7l5JRBxGddjWRAcMLHUtERESk21JB3IF4Y5JZb2wE4ND9crZxX84FqSSNb/2V+Av3QyhMdOr3KDvsFEKhfO69IiIiItLzqCDuwH+/tJ6Ndc1c/pXhRCLdc3e65Ppl1M+6heQqp3TMEUSnfo9w30GFjiUiIiLSI6gg7sCbXkdVRZivHTW40FH+QZBspuHVR2l45RFC5VGqvva/KD3wuF6zrbSIiIhIPqggbseGrc0sWhHnmHF7FDrKP0isWkRs1i0k1y+l9MDjiJ70XcJV3S+niIiISHengrgd0/9cSwiYekT3eZguaG4g/uKDNL75OKHq/lSddQNl+//D4h0iIiIikiUVxO1YWBvj4H2rmDC2eyy31rzsPWKzbyW1eRVlh04l+sXvEKqoKnQsERERkR5NBfFOBEFAQ2OK4YM6vw10zrM01BN7/l6a5j1BuP8wqs/7N0pHH1LoWCIiIiK9ggrinViypoHmZMCYEZUFzdG0+E1iT95GsH0z5RO/TuWx5xMqrShoJhEREZHeRAXxTry1aBsARx5QmAfVUvVbiT09k+YFLxAevDfVZ/yMkuHW8RtFREREpFNUEO/Emwvr2HvPCgb0Lc3rfYMgoHnBC8SemknQGKNi8vlUHHUWoUh+c4iIiIgUCxXEbQiCgKVr4hw2Jr8706XqNhB78naaP3qTyPCxVE27hsiQ0XnNICIiIlJsVBC3Ycv2BPUNKfYbnp/+4SBI0fTOHGLP3QPJJJUnXkL54V8lFI7k5f4iIiIixUwFcRs2bUsAsFceVphIblpFbPZ0Esvfp2TvQ4hOu4pI/2Fdfl8RERERSVNB3IZNdc0A9O/TdX27QSpJ49//QvyFByASITrtasoOOVnbLouIiIjkmQriNny8Og7AkH5dUxAn1y2lftYtJFcvonT/iUSn/hPhPoO65F4iIiIi0j4VxK0kkgF/e30DB+1TxeB+ZTm9dpBopuHVR2h49Y+EKqqpOu0nlH5usmaFRURERApIBXEry9c1sLEuwXlT9szpdRMrF1I/azqpDcsoG3cClSdeSjhamDWORURERORTKohbWbYm3S5he1Xl5HpBUwPxF++n8c2/EOozkOpv/JzSMV/IybVFREREZPepIG5l6doGQiEYOXj3V5hoXvousdnTSW1ZQ/n4aVSecBGh8mgOUoqIiIhIrqggbuWjVXGG9CulrDS8y9dINWwn/tw9NL0zh/CA4VR/8zeUjhqXw5QiIiIikisqiFupXdfIXrsxO9y06DViT84gqN9C+ZFnUDn5fEKlXb+esYiIiIjsGhXELWzZnmD91maOO6Rfp9+bqt9C7KmZNH/4IpEh+xA96wZKhu3fBSlFREREJJdUELfw+CvrATo1QxwEAU0fPE/86TsJmuJUHHsBFZPOJBTRl1ZERESkJ1DVlrElBo/MXUdleZhD9+uT1XtSdeupf+I2Eh+/RWTEAVRNu4bI4FFdnFREREREckkFcUZDU/rfi780jD37t78hRxCkaHr7CWLP3wtBisoTL6P88FMJhSN5SCoiIiIiuaSCOGPF5vRuccMGtN8ukdy4ktjs6SRq51My+lCi064i0m9oPiKKiIiISBdQQZyxdEOI8tIQh42pbvPzQSpJ4xt/Jv7Sg4QipUS/fA1lB5+kbZdFREREejgVxBnb4tA3WtJmgZtY+wmxWTeTXPMxpWMnET3lCsJ9BhYgpYiIiIjkmgrijHgT9Il+tgc4SDTT8MofaHjtUUIVfag6/Z8pPeBozQqLiIiI9CIqiDO2xkMMHfzplyOx4kPqZ91CamMtZQdNoXLKJYSjfQuYUERERES6Qt4KYjMbC9wHDAQ2Ahe6++JW50SA6cBUIAB+7e6/7+psDU1J6htDjB5aSdAUJ/7C/TT+/a+E+w6i+uxfULrf4V0dQUREREQKJJzHe90B3O7uY4HbgZltnHM+MAbYH5gE3Ghmo7s62ObtCQD2bV5I3V1X0vj3v1A+4cv0vXSGimERERGRXi4vBbGZDQHGAw9nDj0MjDezwa1OPRu4y91T7r4eeBw4q6vzbdiwjUvL/8T4D34LkRKqv/kboqdcQag82tW3FhEREZECy1fLxEhgpbsnAdw9aWarMsfXtzhvFLCsxevlmXOyNn/+/E6H27DgQ44tncfakZOJ2fEE6xthfU2nr9MT1dQUxzhb0piLg8YsIiLZ6nUP1Y0bN47y8vY312gtGH8Yr760L0cfe1QXpeqeampqmDBhQqFj5JXGXBw05uw1Njbu0kSCiEhvkq8e4lpgROahuR0Pzw3PHG9pObB3i9ej2jgn50KhMBVVnSuiRURERKR3yEtB7O7rgHeAczOHzgXmZfqEW3oUuNTMwpn+4tOAx/KRUURERESKUz5XmbgcuMrMFgFXZV5jZrPNbMdSDvcDnwCLgdeBX7r7kjxmFBEREZEik7ceYndfCExs4/i0Fh8ngSvylUlEREREJJ8zxCIiIiIi3Y4KYhEREREpaiqIRURERKSo9aZ1iCMATU1Nu3yBxsbGnIXpKTTm4qAxF4ddGXOL35mRnIYREelBQkEQFDpDTtTU1BwDvFToHCIiPdTkCRMmvFzoECIihdCbZoj/DkwGVgPJAmcREekpIsAw0r9DRUSKUq+ZIRYRERER2RV6qE5EREREipoKYhEREREpaiqIRURERKSoqSAWERERkaKmglhEREREipoKYhEREREpaiqIRURERKSo9aaNOTpkZmOB+4CBwEbgQndf3OqcCDAdmAoEwK/d/ff5zporWY75euAc0huaNAPXufucfGfNlWzG3OJcA+YBM9z9R/lLmVvZjtnMvgFcD4RI/3yf6O5r85k1V7L82R4C3AuMBEqB54Gr3T2R57i7zcz+HTgDGA0c5O7z2zinV/3+EhHJl2KbIb4DuN3dxwK3AzPbOOd8YAywPzAJuNHMRuctYe5lM+Y3gSPc/WDgO8AjZlaZx4y5ls2YdxQPM4HH85itq3Q4ZjM7HLgROMndxwHHAFvzGTLHsvk+Xwd8mPnZPhiYAHw9fxFz6nHgWGBZO+f0tt9fIiJ5UTQFcWamaDzwcObQw8B4Mxvc6tSzgbvcPeXu60n/n9BZ+UuaO9mO2d3nuHss8/I90rOHA/MWNIc68X0G+CnwN2BRnuJ1iU6M+Vrg3919DYC7b3X3hvwlzZ1OjDkA+phZGCgHyoCVeQuaQ+7+srvXdnBar/n9JSKST0VTEJP+k+lKd08CZP5dlTne0ig+OwOzvI1zeopsx9zShcDH7r4iD/m6QlZjNrNDgFOA3+U9Ye5l+30+ENjXzF40s7fN7F/MLJTnrLmS7Zh/BYwFVgNrgDnu/ko+g+ZZb/r9JSKSN8VUEEsHzOw40gXEuYXO0pXMrBS4E7h8R0FVJCKk2wZOAo4DvgRcUNBEXe8s0n/1GAaMAI41szMLG0lERLqbYiqIa4ERmb7RHf2jwzPHW1oO7N3i9ag2zukpsh0zZjYJeAA4zd09rylzK5sxDwP2A2ab2VLg+8ClZnZnfqPmTGd+th9z90Z33wb8BfhCXpPmTrZjvgp4MNNCsJX0mE/Ia9L86k2/v0RE8qZoCmJ3Xwe8w6ezn+cC8zJ9di09Sro4Cmf6EU8DHstf0tzJdsxmdgTwCHCmu7+d35S5lc2Y3X25uw9y99HuPhq4mXTf5WV5D5wDnfjZfgg42cxCmVnyKcC7+UuaO50Y8xLSKy5gZmXAicA/rM7Qi/Sa318iIvlUNAVxxuXAVWa2iPTM0eUAZjY78wQ+wP3AJ8Bi4HXgl+6+pBBhcySbMc8AKoGZZvZO5r+DChM3J7IZc2+TzZj/AKwDFpAuJj8A7i5A1lzJZszfByab2fukx7wIuKsQYXeXmU03sxXAXsAzZvZB5nhv/v0lIpIXoSAICp1BRERERKRgim2GWERERETkM1QQi4iIiEhRU0EsIiIiIkVNBbGIiIiIFDUVxCIiIiJS1FQQS49mZnPN7JJC52iPmZ1vZk+18/nJZtaTN0MRERHp0bTsmnQbmV3j9gRabqc81t1XtfOeucAD7v77HOaYCxwJJIAG4EXgSndfnaPrB8D+7v5RLq7Xzn1uBH4GNJIeywLgh+7+Wpbvz0tOERGRQtMMsXQ3X3H36hb/7bQY7mLfc/dqYCzQD/hdgXLsrkcy4xgEPE96JzMRERFpoaTQAUTaY2b9Se++NZH0z+srwOXuvqKNc8eQ3nntUKAZeNbdz8587gDgVmACsB643t3/2NH93X2Tmf0JuCJznaOAW0gXyouAa9z91cznvg3cAAwGNgD/4u4PZo5f4u7HmNmLmUu/m5mBvRhYS3qWey8z+wlwhLuf2WJctwAhd7/azPYA/i8wDUgB9wI/d/eWs+ptjSNhZg8C15nZYHdfb2ZfyIzlc0Ac+BPwA3dvaiunuz9iZqcC/wqMJj3jfLm7v9fR11FERKQ70wyxdHdh0kXf3sAo0oXbbTs591fAU0B/0tvb3gpgZlXA08BDwBDgHGCGmR3Y0c3NbBBwBjB9AR/FAAAD60lEQVTPzAYAs4DpwEDSheksMxuYucd04Evu3gc4ivRWwZ/h7sdmPjwkMwP+SKtT/gBMM7M+mftHgG9ksgP8J+n2hzHAYcDJQIc91GZWBlwIbAQ2Zw4ngWtJzx5PAqYA/7SznGZ2GHAP8N3M+GcCfzWz8o7uLyIi0p1phli6m8fNLJH5eK67n0Z65hIAM7uJ9J/+29JMunAenplBfjlz/FRgqbvfm3k9LzPrexbwi51ca7qZ/TtQD8wFfgB8GVjs7vdnznnYzK4GvkK6FSEFjDOz5Zl+4073HLv7MjN7Gzgd+C/gi0DM3V83sz1Jzwz3c/c4UG9mvwMuI12ctuUbmVndPsAW4Ax3T2TuVdPivKVmNhM4Drh5J9e6DJjp7m9kXt9nZteR7rd+obNjFRER6S5UEEt3c5q7P7PjhZlFSffvTiU98wvQx8wibbQJ/Jj0LPGbZrYZ+K2730O6SJ5oZltanFtCuhVjZ65u/aCemQ0HlrU6bxkwwt3rzexs4EfA3Wb2CukH2BZmMebWHgLOJV0Qn8ens8N7A6XAajPbcW4YqG3nWn90929mZrr/RLplZG5mPGNJz3IfDkRJf01qdnKdHff/lpld1eJYGTA824GJiIh0RyqIpbv7IWDARHdfY2aHAvOAUOsT3X0NcCmAmR0DPJPpha0FXnD3k3YzyyrSRWFLo4AnM/efA8wxs0rSfbZ3AZN34T6PAr81s71IzxRPyhyvJb1ixKAds7zZcvcNZnYZ8JaZPZSZwf4P0l/Lc919m5l9HzizncvUAje5+02dHI+IiEi3poJYurs+pPuGt2R6eH++sxPN7CzgtUy7xGYgIN3G8Dfg12Z2AekeXUg/eLfd3T/sRJbZwK1mdh7wR9K9xQcCf8u0MxwJPJPJuz1z77asBfYF2lzOLPPA21zSvdNLdmR099WZ9Yx/a2bXZ+6xD7CXu3fYsuDubmZzSM+kX0v6a1sHbM88dHgF6QcOd5bzLuDPZvYM8CbpWeXjgRfdfVtH9xcREemu9FCddHc3A5WkV214ncxs7E4cAbxhZtuBv5JeAeKTTLF2MumH6VYBa4DfAJ16GMzdN5LuR/4h6YfTfgyc6u4bSP9v6QeZ628i3Yt7xU4udSPp/tstZvaNnZzzEHAin7ZL7HAh6TaFBaSL/seAYZ0Yxv8BLjOzIaTbO84DtpEudls/4PeZnO7+FukZ+Nsy9/4I+HYn7i0iItItaWMOERERESlqmiEWERERkaKmglhEREREipoKYhEREREpaiqIRURERKSoqSAWERERkaKmglhEREREipoKYhEREREpaiqIRURERKSoqSAWERERkaL2/wFpiMf1qMgfwAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.136 and 0.837 for Random Forest Classifier (top 250) on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "8Mo702Rt-V23",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "14f89e36-57b0-40be-e77c-ce2a0cc496d8"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133739</td>\n",
              "      <td>0.845052</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>LightGBM Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.134520</td>\n",
              "      <td>0.845469</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Random Forest Classifier (top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136440</td>\n",
              "      <td>0.837497</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                Model  ...       auc\n",
              "0                  Logisic Regression  ...  0.823515\n",
              "1            Decision Tree Classifier  ...  0.769599\n",
              "2            Random Forest Classifier  ...  0.836315\n",
              "3                   XGBoost Classifer  ...  0.845052\n",
              "4                  LightGBM Classifer  ...  0.845469\n",
              "5  Random Forest Classifier (top 250)  ...  0.837497\n",
              "\n",
              "[6 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "i-y4IONn-V26"
      },
      "source": [
        "# XGB Classifier (top 250)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Gjg5rhGa-V27",
        "colab": {}
      },
      "source": [
        "model = xgb.XGBClassifier(n_jobs=-1,\n",
        "                      nthread=-1,\n",
        "                      scale_pos_weight=1,\n",
        "                      learning_rate=0.01,  \n",
        "                      colsample_bytree = 0.8, \n",
        "                      subsample = 0.7,\n",
        "                      objective='binary:logistic', \n",
        "                      n_estimators=865,\n",
        "                      reg_alpha = 0.3,\n",
        "                      max_depth=5, \n",
        "                      gamma=1,\n",
        "                      random_state=42\n",
        "                     )\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "8Q0dR9Kt-V2-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "d418014d-2860-41a5-bf89-0e8659fa06be"
      },
      "source": [
        "#https://towardsdatascience.com/fine-tuning-xgboost-in-python-like-a-boss-b4543ed8b1e\n",
        "\"\"\"\n",
        "manually changed each parameters 1 by 1 and compared the validation auc\n",
        "and selected the best parameter value\n",
        "\"\"\"\n",
        "eval_metric = ['auc','error']\n",
        "%time model.fit(X_train_200,y_train,eval_set=[(X_val_200,y_val)],eval_metric=eval_metric,verbose=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0]\tvalidation_0-auc:0.802858\tvalidation_0-error:0.039726\n",
            "[1]\tvalidation_0-auc:0.809126\tvalidation_0-error:0.039639\n",
            "[2]\tvalidation_0-auc:0.813553\tvalidation_0-error:0.039551\n",
            "[3]\tvalidation_0-auc:0.815445\tvalidation_0-error:0.039551\n",
            "[4]\tvalidation_0-auc:0.820435\tvalidation_0-error:0.039551\n",
            "[5]\tvalidation_0-auc:0.824176\tvalidation_0-error:0.039551\n",
            "[6]\tvalidation_0-auc:0.82459\tvalidation_0-error:0.039551\n",
            "[7]\tvalidation_0-auc:0.824298\tvalidation_0-error:0.039551\n",
            "[8]\tvalidation_0-auc:0.824797\tvalidation_0-error:0.039551\n",
            "[9]\tvalidation_0-auc:0.825743\tvalidation_0-error:0.039551\n",
            "[10]\tvalidation_0-auc:0.826107\tvalidation_0-error:0.039551\n",
            "[11]\tvalidation_0-auc:0.826179\tvalidation_0-error:0.039551\n",
            "[12]\tvalidation_0-auc:0.826059\tvalidation_0-error:0.039551\n",
            "[13]\tvalidation_0-auc:0.826039\tvalidation_0-error:0.039551\n",
            "[14]\tvalidation_0-auc:0.82635\tvalidation_0-error:0.039551\n",
            "[15]\tvalidation_0-auc:0.826357\tvalidation_0-error:0.039551\n",
            "[16]\tvalidation_0-auc:0.826251\tvalidation_0-error:0.039551\n",
            "[17]\tvalidation_0-auc:0.825965\tvalidation_0-error:0.039551\n",
            "[18]\tvalidation_0-auc:0.82593\tvalidation_0-error:0.039551\n",
            "[19]\tvalidation_0-auc:0.826123\tvalidation_0-error:0.039551\n",
            "[20]\tvalidation_0-auc:0.825758\tvalidation_0-error:0.039551\n",
            "[21]\tvalidation_0-auc:0.82563\tvalidation_0-error:0.039551\n",
            "[22]\tvalidation_0-auc:0.825869\tvalidation_0-error:0.039551\n",
            "[23]\tvalidation_0-auc:0.82577\tvalidation_0-error:0.039551\n",
            "[24]\tvalidation_0-auc:0.825901\tvalidation_0-error:0.039551\n",
            "[25]\tvalidation_0-auc:0.825894\tvalidation_0-error:0.039551\n",
            "[26]\tvalidation_0-auc:0.826051\tvalidation_0-error:0.039551\n",
            "[27]\tvalidation_0-auc:0.825848\tvalidation_0-error:0.039551\n",
            "[28]\tvalidation_0-auc:0.825758\tvalidation_0-error:0.039551\n",
            "[29]\tvalidation_0-auc:0.825779\tvalidation_0-error:0.039551\n",
            "[30]\tvalidation_0-auc:0.826117\tvalidation_0-error:0.039551\n",
            "[31]\tvalidation_0-auc:0.825971\tvalidation_0-error:0.039551\n",
            "[32]\tvalidation_0-auc:0.826163\tvalidation_0-error:0.039551\n",
            "[33]\tvalidation_0-auc:0.826309\tvalidation_0-error:0.039551\n",
            "[34]\tvalidation_0-auc:0.825719\tvalidation_0-error:0.039551\n",
            "[35]\tvalidation_0-auc:0.825957\tvalidation_0-error:0.039551\n",
            "[36]\tvalidation_0-auc:0.825916\tvalidation_0-error:0.039551\n",
            "[37]\tvalidation_0-auc:0.826349\tvalidation_0-error:0.039551\n",
            "[38]\tvalidation_0-auc:0.826348\tvalidation_0-error:0.039551\n",
            "[39]\tvalidation_0-auc:0.8267\tvalidation_0-error:0.039551\n",
            "[40]\tvalidation_0-auc:0.826704\tvalidation_0-error:0.039551\n",
            "[41]\tvalidation_0-auc:0.826638\tvalidation_0-error:0.039551\n",
            "[42]\tvalidation_0-auc:0.826647\tvalidation_0-error:0.039551\n",
            "[43]\tvalidation_0-auc:0.826956\tvalidation_0-error:0.039551\n",
            "[44]\tvalidation_0-auc:0.826995\tvalidation_0-error:0.039551\n",
            "[45]\tvalidation_0-auc:0.827123\tvalidation_0-error:0.039551\n",
            "[46]\tvalidation_0-auc:0.827167\tvalidation_0-error:0.039551\n",
            "[47]\tvalidation_0-auc:0.827198\tvalidation_0-error:0.039551\n",
            "[48]\tvalidation_0-auc:0.82778\tvalidation_0-error:0.039551\n",
            "[49]\tvalidation_0-auc:0.828004\tvalidation_0-error:0.039551\n",
            "[50]\tvalidation_0-auc:0.827978\tvalidation_0-error:0.039551\n",
            "[51]\tvalidation_0-auc:0.828172\tvalidation_0-error:0.039551\n",
            "[52]\tvalidation_0-auc:0.828167\tvalidation_0-error:0.039551\n",
            "[53]\tvalidation_0-auc:0.828357\tvalidation_0-error:0.039551\n",
            "[54]\tvalidation_0-auc:0.828378\tvalidation_0-error:0.039551\n",
            "[55]\tvalidation_0-auc:0.828747\tvalidation_0-error:0.039551\n",
            "[56]\tvalidation_0-auc:0.828797\tvalidation_0-error:0.039551\n",
            "[57]\tvalidation_0-auc:0.828846\tvalidation_0-error:0.039551\n",
            "[58]\tvalidation_0-auc:0.828751\tvalidation_0-error:0.039551\n",
            "[59]\tvalidation_0-auc:0.828681\tvalidation_0-error:0.039551\n",
            "[60]\tvalidation_0-auc:0.829071\tvalidation_0-error:0.039551\n",
            "[61]\tvalidation_0-auc:0.829214\tvalidation_0-error:0.039551\n",
            "[62]\tvalidation_0-auc:0.82902\tvalidation_0-error:0.039551\n",
            "[63]\tvalidation_0-auc:0.828941\tvalidation_0-error:0.039551\n",
            "[64]\tvalidation_0-auc:0.82883\tvalidation_0-error:0.039551\n",
            "[65]\tvalidation_0-auc:0.828919\tvalidation_0-error:0.039551\n",
            "[66]\tvalidation_0-auc:0.82892\tvalidation_0-error:0.039551\n",
            "[67]\tvalidation_0-auc:0.829186\tvalidation_0-error:0.039551\n",
            "[68]\tvalidation_0-auc:0.829421\tvalidation_0-error:0.039551\n",
            "[69]\tvalidation_0-auc:0.829494\tvalidation_0-error:0.039551\n",
            "[70]\tvalidation_0-auc:0.829501\tvalidation_0-error:0.039551\n",
            "[71]\tvalidation_0-auc:0.829524\tvalidation_0-error:0.039551\n",
            "[72]\tvalidation_0-auc:0.82951\tvalidation_0-error:0.039551\n",
            "[73]\tvalidation_0-auc:0.829489\tvalidation_0-error:0.039551\n",
            "[74]\tvalidation_0-auc:0.829356\tvalidation_0-error:0.039551\n",
            "[75]\tvalidation_0-auc:0.829469\tvalidation_0-error:0.039551\n",
            "[76]\tvalidation_0-auc:0.82954\tvalidation_0-error:0.039551\n",
            "[77]\tvalidation_0-auc:0.82961\tvalidation_0-error:0.039551\n",
            "[78]\tvalidation_0-auc:0.829416\tvalidation_0-error:0.039551\n",
            "[79]\tvalidation_0-auc:0.829195\tvalidation_0-error:0.039551\n",
            "[80]\tvalidation_0-auc:0.82927\tvalidation_0-error:0.039551\n",
            "[81]\tvalidation_0-auc:0.82923\tvalidation_0-error:0.039551\n",
            "[82]\tvalidation_0-auc:0.829425\tvalidation_0-error:0.039551\n",
            "[83]\tvalidation_0-auc:0.829722\tvalidation_0-error:0.039551\n",
            "[84]\tvalidation_0-auc:0.829691\tvalidation_0-error:0.039551\n",
            "[85]\tvalidation_0-auc:0.830007\tvalidation_0-error:0.039551\n",
            "[86]\tvalidation_0-auc:0.830016\tvalidation_0-error:0.039551\n",
            "[87]\tvalidation_0-auc:0.829873\tvalidation_0-error:0.039551\n",
            "[88]\tvalidation_0-auc:0.830179\tvalidation_0-error:0.039551\n",
            "[89]\tvalidation_0-auc:0.830307\tvalidation_0-error:0.039551\n",
            "[90]\tvalidation_0-auc:0.83028\tvalidation_0-error:0.039551\n",
            "[91]\tvalidation_0-auc:0.830565\tvalidation_0-error:0.039551\n",
            "[92]\tvalidation_0-auc:0.830712\tvalidation_0-error:0.039551\n",
            "[93]\tvalidation_0-auc:0.830748\tvalidation_0-error:0.039551\n",
            "[94]\tvalidation_0-auc:0.830865\tvalidation_0-error:0.039551\n",
            "[95]\tvalidation_0-auc:0.830795\tvalidation_0-error:0.039551\n",
            "[96]\tvalidation_0-auc:0.830874\tvalidation_0-error:0.039551\n",
            "[97]\tvalidation_0-auc:0.830949\tvalidation_0-error:0.039551\n",
            "[98]\tvalidation_0-auc:0.830977\tvalidation_0-error:0.039551\n",
            "[99]\tvalidation_0-auc:0.831144\tvalidation_0-error:0.039551\n",
            "[100]\tvalidation_0-auc:0.831204\tvalidation_0-error:0.039551\n",
            "[101]\tvalidation_0-auc:0.831119\tvalidation_0-error:0.039551\n",
            "[102]\tvalidation_0-auc:0.831329\tvalidation_0-error:0.039551\n",
            "[103]\tvalidation_0-auc:0.831388\tvalidation_0-error:0.039551\n",
            "[104]\tvalidation_0-auc:0.831426\tvalidation_0-error:0.039551\n",
            "[105]\tvalidation_0-auc:0.831448\tvalidation_0-error:0.039551\n",
            "[106]\tvalidation_0-auc:0.83151\tvalidation_0-error:0.039551\n",
            "[107]\tvalidation_0-auc:0.831551\tvalidation_0-error:0.039551\n",
            "[108]\tvalidation_0-auc:0.831545\tvalidation_0-error:0.039551\n",
            "[109]\tvalidation_0-auc:0.83159\tvalidation_0-error:0.039551\n",
            "[110]\tvalidation_0-auc:0.83151\tvalidation_0-error:0.039551\n",
            "[111]\tvalidation_0-auc:0.831449\tvalidation_0-error:0.039551\n",
            "[112]\tvalidation_0-auc:0.831491\tvalidation_0-error:0.039551\n",
            "[113]\tvalidation_0-auc:0.831524\tvalidation_0-error:0.039551\n",
            "[114]\tvalidation_0-auc:0.831553\tvalidation_0-error:0.039551\n",
            "[115]\tvalidation_0-auc:0.831628\tvalidation_0-error:0.039551\n",
            "[116]\tvalidation_0-auc:0.831722\tvalidation_0-error:0.039551\n",
            "[117]\tvalidation_0-auc:0.831887\tvalidation_0-error:0.039551\n",
            "[118]\tvalidation_0-auc:0.831851\tvalidation_0-error:0.039551\n",
            "[119]\tvalidation_0-auc:0.831852\tvalidation_0-error:0.039551\n",
            "[120]\tvalidation_0-auc:0.832011\tvalidation_0-error:0.039551\n",
            "[121]\tvalidation_0-auc:0.832072\tvalidation_0-error:0.039551\n",
            "[122]\tvalidation_0-auc:0.83199\tvalidation_0-error:0.039551\n",
            "[123]\tvalidation_0-auc:0.831988\tvalidation_0-error:0.039551\n",
            "[124]\tvalidation_0-auc:0.83194\tvalidation_0-error:0.039551\n",
            "[125]\tvalidation_0-auc:0.832008\tvalidation_0-error:0.039551\n",
            "[126]\tvalidation_0-auc:0.832137\tvalidation_0-error:0.039551\n",
            "[127]\tvalidation_0-auc:0.832147\tvalidation_0-error:0.039551\n",
            "[128]\tvalidation_0-auc:0.832172\tvalidation_0-error:0.039551\n",
            "[129]\tvalidation_0-auc:0.832084\tvalidation_0-error:0.039551\n",
            "[130]\tvalidation_0-auc:0.832401\tvalidation_0-error:0.039551\n",
            "[131]\tvalidation_0-auc:0.832554\tvalidation_0-error:0.039551\n",
            "[132]\tvalidation_0-auc:0.83261\tvalidation_0-error:0.039551\n",
            "[133]\tvalidation_0-auc:0.832669\tvalidation_0-error:0.039551\n",
            "[134]\tvalidation_0-auc:0.832806\tvalidation_0-error:0.039551\n",
            "[135]\tvalidation_0-auc:0.832737\tvalidation_0-error:0.039551\n",
            "[136]\tvalidation_0-auc:0.83275\tvalidation_0-error:0.039551\n",
            "[137]\tvalidation_0-auc:0.832836\tvalidation_0-error:0.039551\n",
            "[138]\tvalidation_0-auc:0.832753\tvalidation_0-error:0.039551\n",
            "[139]\tvalidation_0-auc:0.832778\tvalidation_0-error:0.039551\n",
            "[140]\tvalidation_0-auc:0.832725\tvalidation_0-error:0.039551\n",
            "[141]\tvalidation_0-auc:0.832748\tvalidation_0-error:0.039551\n",
            "[142]\tvalidation_0-auc:0.832814\tvalidation_0-error:0.039551\n",
            "[143]\tvalidation_0-auc:0.832834\tvalidation_0-error:0.039551\n",
            "[144]\tvalidation_0-auc:0.832819\tvalidation_0-error:0.039551\n",
            "[145]\tvalidation_0-auc:0.832885\tvalidation_0-error:0.039551\n",
            "[146]\tvalidation_0-auc:0.832904\tvalidation_0-error:0.039551\n",
            "[147]\tvalidation_0-auc:0.8332\tvalidation_0-error:0.039551\n",
            "[148]\tvalidation_0-auc:0.833255\tvalidation_0-error:0.039551\n",
            "[149]\tvalidation_0-auc:0.833633\tvalidation_0-error:0.039551\n",
            "[150]\tvalidation_0-auc:0.83364\tvalidation_0-error:0.039551\n",
            "[151]\tvalidation_0-auc:0.833674\tvalidation_0-error:0.039551\n",
            "[152]\tvalidation_0-auc:0.833883\tvalidation_0-error:0.039551\n",
            "[153]\tvalidation_0-auc:0.833913\tvalidation_0-error:0.039551\n",
            "[154]\tvalidation_0-auc:0.834026\tvalidation_0-error:0.039551\n",
            "[155]\tvalidation_0-auc:0.834046\tvalidation_0-error:0.039551\n",
            "[156]\tvalidation_0-auc:0.834018\tvalidation_0-error:0.039551\n",
            "[157]\tvalidation_0-auc:0.83385\tvalidation_0-error:0.039551\n",
            "[158]\tvalidation_0-auc:0.833763\tvalidation_0-error:0.039551\n",
            "[159]\tvalidation_0-auc:0.833585\tvalidation_0-error:0.039551\n",
            "[160]\tvalidation_0-auc:0.833597\tvalidation_0-error:0.039551\n",
            "[161]\tvalidation_0-auc:0.83363\tvalidation_0-error:0.039551\n",
            "[162]\tvalidation_0-auc:0.83369\tvalidation_0-error:0.039551\n",
            "[163]\tvalidation_0-auc:0.833811\tvalidation_0-error:0.039551\n",
            "[164]\tvalidation_0-auc:0.833965\tvalidation_0-error:0.039551\n",
            "[165]\tvalidation_0-auc:0.833947\tvalidation_0-error:0.039551\n",
            "[166]\tvalidation_0-auc:0.834051\tvalidation_0-error:0.039551\n",
            "[167]\tvalidation_0-auc:0.834039\tvalidation_0-error:0.039551\n",
            "[168]\tvalidation_0-auc:0.834071\tvalidation_0-error:0.039551\n",
            "[169]\tvalidation_0-auc:0.834197\tvalidation_0-error:0.039551\n",
            "[170]\tvalidation_0-auc:0.834165\tvalidation_0-error:0.039551\n",
            "[171]\tvalidation_0-auc:0.834115\tvalidation_0-error:0.039551\n",
            "[172]\tvalidation_0-auc:0.834195\tvalidation_0-error:0.039551\n",
            "[173]\tvalidation_0-auc:0.83414\tvalidation_0-error:0.039551\n",
            "[174]\tvalidation_0-auc:0.834249\tvalidation_0-error:0.039551\n",
            "[175]\tvalidation_0-auc:0.834225\tvalidation_0-error:0.039551\n",
            "[176]\tvalidation_0-auc:0.834296\tvalidation_0-error:0.039551\n",
            "[177]\tvalidation_0-auc:0.834306\tvalidation_0-error:0.039551\n",
            "[178]\tvalidation_0-auc:0.83444\tvalidation_0-error:0.039551\n",
            "[179]\tvalidation_0-auc:0.834567\tvalidation_0-error:0.039551\n",
            "[180]\tvalidation_0-auc:0.834677\tvalidation_0-error:0.039551\n",
            "[181]\tvalidation_0-auc:0.83461\tvalidation_0-error:0.039551\n",
            "[182]\tvalidation_0-auc:0.83454\tvalidation_0-error:0.039551\n",
            "[183]\tvalidation_0-auc:0.834495\tvalidation_0-error:0.039551\n",
            "[184]\tvalidation_0-auc:0.834438\tvalidation_0-error:0.039551\n",
            "[185]\tvalidation_0-auc:0.834468\tvalidation_0-error:0.039551\n",
            "[186]\tvalidation_0-auc:0.834554\tvalidation_0-error:0.039551\n",
            "[187]\tvalidation_0-auc:0.834827\tvalidation_0-error:0.039551\n",
            "[188]\tvalidation_0-auc:0.834826\tvalidation_0-error:0.039551\n",
            "[189]\tvalidation_0-auc:0.834908\tvalidation_0-error:0.039551\n",
            "[190]\tvalidation_0-auc:0.835064\tvalidation_0-error:0.039551\n",
            "[191]\tvalidation_0-auc:0.835067\tvalidation_0-error:0.039551\n",
            "[192]\tvalidation_0-auc:0.834938\tvalidation_0-error:0.039551\n",
            "[193]\tvalidation_0-auc:0.834981\tvalidation_0-error:0.039551\n",
            "[194]\tvalidation_0-auc:0.835267\tvalidation_0-error:0.039551\n",
            "[195]\tvalidation_0-auc:0.835384\tvalidation_0-error:0.039551\n",
            "[196]\tvalidation_0-auc:0.835718\tvalidation_0-error:0.039551\n",
            "[197]\tvalidation_0-auc:0.835727\tvalidation_0-error:0.039551\n",
            "[198]\tvalidation_0-auc:0.835593\tvalidation_0-error:0.039551\n",
            "[199]\tvalidation_0-auc:0.835654\tvalidation_0-error:0.039551\n",
            "[200]\tvalidation_0-auc:0.835774\tvalidation_0-error:0.039551\n",
            "[201]\tvalidation_0-auc:0.835836\tvalidation_0-error:0.039551\n",
            "[202]\tvalidation_0-auc:0.835949\tvalidation_0-error:0.039551\n",
            "[203]\tvalidation_0-auc:0.835906\tvalidation_0-error:0.039551\n",
            "[204]\tvalidation_0-auc:0.836162\tvalidation_0-error:0.039551\n",
            "[205]\tvalidation_0-auc:0.836384\tvalidation_0-error:0.039551\n",
            "[206]\tvalidation_0-auc:0.836367\tvalidation_0-error:0.039551\n",
            "[207]\tvalidation_0-auc:0.836354\tvalidation_0-error:0.039551\n",
            "[208]\tvalidation_0-auc:0.836444\tvalidation_0-error:0.039551\n",
            "[209]\tvalidation_0-auc:0.836479\tvalidation_0-error:0.039551\n",
            "[210]\tvalidation_0-auc:0.836517\tvalidation_0-error:0.039551\n",
            "[211]\tvalidation_0-auc:0.836575\tvalidation_0-error:0.039551\n",
            "[212]\tvalidation_0-auc:0.836631\tvalidation_0-error:0.039551\n",
            "[213]\tvalidation_0-auc:0.83668\tvalidation_0-error:0.039551\n",
            "[214]\tvalidation_0-auc:0.836666\tvalidation_0-error:0.039551\n",
            "[215]\tvalidation_0-auc:0.836824\tvalidation_0-error:0.039551\n",
            "[216]\tvalidation_0-auc:0.836633\tvalidation_0-error:0.039551\n",
            "[217]\tvalidation_0-auc:0.836596\tvalidation_0-error:0.039551\n",
            "[218]\tvalidation_0-auc:0.83657\tvalidation_0-error:0.039551\n",
            "[219]\tvalidation_0-auc:0.836715\tvalidation_0-error:0.039551\n",
            "[220]\tvalidation_0-auc:0.836778\tvalidation_0-error:0.039551\n",
            "[221]\tvalidation_0-auc:0.836903\tvalidation_0-error:0.039551\n",
            "[222]\tvalidation_0-auc:0.837082\tvalidation_0-error:0.039551\n",
            "[223]\tvalidation_0-auc:0.837045\tvalidation_0-error:0.039551\n",
            "[224]\tvalidation_0-auc:0.837111\tvalidation_0-error:0.039551\n",
            "[225]\tvalidation_0-auc:0.837147\tvalidation_0-error:0.039551\n",
            "[226]\tvalidation_0-auc:0.83717\tvalidation_0-error:0.039551\n",
            "[227]\tvalidation_0-auc:0.837069\tvalidation_0-error:0.039551\n",
            "[228]\tvalidation_0-auc:0.837082\tvalidation_0-error:0.039551\n",
            "[229]\tvalidation_0-auc:0.837017\tvalidation_0-error:0.039551\n",
            "[230]\tvalidation_0-auc:0.837196\tvalidation_0-error:0.039551\n",
            "[231]\tvalidation_0-auc:0.837115\tvalidation_0-error:0.039551\n",
            "[232]\tvalidation_0-auc:0.837016\tvalidation_0-error:0.039551\n",
            "[233]\tvalidation_0-auc:0.837008\tvalidation_0-error:0.039551\n",
            "[234]\tvalidation_0-auc:0.837254\tvalidation_0-error:0.039551\n",
            "[235]\tvalidation_0-auc:0.837163\tvalidation_0-error:0.039551\n",
            "[236]\tvalidation_0-auc:0.837178\tvalidation_0-error:0.039551\n",
            "[237]\tvalidation_0-auc:0.837298\tvalidation_0-error:0.039551\n",
            "[238]\tvalidation_0-auc:0.837308\tvalidation_0-error:0.039551\n",
            "[239]\tvalidation_0-auc:0.83723\tvalidation_0-error:0.039551\n",
            "[240]\tvalidation_0-auc:0.837311\tvalidation_0-error:0.039551\n",
            "[241]\tvalidation_0-auc:0.837246\tvalidation_0-error:0.039551\n",
            "[242]\tvalidation_0-auc:0.837603\tvalidation_0-error:0.039551\n",
            "[243]\tvalidation_0-auc:0.837547\tvalidation_0-error:0.039551\n",
            "[244]\tvalidation_0-auc:0.837532\tvalidation_0-error:0.039551\n",
            "[245]\tvalidation_0-auc:0.837478\tvalidation_0-error:0.039551\n",
            "[246]\tvalidation_0-auc:0.837578\tvalidation_0-error:0.039551\n",
            "[247]\tvalidation_0-auc:0.837547\tvalidation_0-error:0.039551\n",
            "[248]\tvalidation_0-auc:0.837454\tvalidation_0-error:0.039551\n",
            "[249]\tvalidation_0-auc:0.837416\tvalidation_0-error:0.039551\n",
            "[250]\tvalidation_0-auc:0.837527\tvalidation_0-error:0.039551\n",
            "[251]\tvalidation_0-auc:0.837763\tvalidation_0-error:0.039551\n",
            "[252]\tvalidation_0-auc:0.837651\tvalidation_0-error:0.039551\n",
            "[253]\tvalidation_0-auc:0.83785\tvalidation_0-error:0.039551\n",
            "[254]\tvalidation_0-auc:0.837729\tvalidation_0-error:0.039551\n",
            "[255]\tvalidation_0-auc:0.83781\tvalidation_0-error:0.039551\n",
            "[256]\tvalidation_0-auc:0.837818\tvalidation_0-error:0.039551\n",
            "[257]\tvalidation_0-auc:0.837806\tvalidation_0-error:0.039551\n",
            "[258]\tvalidation_0-auc:0.837992\tvalidation_0-error:0.039551\n",
            "[259]\tvalidation_0-auc:0.838168\tvalidation_0-error:0.039551\n",
            "[260]\tvalidation_0-auc:0.838176\tvalidation_0-error:0.039551\n",
            "[261]\tvalidation_0-auc:0.838203\tvalidation_0-error:0.039551\n",
            "[262]\tvalidation_0-auc:0.838221\tvalidation_0-error:0.039551\n",
            "[263]\tvalidation_0-auc:0.838346\tvalidation_0-error:0.039551\n",
            "[264]\tvalidation_0-auc:0.838352\tvalidation_0-error:0.039551\n",
            "[265]\tvalidation_0-auc:0.838444\tvalidation_0-error:0.039551\n",
            "[266]\tvalidation_0-auc:0.838582\tvalidation_0-error:0.039551\n",
            "[267]\tvalidation_0-auc:0.838643\tvalidation_0-error:0.039551\n",
            "[268]\tvalidation_0-auc:0.838638\tvalidation_0-error:0.039551\n",
            "[269]\tvalidation_0-auc:0.838753\tvalidation_0-error:0.039551\n",
            "[270]\tvalidation_0-auc:0.83878\tvalidation_0-error:0.039551\n",
            "[271]\tvalidation_0-auc:0.838756\tvalidation_0-error:0.039551\n",
            "[272]\tvalidation_0-auc:0.838755\tvalidation_0-error:0.039551\n",
            "[273]\tvalidation_0-auc:0.838703\tvalidation_0-error:0.039551\n",
            "[274]\tvalidation_0-auc:0.838906\tvalidation_0-error:0.039551\n",
            "[275]\tvalidation_0-auc:0.838959\tvalidation_0-error:0.039551\n",
            "[276]\tvalidation_0-auc:0.839034\tvalidation_0-error:0.039551\n",
            "[277]\tvalidation_0-auc:0.839074\tvalidation_0-error:0.039551\n",
            "[278]\tvalidation_0-auc:0.839144\tvalidation_0-error:0.039551\n",
            "[279]\tvalidation_0-auc:0.839043\tvalidation_0-error:0.039551\n",
            "[280]\tvalidation_0-auc:0.839083\tvalidation_0-error:0.039551\n",
            "[281]\tvalidation_0-auc:0.839228\tvalidation_0-error:0.039551\n",
            "[282]\tvalidation_0-auc:0.839334\tvalidation_0-error:0.039551\n",
            "[283]\tvalidation_0-auc:0.839374\tvalidation_0-error:0.039551\n",
            "[284]\tvalidation_0-auc:0.839773\tvalidation_0-error:0.039551\n",
            "[285]\tvalidation_0-auc:0.83982\tvalidation_0-error:0.039551\n",
            "[286]\tvalidation_0-auc:0.839879\tvalidation_0-error:0.039551\n",
            "[287]\tvalidation_0-auc:0.83976\tvalidation_0-error:0.039551\n",
            "[288]\tvalidation_0-auc:0.839828\tvalidation_0-error:0.039551\n",
            "[289]\tvalidation_0-auc:0.839939\tvalidation_0-error:0.039551\n",
            "[290]\tvalidation_0-auc:0.839968\tvalidation_0-error:0.039551\n",
            "[291]\tvalidation_0-auc:0.840092\tvalidation_0-error:0.039551\n",
            "[292]\tvalidation_0-auc:0.840181\tvalidation_0-error:0.039551\n",
            "[293]\tvalidation_0-auc:0.840283\tvalidation_0-error:0.039551\n",
            "[294]\tvalidation_0-auc:0.840371\tvalidation_0-error:0.039551\n",
            "[295]\tvalidation_0-auc:0.840328\tvalidation_0-error:0.039551\n",
            "[296]\tvalidation_0-auc:0.840337\tvalidation_0-error:0.039551\n",
            "[297]\tvalidation_0-auc:0.840352\tvalidation_0-error:0.039551\n",
            "[298]\tvalidation_0-auc:0.840415\tvalidation_0-error:0.039551\n",
            "[299]\tvalidation_0-auc:0.840416\tvalidation_0-error:0.039551\n",
            "[300]\tvalidation_0-auc:0.840475\tvalidation_0-error:0.039551\n",
            "[301]\tvalidation_0-auc:0.840527\tvalidation_0-error:0.039551\n",
            "[302]\tvalidation_0-auc:0.840548\tvalidation_0-error:0.039551\n",
            "[303]\tvalidation_0-auc:0.84059\tvalidation_0-error:0.039551\n",
            "[304]\tvalidation_0-auc:0.840586\tvalidation_0-error:0.039551\n",
            "[305]\tvalidation_0-auc:0.840649\tvalidation_0-error:0.039551\n",
            "[306]\tvalidation_0-auc:0.840624\tvalidation_0-error:0.039551\n",
            "[307]\tvalidation_0-auc:0.840566\tvalidation_0-error:0.039551\n",
            "[308]\tvalidation_0-auc:0.840485\tvalidation_0-error:0.039551\n",
            "[309]\tvalidation_0-auc:0.840328\tvalidation_0-error:0.039551\n",
            "[310]\tvalidation_0-auc:0.84022\tvalidation_0-error:0.039551\n",
            "[311]\tvalidation_0-auc:0.840308\tvalidation_0-error:0.039551\n",
            "[312]\tvalidation_0-auc:0.840271\tvalidation_0-error:0.039551\n",
            "[313]\tvalidation_0-auc:0.840261\tvalidation_0-error:0.039551\n",
            "[314]\tvalidation_0-auc:0.840388\tvalidation_0-error:0.039551\n",
            "[315]\tvalidation_0-auc:0.840371\tvalidation_0-error:0.039551\n",
            "[316]\tvalidation_0-auc:0.840444\tvalidation_0-error:0.039551\n",
            "[317]\tvalidation_0-auc:0.840426\tvalidation_0-error:0.039551\n",
            "[318]\tvalidation_0-auc:0.840472\tvalidation_0-error:0.039551\n",
            "[319]\tvalidation_0-auc:0.840533\tvalidation_0-error:0.039551\n",
            "[320]\tvalidation_0-auc:0.840558\tvalidation_0-error:0.039551\n",
            "[321]\tvalidation_0-auc:0.840585\tvalidation_0-error:0.039551\n",
            "[322]\tvalidation_0-auc:0.84065\tvalidation_0-error:0.039551\n",
            "[323]\tvalidation_0-auc:0.84062\tvalidation_0-error:0.039551\n",
            "[324]\tvalidation_0-auc:0.840628\tvalidation_0-error:0.039551\n",
            "[325]\tvalidation_0-auc:0.840691\tvalidation_0-error:0.039551\n",
            "[326]\tvalidation_0-auc:0.840706\tvalidation_0-error:0.039551\n",
            "[327]\tvalidation_0-auc:0.840746\tvalidation_0-error:0.039551\n",
            "[328]\tvalidation_0-auc:0.840778\tvalidation_0-error:0.039551\n",
            "[329]\tvalidation_0-auc:0.840875\tvalidation_0-error:0.039551\n",
            "[330]\tvalidation_0-auc:0.840929\tvalidation_0-error:0.039551\n",
            "[331]\tvalidation_0-auc:0.840947\tvalidation_0-error:0.039551\n",
            "[332]\tvalidation_0-auc:0.840906\tvalidation_0-error:0.039551\n",
            "[333]\tvalidation_0-auc:0.840995\tvalidation_0-error:0.039551\n",
            "[334]\tvalidation_0-auc:0.841059\tvalidation_0-error:0.039551\n",
            "[335]\tvalidation_0-auc:0.841136\tvalidation_0-error:0.039551\n",
            "[336]\tvalidation_0-auc:0.841258\tvalidation_0-error:0.039551\n",
            "[337]\tvalidation_0-auc:0.841265\tvalidation_0-error:0.039551\n",
            "[338]\tvalidation_0-auc:0.841394\tvalidation_0-error:0.039551\n",
            "[339]\tvalidation_0-auc:0.841531\tvalidation_0-error:0.039551\n",
            "[340]\tvalidation_0-auc:0.841376\tvalidation_0-error:0.039551\n",
            "[341]\tvalidation_0-auc:0.841458\tvalidation_0-error:0.039551\n",
            "[342]\tvalidation_0-auc:0.84154\tvalidation_0-error:0.039551\n",
            "[343]\tvalidation_0-auc:0.841509\tvalidation_0-error:0.039551\n",
            "[344]\tvalidation_0-auc:0.841412\tvalidation_0-error:0.039551\n",
            "[345]\tvalidation_0-auc:0.841383\tvalidation_0-error:0.039551\n",
            "[346]\tvalidation_0-auc:0.841496\tvalidation_0-error:0.039551\n",
            "[347]\tvalidation_0-auc:0.841538\tvalidation_0-error:0.039551\n",
            "[348]\tvalidation_0-auc:0.841589\tvalidation_0-error:0.039551\n",
            "[349]\tvalidation_0-auc:0.841582\tvalidation_0-error:0.039551\n",
            "[350]\tvalidation_0-auc:0.841627\tvalidation_0-error:0.039551\n",
            "[351]\tvalidation_0-auc:0.841563\tvalidation_0-error:0.039551\n",
            "[352]\tvalidation_0-auc:0.841603\tvalidation_0-error:0.039551\n",
            "[353]\tvalidation_0-auc:0.841633\tvalidation_0-error:0.039551\n",
            "[354]\tvalidation_0-auc:0.841687\tvalidation_0-error:0.039551\n",
            "[355]\tvalidation_0-auc:0.841713\tvalidation_0-error:0.039551\n",
            "[356]\tvalidation_0-auc:0.841737\tvalidation_0-error:0.039551\n",
            "[357]\tvalidation_0-auc:0.841747\tvalidation_0-error:0.039551\n",
            "[358]\tvalidation_0-auc:0.841969\tvalidation_0-error:0.039551\n",
            "[359]\tvalidation_0-auc:0.841982\tvalidation_0-error:0.039551\n",
            "[360]\tvalidation_0-auc:0.841932\tvalidation_0-error:0.039551\n",
            "[361]\tvalidation_0-auc:0.841897\tvalidation_0-error:0.039551\n",
            "[362]\tvalidation_0-auc:0.841918\tvalidation_0-error:0.039551\n",
            "[363]\tvalidation_0-auc:0.841941\tvalidation_0-error:0.039551\n",
            "[364]\tvalidation_0-auc:0.842013\tvalidation_0-error:0.039551\n",
            "[365]\tvalidation_0-auc:0.84202\tvalidation_0-error:0.039551\n",
            "[366]\tvalidation_0-auc:0.842013\tvalidation_0-error:0.039551\n",
            "[367]\tvalidation_0-auc:0.842041\tvalidation_0-error:0.039551\n",
            "[368]\tvalidation_0-auc:0.842036\tvalidation_0-error:0.039551\n",
            "[369]\tvalidation_0-auc:0.842018\tvalidation_0-error:0.039551\n",
            "[370]\tvalidation_0-auc:0.842066\tvalidation_0-error:0.039551\n",
            "[371]\tvalidation_0-auc:0.842068\tvalidation_0-error:0.039551\n",
            "[372]\tvalidation_0-auc:0.842038\tvalidation_0-error:0.039551\n",
            "[373]\tvalidation_0-auc:0.842001\tvalidation_0-error:0.039551\n",
            "[374]\tvalidation_0-auc:0.841977\tvalidation_0-error:0.039551\n",
            "[375]\tvalidation_0-auc:0.842031\tvalidation_0-error:0.039551\n",
            "[376]\tvalidation_0-auc:0.842022\tvalidation_0-error:0.039551\n",
            "[377]\tvalidation_0-auc:0.842076\tvalidation_0-error:0.039551\n",
            "[378]\tvalidation_0-auc:0.842051\tvalidation_0-error:0.039551\n",
            "[379]\tvalidation_0-auc:0.841943\tvalidation_0-error:0.039551\n",
            "[380]\tvalidation_0-auc:0.841969\tvalidation_0-error:0.039551\n",
            "[381]\tvalidation_0-auc:0.842122\tvalidation_0-error:0.039551\n",
            "[382]\tvalidation_0-auc:0.842108\tvalidation_0-error:0.039551\n",
            "[383]\tvalidation_0-auc:0.842125\tvalidation_0-error:0.039551\n",
            "[384]\tvalidation_0-auc:0.842161\tvalidation_0-error:0.039551\n",
            "[385]\tvalidation_0-auc:0.842232\tvalidation_0-error:0.039551\n",
            "[386]\tvalidation_0-auc:0.842216\tvalidation_0-error:0.039551\n",
            "[387]\tvalidation_0-auc:0.842249\tvalidation_0-error:0.039551\n",
            "[388]\tvalidation_0-auc:0.842116\tvalidation_0-error:0.039551\n",
            "[389]\tvalidation_0-auc:0.841928\tvalidation_0-error:0.039551\n",
            "[390]\tvalidation_0-auc:0.841955\tvalidation_0-error:0.039551\n",
            "[391]\tvalidation_0-auc:0.841982\tvalidation_0-error:0.039551\n",
            "[392]\tvalidation_0-auc:0.841972\tvalidation_0-error:0.039551\n",
            "[393]\tvalidation_0-auc:0.841968\tvalidation_0-error:0.039551\n",
            "[394]\tvalidation_0-auc:0.842046\tvalidation_0-error:0.039551\n",
            "[395]\tvalidation_0-auc:0.842068\tvalidation_0-error:0.039551\n",
            "[396]\tvalidation_0-auc:0.842129\tvalidation_0-error:0.039551\n",
            "[397]\tvalidation_0-auc:0.842128\tvalidation_0-error:0.039551\n",
            "[398]\tvalidation_0-auc:0.842085\tvalidation_0-error:0.039551\n",
            "[399]\tvalidation_0-auc:0.842064\tvalidation_0-error:0.039551\n",
            "[400]\tvalidation_0-auc:0.842099\tvalidation_0-error:0.039551\n",
            "[401]\tvalidation_0-auc:0.842076\tvalidation_0-error:0.039551\n",
            "[402]\tvalidation_0-auc:0.842112\tvalidation_0-error:0.039551\n",
            "[403]\tvalidation_0-auc:0.84212\tvalidation_0-error:0.039551\n",
            "[404]\tvalidation_0-auc:0.842079\tvalidation_0-error:0.039551\n",
            "[405]\tvalidation_0-auc:0.842133\tvalidation_0-error:0.039551\n",
            "[406]\tvalidation_0-auc:0.842177\tvalidation_0-error:0.039551\n",
            "[407]\tvalidation_0-auc:0.842195\tvalidation_0-error:0.039551\n",
            "[408]\tvalidation_0-auc:0.842244\tvalidation_0-error:0.039551\n",
            "[409]\tvalidation_0-auc:0.842293\tvalidation_0-error:0.039551\n",
            "[410]\tvalidation_0-auc:0.842269\tvalidation_0-error:0.039551\n",
            "[411]\tvalidation_0-auc:0.842274\tvalidation_0-error:0.039551\n",
            "[412]\tvalidation_0-auc:0.842308\tvalidation_0-error:0.039551\n",
            "[413]\tvalidation_0-auc:0.842261\tvalidation_0-error:0.039551\n",
            "[414]\tvalidation_0-auc:0.842356\tvalidation_0-error:0.039551\n",
            "[415]\tvalidation_0-auc:0.842426\tvalidation_0-error:0.039551\n",
            "[416]\tvalidation_0-auc:0.842526\tvalidation_0-error:0.039551\n",
            "[417]\tvalidation_0-auc:0.842497\tvalidation_0-error:0.039551\n",
            "[418]\tvalidation_0-auc:0.842633\tvalidation_0-error:0.039551\n",
            "[419]\tvalidation_0-auc:0.84263\tvalidation_0-error:0.039551\n",
            "[420]\tvalidation_0-auc:0.842656\tvalidation_0-error:0.039551\n",
            "[421]\tvalidation_0-auc:0.842672\tvalidation_0-error:0.039551\n",
            "[422]\tvalidation_0-auc:0.842691\tvalidation_0-error:0.039551\n",
            "[423]\tvalidation_0-auc:0.842678\tvalidation_0-error:0.039551\n",
            "[424]\tvalidation_0-auc:0.8427\tvalidation_0-error:0.039551\n",
            "[425]\tvalidation_0-auc:0.842727\tvalidation_0-error:0.039551\n",
            "[426]\tvalidation_0-auc:0.842769\tvalidation_0-error:0.039551\n",
            "[427]\tvalidation_0-auc:0.842876\tvalidation_0-error:0.039551\n",
            "[428]\tvalidation_0-auc:0.842854\tvalidation_0-error:0.039551\n",
            "[429]\tvalidation_0-auc:0.842918\tvalidation_0-error:0.039551\n",
            "[430]\tvalidation_0-auc:0.842927\tvalidation_0-error:0.039551\n",
            "[431]\tvalidation_0-auc:0.842918\tvalidation_0-error:0.039551\n",
            "[432]\tvalidation_0-auc:0.842886\tvalidation_0-error:0.039551\n",
            "[433]\tvalidation_0-auc:0.842881\tvalidation_0-error:0.039551\n",
            "[434]\tvalidation_0-auc:0.842916\tvalidation_0-error:0.039551\n",
            "[435]\tvalidation_0-auc:0.842917\tvalidation_0-error:0.039551\n",
            "[436]\tvalidation_0-auc:0.842963\tvalidation_0-error:0.039551\n",
            "[437]\tvalidation_0-auc:0.843014\tvalidation_0-error:0.039551\n",
            "[438]\tvalidation_0-auc:0.843056\tvalidation_0-error:0.039551\n",
            "[439]\tvalidation_0-auc:0.84306\tvalidation_0-error:0.039551\n",
            "[440]\tvalidation_0-auc:0.843073\tvalidation_0-error:0.039551\n",
            "[441]\tvalidation_0-auc:0.843095\tvalidation_0-error:0.039551\n",
            "[442]\tvalidation_0-auc:0.843124\tvalidation_0-error:0.039551\n",
            "[443]\tvalidation_0-auc:0.843177\tvalidation_0-error:0.039551\n",
            "[444]\tvalidation_0-auc:0.843236\tvalidation_0-error:0.039551\n",
            "[445]\tvalidation_0-auc:0.84324\tvalidation_0-error:0.039551\n",
            "[446]\tvalidation_0-auc:0.843316\tvalidation_0-error:0.039551\n",
            "[447]\tvalidation_0-auc:0.843308\tvalidation_0-error:0.039551\n",
            "[448]\tvalidation_0-auc:0.843307\tvalidation_0-error:0.039551\n",
            "[449]\tvalidation_0-auc:0.843306\tvalidation_0-error:0.039551\n",
            "[450]\tvalidation_0-auc:0.843279\tvalidation_0-error:0.039551\n",
            "[451]\tvalidation_0-auc:0.843335\tvalidation_0-error:0.039551\n",
            "[452]\tvalidation_0-auc:0.843271\tvalidation_0-error:0.039551\n",
            "[453]\tvalidation_0-auc:0.843278\tvalidation_0-error:0.039551\n",
            "[454]\tvalidation_0-auc:0.843297\tvalidation_0-error:0.039551\n",
            "[455]\tvalidation_0-auc:0.843323\tvalidation_0-error:0.039551\n",
            "[456]\tvalidation_0-auc:0.843339\tvalidation_0-error:0.039551\n",
            "[457]\tvalidation_0-auc:0.843351\tvalidation_0-error:0.039551\n",
            "[458]\tvalidation_0-auc:0.84336\tvalidation_0-error:0.039551\n",
            "[459]\tvalidation_0-auc:0.843345\tvalidation_0-error:0.039551\n",
            "[460]\tvalidation_0-auc:0.84332\tvalidation_0-error:0.039551\n",
            "[461]\tvalidation_0-auc:0.843315\tvalidation_0-error:0.039551\n",
            "[462]\tvalidation_0-auc:0.843345\tvalidation_0-error:0.039551\n",
            "[463]\tvalidation_0-auc:0.843382\tvalidation_0-error:0.039551\n",
            "[464]\tvalidation_0-auc:0.843386\tvalidation_0-error:0.039551\n",
            "[465]\tvalidation_0-auc:0.843393\tvalidation_0-error:0.039551\n",
            "[466]\tvalidation_0-auc:0.843414\tvalidation_0-error:0.039551\n",
            "[467]\tvalidation_0-auc:0.843496\tvalidation_0-error:0.039551\n",
            "[468]\tvalidation_0-auc:0.843473\tvalidation_0-error:0.039551\n",
            "[469]\tvalidation_0-auc:0.843452\tvalidation_0-error:0.039551\n",
            "[470]\tvalidation_0-auc:0.843494\tvalidation_0-error:0.039551\n",
            "[471]\tvalidation_0-auc:0.843526\tvalidation_0-error:0.039551\n",
            "[472]\tvalidation_0-auc:0.843532\tvalidation_0-error:0.039551\n",
            "[473]\tvalidation_0-auc:0.843507\tvalidation_0-error:0.039551\n",
            "[474]\tvalidation_0-auc:0.843513\tvalidation_0-error:0.039551\n",
            "[475]\tvalidation_0-auc:0.843519\tvalidation_0-error:0.039551\n",
            "[476]\tvalidation_0-auc:0.843537\tvalidation_0-error:0.039551\n",
            "[477]\tvalidation_0-auc:0.84358\tvalidation_0-error:0.039551\n",
            "[478]\tvalidation_0-auc:0.843607\tvalidation_0-error:0.039551\n",
            "[479]\tvalidation_0-auc:0.843625\tvalidation_0-error:0.039551\n",
            "[480]\tvalidation_0-auc:0.843655\tvalidation_0-error:0.039551\n",
            "[481]\tvalidation_0-auc:0.843711\tvalidation_0-error:0.039551\n",
            "[482]\tvalidation_0-auc:0.843715\tvalidation_0-error:0.039551\n",
            "[483]\tvalidation_0-auc:0.843759\tvalidation_0-error:0.039551\n",
            "[484]\tvalidation_0-auc:0.843924\tvalidation_0-error:0.039551\n",
            "[485]\tvalidation_0-auc:0.843914\tvalidation_0-error:0.039551\n",
            "[486]\tvalidation_0-auc:0.843945\tvalidation_0-error:0.039551\n",
            "[487]\tvalidation_0-auc:0.843951\tvalidation_0-error:0.039551\n",
            "[488]\tvalidation_0-auc:0.843962\tvalidation_0-error:0.039551\n",
            "[489]\tvalidation_0-auc:0.843941\tvalidation_0-error:0.039551\n",
            "[490]\tvalidation_0-auc:0.843937\tvalidation_0-error:0.039551\n",
            "[491]\tvalidation_0-auc:0.843893\tvalidation_0-error:0.039551\n",
            "[492]\tvalidation_0-auc:0.84388\tvalidation_0-error:0.039551\n",
            "[493]\tvalidation_0-auc:0.843888\tvalidation_0-error:0.039551\n",
            "[494]\tvalidation_0-auc:0.843891\tvalidation_0-error:0.039551\n",
            "[495]\tvalidation_0-auc:0.843871\tvalidation_0-error:0.039551\n",
            "[496]\tvalidation_0-auc:0.843875\tvalidation_0-error:0.039551\n",
            "[497]\tvalidation_0-auc:0.843875\tvalidation_0-error:0.039551\n",
            "[498]\tvalidation_0-auc:0.843875\tvalidation_0-error:0.039551\n",
            "[499]\tvalidation_0-auc:0.843895\tvalidation_0-error:0.039551\n",
            "[500]\tvalidation_0-auc:0.843935\tvalidation_0-error:0.039551\n",
            "[501]\tvalidation_0-auc:0.843941\tvalidation_0-error:0.039551\n",
            "[502]\tvalidation_0-auc:0.843965\tvalidation_0-error:0.039551\n",
            "[503]\tvalidation_0-auc:0.84399\tvalidation_0-error:0.039551\n",
            "[504]\tvalidation_0-auc:0.843993\tvalidation_0-error:0.039551\n",
            "[505]\tvalidation_0-auc:0.844036\tvalidation_0-error:0.039551\n",
            "[506]\tvalidation_0-auc:0.844025\tvalidation_0-error:0.039551\n",
            "[507]\tvalidation_0-auc:0.844157\tvalidation_0-error:0.039551\n",
            "[508]\tvalidation_0-auc:0.844104\tvalidation_0-error:0.039551\n",
            "[509]\tvalidation_0-auc:0.844152\tvalidation_0-error:0.039551\n",
            "[510]\tvalidation_0-auc:0.844137\tvalidation_0-error:0.039551\n",
            "[511]\tvalidation_0-auc:0.844161\tvalidation_0-error:0.039551\n",
            "[512]\tvalidation_0-auc:0.844183\tvalidation_0-error:0.039551\n",
            "[513]\tvalidation_0-auc:0.844209\tvalidation_0-error:0.039551\n",
            "[514]\tvalidation_0-auc:0.844228\tvalidation_0-error:0.039551\n",
            "[515]\tvalidation_0-auc:0.844257\tvalidation_0-error:0.039551\n",
            "[516]\tvalidation_0-auc:0.844329\tvalidation_0-error:0.039551\n",
            "[517]\tvalidation_0-auc:0.844318\tvalidation_0-error:0.039551\n",
            "[518]\tvalidation_0-auc:0.844342\tvalidation_0-error:0.039551\n",
            "[519]\tvalidation_0-auc:0.844343\tvalidation_0-error:0.039551\n",
            "[520]\tvalidation_0-auc:0.84436\tvalidation_0-error:0.039551\n",
            "[521]\tvalidation_0-auc:0.844363\tvalidation_0-error:0.039551\n",
            "[522]\tvalidation_0-auc:0.844366\tvalidation_0-error:0.039551\n",
            "[523]\tvalidation_0-auc:0.844374\tvalidation_0-error:0.039551\n",
            "[524]\tvalidation_0-auc:0.844379\tvalidation_0-error:0.039551\n",
            "[525]\tvalidation_0-auc:0.844396\tvalidation_0-error:0.039551\n",
            "[526]\tvalidation_0-auc:0.844379\tvalidation_0-error:0.039551\n",
            "[527]\tvalidation_0-auc:0.844423\tvalidation_0-error:0.039551\n",
            "[528]\tvalidation_0-auc:0.844411\tvalidation_0-error:0.039551\n",
            "[529]\tvalidation_0-auc:0.844392\tvalidation_0-error:0.039551\n",
            "[530]\tvalidation_0-auc:0.844412\tvalidation_0-error:0.039551\n",
            "[531]\tvalidation_0-auc:0.844431\tvalidation_0-error:0.039551\n",
            "[532]\tvalidation_0-auc:0.844429\tvalidation_0-error:0.039551\n",
            "[533]\tvalidation_0-auc:0.844456\tvalidation_0-error:0.039551\n",
            "[534]\tvalidation_0-auc:0.844459\tvalidation_0-error:0.039551\n",
            "[535]\tvalidation_0-auc:0.84446\tvalidation_0-error:0.039551\n",
            "[536]\tvalidation_0-auc:0.844477\tvalidation_0-error:0.039551\n",
            "[537]\tvalidation_0-auc:0.844476\tvalidation_0-error:0.039551\n",
            "[538]\tvalidation_0-auc:0.844477\tvalidation_0-error:0.039551\n",
            "[539]\tvalidation_0-auc:0.844476\tvalidation_0-error:0.039551\n",
            "[540]\tvalidation_0-auc:0.844472\tvalidation_0-error:0.039551\n",
            "[541]\tvalidation_0-auc:0.84446\tvalidation_0-error:0.039551\n",
            "[542]\tvalidation_0-auc:0.844468\tvalidation_0-error:0.039551\n",
            "[543]\tvalidation_0-auc:0.844484\tvalidation_0-error:0.039551\n",
            "[544]\tvalidation_0-auc:0.844556\tvalidation_0-error:0.039551\n",
            "[545]\tvalidation_0-auc:0.84458\tvalidation_0-error:0.039551\n",
            "[546]\tvalidation_0-auc:0.844612\tvalidation_0-error:0.039551\n",
            "[547]\tvalidation_0-auc:0.84464\tvalidation_0-error:0.039551\n",
            "[548]\tvalidation_0-auc:0.84463\tvalidation_0-error:0.039551\n",
            "[549]\tvalidation_0-auc:0.844621\tvalidation_0-error:0.039551\n",
            "[550]\tvalidation_0-auc:0.844618\tvalidation_0-error:0.039551\n",
            "[551]\tvalidation_0-auc:0.844588\tvalidation_0-error:0.039551\n",
            "[552]\tvalidation_0-auc:0.844611\tvalidation_0-error:0.039551\n",
            "[553]\tvalidation_0-auc:0.844595\tvalidation_0-error:0.039551\n",
            "[554]\tvalidation_0-auc:0.844599\tvalidation_0-error:0.039551\n",
            "[555]\tvalidation_0-auc:0.844629\tvalidation_0-error:0.039551\n",
            "[556]\tvalidation_0-auc:0.844609\tvalidation_0-error:0.039551\n",
            "[557]\tvalidation_0-auc:0.844628\tvalidation_0-error:0.039551\n",
            "[558]\tvalidation_0-auc:0.84464\tvalidation_0-error:0.039551\n",
            "[559]\tvalidation_0-auc:0.844695\tvalidation_0-error:0.039551\n",
            "[560]\tvalidation_0-auc:0.844681\tvalidation_0-error:0.039551\n",
            "[561]\tvalidation_0-auc:0.844686\tvalidation_0-error:0.039551\n",
            "[562]\tvalidation_0-auc:0.844698\tvalidation_0-error:0.039551\n",
            "[563]\tvalidation_0-auc:0.844743\tvalidation_0-error:0.039551\n",
            "[564]\tvalidation_0-auc:0.844775\tvalidation_0-error:0.039551\n",
            "[565]\tvalidation_0-auc:0.844763\tvalidation_0-error:0.039551\n",
            "[566]\tvalidation_0-auc:0.844792\tvalidation_0-error:0.039551\n",
            "[567]\tvalidation_0-auc:0.844754\tvalidation_0-error:0.039551\n",
            "[568]\tvalidation_0-auc:0.844777\tvalidation_0-error:0.039551\n",
            "[569]\tvalidation_0-auc:0.844765\tvalidation_0-error:0.039551\n",
            "[570]\tvalidation_0-auc:0.844746\tvalidation_0-error:0.039551\n",
            "[571]\tvalidation_0-auc:0.844776\tvalidation_0-error:0.039551\n",
            "[572]\tvalidation_0-auc:0.844797\tvalidation_0-error:0.039551\n",
            "[573]\tvalidation_0-auc:0.844811\tvalidation_0-error:0.039551\n",
            "[574]\tvalidation_0-auc:0.844791\tvalidation_0-error:0.039551\n",
            "[575]\tvalidation_0-auc:0.844829\tvalidation_0-error:0.039551\n",
            "[576]\tvalidation_0-auc:0.844839\tvalidation_0-error:0.039551\n",
            "[577]\tvalidation_0-auc:0.844811\tvalidation_0-error:0.039551\n",
            "[578]\tvalidation_0-auc:0.844787\tvalidation_0-error:0.039551\n",
            "[579]\tvalidation_0-auc:0.844846\tvalidation_0-error:0.039551\n",
            "[580]\tvalidation_0-auc:0.844866\tvalidation_0-error:0.039551\n",
            "[581]\tvalidation_0-auc:0.844771\tvalidation_0-error:0.039551\n",
            "[582]\tvalidation_0-auc:0.84479\tvalidation_0-error:0.039551\n",
            "[583]\tvalidation_0-auc:0.844836\tvalidation_0-error:0.039551\n",
            "[584]\tvalidation_0-auc:0.844898\tvalidation_0-error:0.039551\n",
            "[585]\tvalidation_0-auc:0.84489\tvalidation_0-error:0.039551\n",
            "[586]\tvalidation_0-auc:0.844896\tvalidation_0-error:0.039551\n",
            "[587]\tvalidation_0-auc:0.844947\tvalidation_0-error:0.039551\n",
            "[588]\tvalidation_0-auc:0.844909\tvalidation_0-error:0.039551\n",
            "[589]\tvalidation_0-auc:0.84495\tvalidation_0-error:0.039551\n",
            "[590]\tvalidation_0-auc:0.84493\tvalidation_0-error:0.039551\n",
            "[591]\tvalidation_0-auc:0.844948\tvalidation_0-error:0.039551\n",
            "[592]\tvalidation_0-auc:0.844914\tvalidation_0-error:0.039551\n",
            "[593]\tvalidation_0-auc:0.844934\tvalidation_0-error:0.039551\n",
            "[594]\tvalidation_0-auc:0.844966\tvalidation_0-error:0.039551\n",
            "[595]\tvalidation_0-auc:0.844994\tvalidation_0-error:0.039551\n",
            "[596]\tvalidation_0-auc:0.845007\tvalidation_0-error:0.039551\n",
            "[597]\tvalidation_0-auc:0.845067\tvalidation_0-error:0.039551\n",
            "[598]\tvalidation_0-auc:0.845104\tvalidation_0-error:0.039551\n",
            "[599]\tvalidation_0-auc:0.845088\tvalidation_0-error:0.039551\n",
            "[600]\tvalidation_0-auc:0.84512\tvalidation_0-error:0.039551\n",
            "[601]\tvalidation_0-auc:0.845119\tvalidation_0-error:0.039551\n",
            "[602]\tvalidation_0-auc:0.845167\tvalidation_0-error:0.039551\n",
            "[603]\tvalidation_0-auc:0.845147\tvalidation_0-error:0.039551\n",
            "[604]\tvalidation_0-auc:0.84515\tvalidation_0-error:0.039551\n",
            "[605]\tvalidation_0-auc:0.845175\tvalidation_0-error:0.039551\n",
            "[606]\tvalidation_0-auc:0.845172\tvalidation_0-error:0.039551\n",
            "[607]\tvalidation_0-auc:0.845169\tvalidation_0-error:0.039551\n",
            "[608]\tvalidation_0-auc:0.845205\tvalidation_0-error:0.039551\n",
            "[609]\tvalidation_0-auc:0.845213\tvalidation_0-error:0.039551\n",
            "[610]\tvalidation_0-auc:0.845194\tvalidation_0-error:0.039551\n",
            "[611]\tvalidation_0-auc:0.845237\tvalidation_0-error:0.039551\n",
            "[612]\tvalidation_0-auc:0.84529\tvalidation_0-error:0.039551\n",
            "[613]\tvalidation_0-auc:0.845316\tvalidation_0-error:0.039551\n",
            "[614]\tvalidation_0-auc:0.8453\tvalidation_0-error:0.039551\n",
            "[615]\tvalidation_0-auc:0.845309\tvalidation_0-error:0.039551\n",
            "[616]\tvalidation_0-auc:0.845325\tvalidation_0-error:0.039551\n",
            "[617]\tvalidation_0-auc:0.845377\tvalidation_0-error:0.039551\n",
            "[618]\tvalidation_0-auc:0.845401\tvalidation_0-error:0.039551\n",
            "[619]\tvalidation_0-auc:0.845378\tvalidation_0-error:0.039551\n",
            "[620]\tvalidation_0-auc:0.845329\tvalidation_0-error:0.039551\n",
            "[621]\tvalidation_0-auc:0.845287\tvalidation_0-error:0.039551\n",
            "[622]\tvalidation_0-auc:0.845295\tvalidation_0-error:0.039551\n",
            "[623]\tvalidation_0-auc:0.84533\tvalidation_0-error:0.039551\n",
            "[624]\tvalidation_0-auc:0.845335\tvalidation_0-error:0.039551\n",
            "[625]\tvalidation_0-auc:0.845379\tvalidation_0-error:0.039551\n",
            "[626]\tvalidation_0-auc:0.84542\tvalidation_0-error:0.039551\n",
            "[627]\tvalidation_0-auc:0.845483\tvalidation_0-error:0.039551\n",
            "[628]\tvalidation_0-auc:0.845461\tvalidation_0-error:0.039551\n",
            "[629]\tvalidation_0-auc:0.845453\tvalidation_0-error:0.039551\n",
            "[630]\tvalidation_0-auc:0.845402\tvalidation_0-error:0.039551\n",
            "[631]\tvalidation_0-auc:0.84537\tvalidation_0-error:0.039551\n",
            "[632]\tvalidation_0-auc:0.845377\tvalidation_0-error:0.039551\n",
            "[633]\tvalidation_0-auc:0.845392\tvalidation_0-error:0.039551\n",
            "[634]\tvalidation_0-auc:0.845353\tvalidation_0-error:0.039551\n",
            "[635]\tvalidation_0-auc:0.84538\tvalidation_0-error:0.039551\n",
            "[636]\tvalidation_0-auc:0.845421\tvalidation_0-error:0.039551\n",
            "[637]\tvalidation_0-auc:0.845453\tvalidation_0-error:0.039551\n",
            "[638]\tvalidation_0-auc:0.845508\tvalidation_0-error:0.039551\n",
            "[639]\tvalidation_0-auc:0.845522\tvalidation_0-error:0.039551\n",
            "[640]\tvalidation_0-auc:0.845543\tvalidation_0-error:0.039551\n",
            "[641]\tvalidation_0-auc:0.845524\tvalidation_0-error:0.039551\n",
            "[642]\tvalidation_0-auc:0.845551\tvalidation_0-error:0.039551\n",
            "[643]\tvalidation_0-auc:0.845575\tvalidation_0-error:0.039551\n",
            "[644]\tvalidation_0-auc:0.845568\tvalidation_0-error:0.039551\n",
            "[645]\tvalidation_0-auc:0.845558\tvalidation_0-error:0.039551\n",
            "[646]\tvalidation_0-auc:0.845582\tvalidation_0-error:0.039551\n",
            "[647]\tvalidation_0-auc:0.845569\tvalidation_0-error:0.039551\n",
            "[648]\tvalidation_0-auc:0.845549\tvalidation_0-error:0.039551\n",
            "[649]\tvalidation_0-auc:0.84557\tvalidation_0-error:0.039551\n",
            "[650]\tvalidation_0-auc:0.845582\tvalidation_0-error:0.039551\n",
            "[651]\tvalidation_0-auc:0.84557\tvalidation_0-error:0.039551\n",
            "[652]\tvalidation_0-auc:0.84557\tvalidation_0-error:0.039551\n",
            "[653]\tvalidation_0-auc:0.845605\tvalidation_0-error:0.039551\n",
            "[654]\tvalidation_0-auc:0.845606\tvalidation_0-error:0.039551\n",
            "[655]\tvalidation_0-auc:0.845619\tvalidation_0-error:0.039551\n",
            "[656]\tvalidation_0-auc:0.845606\tvalidation_0-error:0.039551\n",
            "[657]\tvalidation_0-auc:0.845601\tvalidation_0-error:0.039551\n",
            "[658]\tvalidation_0-auc:0.845562\tvalidation_0-error:0.039551\n",
            "[659]\tvalidation_0-auc:0.845559\tvalidation_0-error:0.039551\n",
            "[660]\tvalidation_0-auc:0.845571\tvalidation_0-error:0.039551\n",
            "[661]\tvalidation_0-auc:0.845539\tvalidation_0-error:0.039551\n",
            "[662]\tvalidation_0-auc:0.845542\tvalidation_0-error:0.039551\n",
            "[663]\tvalidation_0-auc:0.845572\tvalidation_0-error:0.039551\n",
            "[664]\tvalidation_0-auc:0.845598\tvalidation_0-error:0.039551\n",
            "[665]\tvalidation_0-auc:0.845657\tvalidation_0-error:0.039551\n",
            "[666]\tvalidation_0-auc:0.845682\tvalidation_0-error:0.039551\n",
            "[667]\tvalidation_0-auc:0.845686\tvalidation_0-error:0.039551\n",
            "[668]\tvalidation_0-auc:0.845645\tvalidation_0-error:0.039551\n",
            "[669]\tvalidation_0-auc:0.8456\tvalidation_0-error:0.039551\n",
            "[670]\tvalidation_0-auc:0.845641\tvalidation_0-error:0.039551\n",
            "[671]\tvalidation_0-auc:0.845669\tvalidation_0-error:0.039551\n",
            "[672]\tvalidation_0-auc:0.845692\tvalidation_0-error:0.039551\n",
            "[673]\tvalidation_0-auc:0.845711\tvalidation_0-error:0.039551\n",
            "[674]\tvalidation_0-auc:0.845643\tvalidation_0-error:0.039551\n",
            "[675]\tvalidation_0-auc:0.845693\tvalidation_0-error:0.039551\n",
            "[676]\tvalidation_0-auc:0.845688\tvalidation_0-error:0.039551\n",
            "[677]\tvalidation_0-auc:0.845663\tvalidation_0-error:0.039551\n",
            "[678]\tvalidation_0-auc:0.845677\tvalidation_0-error:0.039551\n",
            "[679]\tvalidation_0-auc:0.845659\tvalidation_0-error:0.039551\n",
            "[680]\tvalidation_0-auc:0.845679\tvalidation_0-error:0.039551\n",
            "[681]\tvalidation_0-auc:0.845685\tvalidation_0-error:0.039551\n",
            "[682]\tvalidation_0-auc:0.845698\tvalidation_0-error:0.039551\n",
            "[683]\tvalidation_0-auc:0.845671\tvalidation_0-error:0.039551\n",
            "[684]\tvalidation_0-auc:0.845665\tvalidation_0-error:0.039551\n",
            "[685]\tvalidation_0-auc:0.845653\tvalidation_0-error:0.039551\n",
            "[686]\tvalidation_0-auc:0.84566\tvalidation_0-error:0.039551\n",
            "[687]\tvalidation_0-auc:0.845617\tvalidation_0-error:0.039551\n",
            "[688]\tvalidation_0-auc:0.845631\tvalidation_0-error:0.039551\n",
            "[689]\tvalidation_0-auc:0.845702\tvalidation_0-error:0.039551\n",
            "[690]\tvalidation_0-auc:0.845711\tvalidation_0-error:0.039551\n",
            "[691]\tvalidation_0-auc:0.845755\tvalidation_0-error:0.039551\n",
            "[692]\tvalidation_0-auc:0.845743\tvalidation_0-error:0.039551\n",
            "[693]\tvalidation_0-auc:0.845661\tvalidation_0-error:0.039551\n",
            "[694]\tvalidation_0-auc:0.845673\tvalidation_0-error:0.039551\n",
            "[695]\tvalidation_0-auc:0.845679\tvalidation_0-error:0.039551\n",
            "[696]\tvalidation_0-auc:0.845659\tvalidation_0-error:0.039551\n",
            "[697]\tvalidation_0-auc:0.845653\tvalidation_0-error:0.039551\n",
            "[698]\tvalidation_0-auc:0.845618\tvalidation_0-error:0.039551\n",
            "[699]\tvalidation_0-auc:0.845668\tvalidation_0-error:0.039551\n",
            "[700]\tvalidation_0-auc:0.845778\tvalidation_0-error:0.039551\n",
            "[701]\tvalidation_0-auc:0.845761\tvalidation_0-error:0.039551\n",
            "[702]\tvalidation_0-auc:0.845741\tvalidation_0-error:0.039551\n",
            "[703]\tvalidation_0-auc:0.84571\tvalidation_0-error:0.039551\n",
            "[704]\tvalidation_0-auc:0.845698\tvalidation_0-error:0.039551\n",
            "[705]\tvalidation_0-auc:0.845722\tvalidation_0-error:0.039551\n",
            "[706]\tvalidation_0-auc:0.845723\tvalidation_0-error:0.039551\n",
            "[707]\tvalidation_0-auc:0.845678\tvalidation_0-error:0.039551\n",
            "[708]\tvalidation_0-auc:0.845649\tvalidation_0-error:0.039551\n",
            "[709]\tvalidation_0-auc:0.845668\tvalidation_0-error:0.039551\n",
            "[710]\tvalidation_0-auc:0.845695\tvalidation_0-error:0.039551\n",
            "[711]\tvalidation_0-auc:0.845687\tvalidation_0-error:0.039551\n",
            "[712]\tvalidation_0-auc:0.845713\tvalidation_0-error:0.039551\n",
            "[713]\tvalidation_0-auc:0.845687\tvalidation_0-error:0.039551\n",
            "[714]\tvalidation_0-auc:0.845701\tvalidation_0-error:0.039551\n",
            "[715]\tvalidation_0-auc:0.845702\tvalidation_0-error:0.039551\n",
            "[716]\tvalidation_0-auc:0.845729\tvalidation_0-error:0.039551\n",
            "[717]\tvalidation_0-auc:0.845697\tvalidation_0-error:0.039551\n",
            "[718]\tvalidation_0-auc:0.845704\tvalidation_0-error:0.039551\n",
            "[719]\tvalidation_0-auc:0.845739\tvalidation_0-error:0.039551\n",
            "[720]\tvalidation_0-auc:0.845737\tvalidation_0-error:0.039551\n",
            "[721]\tvalidation_0-auc:0.845742\tvalidation_0-error:0.039551\n",
            "[722]\tvalidation_0-auc:0.845696\tvalidation_0-error:0.039551\n",
            "[723]\tvalidation_0-auc:0.845667\tvalidation_0-error:0.039551\n",
            "[724]\tvalidation_0-auc:0.845649\tvalidation_0-error:0.039551\n",
            "[725]\tvalidation_0-auc:0.845662\tvalidation_0-error:0.039551\n",
            "[726]\tvalidation_0-auc:0.845686\tvalidation_0-error:0.039551\n",
            "[727]\tvalidation_0-auc:0.845695\tvalidation_0-error:0.039551\n",
            "[728]\tvalidation_0-auc:0.845653\tvalidation_0-error:0.039551\n",
            "[729]\tvalidation_0-auc:0.845676\tvalidation_0-error:0.039551\n",
            "[730]\tvalidation_0-auc:0.845649\tvalidation_0-error:0.039551\n",
            "[731]\tvalidation_0-auc:0.845691\tvalidation_0-error:0.039551\n",
            "[732]\tvalidation_0-auc:0.845772\tvalidation_0-error:0.039551\n",
            "[733]\tvalidation_0-auc:0.845782\tvalidation_0-error:0.039551\n",
            "[734]\tvalidation_0-auc:0.845764\tvalidation_0-error:0.039551\n",
            "[735]\tvalidation_0-auc:0.845763\tvalidation_0-error:0.039551\n",
            "[736]\tvalidation_0-auc:0.84579\tvalidation_0-error:0.039551\n",
            "[737]\tvalidation_0-auc:0.845833\tvalidation_0-error:0.039551\n",
            "[738]\tvalidation_0-auc:0.845807\tvalidation_0-error:0.039551\n",
            "[739]\tvalidation_0-auc:0.845787\tvalidation_0-error:0.039551\n",
            "[740]\tvalidation_0-auc:0.845748\tvalidation_0-error:0.039551\n",
            "[741]\tvalidation_0-auc:0.845756\tvalidation_0-error:0.039551\n",
            "[742]\tvalidation_0-auc:0.845805\tvalidation_0-error:0.039551\n",
            "[743]\tvalidation_0-auc:0.845794\tvalidation_0-error:0.039551\n",
            "[744]\tvalidation_0-auc:0.845836\tvalidation_0-error:0.039551\n",
            "[745]\tvalidation_0-auc:0.845841\tvalidation_0-error:0.039551\n",
            "[746]\tvalidation_0-auc:0.845862\tvalidation_0-error:0.039551\n",
            "[747]\tvalidation_0-auc:0.84587\tvalidation_0-error:0.039551\n",
            "[748]\tvalidation_0-auc:0.845835\tvalidation_0-error:0.039551\n",
            "[749]\tvalidation_0-auc:0.845815\tvalidation_0-error:0.039551\n",
            "[750]\tvalidation_0-auc:0.845824\tvalidation_0-error:0.039551\n",
            "[751]\tvalidation_0-auc:0.845869\tvalidation_0-error:0.039551\n",
            "[752]\tvalidation_0-auc:0.845833\tvalidation_0-error:0.039551\n",
            "[753]\tvalidation_0-auc:0.845852\tvalidation_0-error:0.039551\n",
            "[754]\tvalidation_0-auc:0.845883\tvalidation_0-error:0.039551\n",
            "[755]\tvalidation_0-auc:0.845873\tvalidation_0-error:0.039551\n",
            "[756]\tvalidation_0-auc:0.84587\tvalidation_0-error:0.039551\n",
            "[757]\tvalidation_0-auc:0.845867\tvalidation_0-error:0.039551\n",
            "[758]\tvalidation_0-auc:0.845864\tvalidation_0-error:0.039551\n",
            "[759]\tvalidation_0-auc:0.845871\tvalidation_0-error:0.039551\n",
            "[760]\tvalidation_0-auc:0.845861\tvalidation_0-error:0.039551\n",
            "[761]\tvalidation_0-auc:0.845862\tvalidation_0-error:0.039551\n",
            "[762]\tvalidation_0-auc:0.845894\tvalidation_0-error:0.039551\n",
            "[763]\tvalidation_0-auc:0.845879\tvalidation_0-error:0.039551\n",
            "[764]\tvalidation_0-auc:0.845891\tvalidation_0-error:0.039551\n",
            "[765]\tvalidation_0-auc:0.845911\tvalidation_0-error:0.039551\n",
            "[766]\tvalidation_0-auc:0.845911\tvalidation_0-error:0.039551\n",
            "[767]\tvalidation_0-auc:0.845932\tvalidation_0-error:0.039551\n",
            "[768]\tvalidation_0-auc:0.845914\tvalidation_0-error:0.039551\n",
            "[769]\tvalidation_0-auc:0.845931\tvalidation_0-error:0.039551\n",
            "[770]\tvalidation_0-auc:0.845918\tvalidation_0-error:0.039551\n",
            "[771]\tvalidation_0-auc:0.845917\tvalidation_0-error:0.039551\n",
            "[772]\tvalidation_0-auc:0.845924\tvalidation_0-error:0.039551\n",
            "[773]\tvalidation_0-auc:0.845946\tvalidation_0-error:0.039551\n",
            "[774]\tvalidation_0-auc:0.846\tvalidation_0-error:0.039551\n",
            "[775]\tvalidation_0-auc:0.84604\tvalidation_0-error:0.039551\n",
            "[776]\tvalidation_0-auc:0.846049\tvalidation_0-error:0.039551\n",
            "[777]\tvalidation_0-auc:0.846082\tvalidation_0-error:0.039551\n",
            "[778]\tvalidation_0-auc:0.846036\tvalidation_0-error:0.039551\n",
            "[779]\tvalidation_0-auc:0.845999\tvalidation_0-error:0.039551\n",
            "[780]\tvalidation_0-auc:0.845972\tvalidation_0-error:0.039551\n",
            "[781]\tvalidation_0-auc:0.845976\tvalidation_0-error:0.039551\n",
            "[782]\tvalidation_0-auc:0.845926\tvalidation_0-error:0.039551\n",
            "[783]\tvalidation_0-auc:0.845924\tvalidation_0-error:0.039551\n",
            "[784]\tvalidation_0-auc:0.845908\tvalidation_0-error:0.039551\n",
            "[785]\tvalidation_0-auc:0.845902\tvalidation_0-error:0.039551\n",
            "[786]\tvalidation_0-auc:0.84593\tvalidation_0-error:0.039551\n",
            "[787]\tvalidation_0-auc:0.845894\tvalidation_0-error:0.039551\n",
            "[788]\tvalidation_0-auc:0.845908\tvalidation_0-error:0.039551\n",
            "[789]\tvalidation_0-auc:0.845898\tvalidation_0-error:0.039551\n",
            "[790]\tvalidation_0-auc:0.845953\tvalidation_0-error:0.039551\n",
            "[791]\tvalidation_0-auc:0.84589\tvalidation_0-error:0.039551\n",
            "[792]\tvalidation_0-auc:0.845903\tvalidation_0-error:0.039551\n",
            "[793]\tvalidation_0-auc:0.845843\tvalidation_0-error:0.039551\n",
            "[794]\tvalidation_0-auc:0.845845\tvalidation_0-error:0.039551\n",
            "[795]\tvalidation_0-auc:0.845848\tvalidation_0-error:0.039551\n",
            "[796]\tvalidation_0-auc:0.845839\tvalidation_0-error:0.039551\n",
            "[797]\tvalidation_0-auc:0.845841\tvalidation_0-error:0.039551\n",
            "[798]\tvalidation_0-auc:0.845807\tvalidation_0-error:0.039551\n",
            "[799]\tvalidation_0-auc:0.845834\tvalidation_0-error:0.039551\n",
            "[800]\tvalidation_0-auc:0.845829\tvalidation_0-error:0.039551\n",
            "[801]\tvalidation_0-auc:0.845816\tvalidation_0-error:0.039551\n",
            "[802]\tvalidation_0-auc:0.845817\tvalidation_0-error:0.039551\n",
            "[803]\tvalidation_0-auc:0.845814\tvalidation_0-error:0.039551\n",
            "[804]\tvalidation_0-auc:0.845805\tvalidation_0-error:0.039551\n",
            "[805]\tvalidation_0-auc:0.845777\tvalidation_0-error:0.039551\n",
            "[806]\tvalidation_0-auc:0.845754\tvalidation_0-error:0.039551\n",
            "[807]\tvalidation_0-auc:0.845752\tvalidation_0-error:0.039551\n",
            "[808]\tvalidation_0-auc:0.845774\tvalidation_0-error:0.039551\n",
            "[809]\tvalidation_0-auc:0.845786\tvalidation_0-error:0.039551\n",
            "[810]\tvalidation_0-auc:0.845778\tvalidation_0-error:0.039551\n",
            "[811]\tvalidation_0-auc:0.845789\tvalidation_0-error:0.039551\n",
            "[812]\tvalidation_0-auc:0.845807\tvalidation_0-error:0.039551\n",
            "[813]\tvalidation_0-auc:0.845814\tvalidation_0-error:0.039551\n",
            "[814]\tvalidation_0-auc:0.845798\tvalidation_0-error:0.039551\n",
            "[815]\tvalidation_0-auc:0.845772\tvalidation_0-error:0.039551\n",
            "[816]\tvalidation_0-auc:0.845848\tvalidation_0-error:0.039551\n",
            "[817]\tvalidation_0-auc:0.845827\tvalidation_0-error:0.039551\n",
            "[818]\tvalidation_0-auc:0.845809\tvalidation_0-error:0.039551\n",
            "[819]\tvalidation_0-auc:0.845812\tvalidation_0-error:0.039551\n",
            "[820]\tvalidation_0-auc:0.845824\tvalidation_0-error:0.039551\n",
            "[821]\tvalidation_0-auc:0.845862\tvalidation_0-error:0.039551\n",
            "[822]\tvalidation_0-auc:0.845886\tvalidation_0-error:0.039551\n",
            "[823]\tvalidation_0-auc:0.845918\tvalidation_0-error:0.039551\n",
            "[824]\tvalidation_0-auc:0.845891\tvalidation_0-error:0.039551\n",
            "[825]\tvalidation_0-auc:0.845896\tvalidation_0-error:0.039551\n",
            "[826]\tvalidation_0-auc:0.845899\tvalidation_0-error:0.039551\n",
            "[827]\tvalidation_0-auc:0.845888\tvalidation_0-error:0.039551\n",
            "[828]\tvalidation_0-auc:0.845926\tvalidation_0-error:0.039551\n",
            "[829]\tvalidation_0-auc:0.845899\tvalidation_0-error:0.039551\n",
            "[830]\tvalidation_0-auc:0.845925\tvalidation_0-error:0.039551\n",
            "[831]\tvalidation_0-auc:0.845934\tvalidation_0-error:0.039551\n",
            "[832]\tvalidation_0-auc:0.845897\tvalidation_0-error:0.039551\n",
            "[833]\tvalidation_0-auc:0.845874\tvalidation_0-error:0.039551\n",
            "[834]\tvalidation_0-auc:0.845876\tvalidation_0-error:0.039551\n",
            "[835]\tvalidation_0-auc:0.845854\tvalidation_0-error:0.039551\n",
            "[836]\tvalidation_0-auc:0.84586\tvalidation_0-error:0.039551\n",
            "[837]\tvalidation_0-auc:0.845855\tvalidation_0-error:0.039551\n",
            "[838]\tvalidation_0-auc:0.845881\tvalidation_0-error:0.039551\n",
            "[839]\tvalidation_0-auc:0.845865\tvalidation_0-error:0.039551\n",
            "[840]\tvalidation_0-auc:0.845885\tvalidation_0-error:0.039551\n",
            "[841]\tvalidation_0-auc:0.84586\tvalidation_0-error:0.039551\n",
            "[842]\tvalidation_0-auc:0.845909\tvalidation_0-error:0.039551\n",
            "[843]\tvalidation_0-auc:0.845927\tvalidation_0-error:0.039551\n",
            "[844]\tvalidation_0-auc:0.845949\tvalidation_0-error:0.039551\n",
            "[845]\tvalidation_0-auc:0.845926\tvalidation_0-error:0.039551\n",
            "[846]\tvalidation_0-auc:0.845931\tvalidation_0-error:0.039551\n",
            "[847]\tvalidation_0-auc:0.845895\tvalidation_0-error:0.039551\n",
            "[848]\tvalidation_0-auc:0.845912\tvalidation_0-error:0.039551\n",
            "[849]\tvalidation_0-auc:0.845901\tvalidation_0-error:0.039551\n",
            "[850]\tvalidation_0-auc:0.845893\tvalidation_0-error:0.039551\n",
            "[851]\tvalidation_0-auc:0.84584\tvalidation_0-error:0.039551\n",
            "[852]\tvalidation_0-auc:0.845836\tvalidation_0-error:0.039551\n",
            "[853]\tvalidation_0-auc:0.845843\tvalidation_0-error:0.039551\n",
            "[854]\tvalidation_0-auc:0.845832\tvalidation_0-error:0.039551\n",
            "[855]\tvalidation_0-auc:0.845823\tvalidation_0-error:0.039551\n",
            "[856]\tvalidation_0-auc:0.845824\tvalidation_0-error:0.039551\n",
            "[857]\tvalidation_0-auc:0.845799\tvalidation_0-error:0.039551\n",
            "[858]\tvalidation_0-auc:0.845775\tvalidation_0-error:0.039551\n",
            "[859]\tvalidation_0-auc:0.84576\tvalidation_0-error:0.039551\n",
            "[860]\tvalidation_0-auc:0.845747\tvalidation_0-error:0.039551\n",
            "[861]\tvalidation_0-auc:0.845723\tvalidation_0-error:0.039551\n",
            "[862]\tvalidation_0-auc:0.845754\tvalidation_0-error:0.039551\n",
            "[863]\tvalidation_0-auc:0.845744\tvalidation_0-error:0.039551\n",
            "[864]\tvalidation_0-auc:0.84576\tvalidation_0-error:0.039551\n",
            "CPU times: user 12min 59s, sys: 1.39 s, total: 13min 1s\n",
            "Wall time: 6min 41s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
              "              colsample_bynode=1, colsample_bytree=0.8, gamma=1,\n",
              "              learning_rate=0.01, max_delta_step=0, max_depth=5,\n",
              "              min_child_weight=1, missing=None, n_estimators=865, n_jobs=-1,\n",
              "              nthread=-1, objective='binary:logistic', random_state=42,\n",
              "              reg_alpha=0.3, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
              "              silent=None, subsample=0.7, verbosity=1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "9ZI4N9TM-V3D",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "c9822e25-df42-4fcb-9ff0-709416ea2b0f"
      },
      "source": [
        "y_pred = model.predict_proba(X_val_200)[:,1]\n",
        "label=\"XGBoost Classifer(top 250)\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "i+=1\n",
        "result = plot_auc(y_val,y_pred,label)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t XGBoost Classifer(top 250) on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.133 and 0.846 for XGBoost Classifer(top 250) on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "LN2FW-s2-V3G",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "outputId": "e4656801-fb6e-4092-8a2b-681a1ed19eae"
      },
      "source": [
        "result_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133739</td>\n",
              "      <td>0.845052</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>LightGBM Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.134520</td>\n",
              "      <td>0.845469</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Random Forest Classifier (top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136440</td>\n",
              "      <td>0.837497</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>XGBoost Classifer(top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133050</td>\n",
              "      <td>0.845760</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                Model  ...       auc\n",
              "0                  Logisic Regression  ...  0.823515\n",
              "1            Decision Tree Classifier  ...  0.769599\n",
              "2            Random Forest Classifier  ...  0.836315\n",
              "3                   XGBoost Classifer  ...  0.845052\n",
              "4                  LightGBM Classifer  ...  0.845469\n",
              "5  Random Forest Classifier (top 250)  ...  0.837497\n",
              "6          XGBoost Classifer(top 250)  ...  0.845760\n",
              "\n",
              "[7 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "mYW7_x88-V3J"
      },
      "source": [
        "# LightGBM Classifier (top 250)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "EYaevQf8-V3J",
        "colab": {}
      },
      "source": [
        "model = lgb.LGBMClassifier(n_jobs=-1,\n",
        "                      nthread=-1,\n",
        "                      scale_pos_weight=1,\n",
        "                      learning_rate=0.01,  \n",
        "                      colsample_bytree = 0.5,\n",
        "                      subsample = 1,\n",
        "                      objective='binary', \n",
        "                      n_estimators=447,\n",
        "                      reg_alpha = 0,\n",
        "                      max_depth=7,\n",
        "                      gamma=5,\n",
        "                      random_state=42\n",
        "                     )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "JhKg3-Lv-V3M",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "5fe9a400-a9eb-465c-c904-f7e4dddd1654"
      },
      "source": [
        "#https://towardsdatascience.com/fine-tuning-xgboost-in-python-like-a-boss-b4543ed8b1e\n",
        "\"\"\"\n",
        "manually changed each parameters 1 by 1 and compared the validation auc\n",
        "and selected the best parameter value\n",
        "\"\"\"\n",
        "eval_metric = ['auc','error']\n",
        "%time model.fit(X_train_200,y_train,eval_set=[(X_val_200,y_val)],eval_metric=eval_metric,verbose=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1]\tvalid_0's auc: 0.808366\tvalid_0's binary_logloss: 0.165786\n",
            "[2]\tvalid_0's auc: 0.816347\tvalid_0's binary_logloss: 0.165143\n",
            "[3]\tvalid_0's auc: 0.818707\tvalid_0's binary_logloss: 0.164478\n",
            "[4]\tvalid_0's auc: 0.81881\tvalid_0's binary_logloss: 0.163828\n",
            "[5]\tvalid_0's auc: 0.823873\tvalid_0's binary_logloss: 0.163312\n",
            "[6]\tvalid_0's auc: 0.823931\tvalid_0's binary_logloss: 0.162781\n",
            "[7]\tvalid_0's auc: 0.825442\tvalid_0's binary_logloss: 0.162189\n",
            "[8]\tvalid_0's auc: 0.829788\tvalid_0's binary_logloss: 0.161575\n",
            "[9]\tvalid_0's auc: 0.830491\tvalid_0's binary_logloss: 0.161072\n",
            "[10]\tvalid_0's auc: 0.83084\tvalid_0's binary_logloss: 0.160553\n",
            "[11]\tvalid_0's auc: 0.831279\tvalid_0's binary_logloss: 0.160053\n",
            "[12]\tvalid_0's auc: 0.832585\tvalid_0's binary_logloss: 0.159541\n",
            "[13]\tvalid_0's auc: 0.833993\tvalid_0's binary_logloss: 0.159033\n",
            "[14]\tvalid_0's auc: 0.834333\tvalid_0's binary_logloss: 0.15858\n",
            "[15]\tvalid_0's auc: 0.834601\tvalid_0's binary_logloss: 0.158128\n",
            "[16]\tvalid_0's auc: 0.835515\tvalid_0's binary_logloss: 0.157675\n",
            "[17]\tvalid_0's auc: 0.835605\tvalid_0's binary_logloss: 0.157238\n",
            "[18]\tvalid_0's auc: 0.836244\tvalid_0's binary_logloss: 0.156802\n",
            "[19]\tvalid_0's auc: 0.836408\tvalid_0's binary_logloss: 0.156401\n",
            "[20]\tvalid_0's auc: 0.83653\tvalid_0's binary_logloss: 0.156011\n",
            "[21]\tvalid_0's auc: 0.836213\tvalid_0's binary_logloss: 0.155652\n",
            "[22]\tvalid_0's auc: 0.835584\tvalid_0's binary_logloss: 0.1553\n",
            "[23]\tvalid_0's auc: 0.835829\tvalid_0's binary_logloss: 0.154941\n",
            "[24]\tvalid_0's auc: 0.835723\tvalid_0's binary_logloss: 0.154623\n",
            "[25]\tvalid_0's auc: 0.835542\tvalid_0's binary_logloss: 0.154283\n",
            "[26]\tvalid_0's auc: 0.83546\tvalid_0's binary_logloss: 0.153939\n",
            "[27]\tvalid_0's auc: 0.835543\tvalid_0's binary_logloss: 0.153614\n",
            "[28]\tvalid_0's auc: 0.836095\tvalid_0's binary_logloss: 0.153291\n",
            "[29]\tvalid_0's auc: 0.835728\tvalid_0's binary_logloss: 0.152967\n",
            "[30]\tvalid_0's auc: 0.835932\tvalid_0's binary_logloss: 0.152659\n",
            "[31]\tvalid_0's auc: 0.83576\tvalid_0's binary_logloss: 0.152361\n",
            "[32]\tvalid_0's auc: 0.835739\tvalid_0's binary_logloss: 0.152075\n",
            "[33]\tvalid_0's auc: 0.835385\tvalid_0's binary_logloss: 0.151815\n",
            "[34]\tvalid_0's auc: 0.835715\tvalid_0's binary_logloss: 0.151549\n",
            "[35]\tvalid_0's auc: 0.836224\tvalid_0's binary_logloss: 0.151299\n",
            "[36]\tvalid_0's auc: 0.836482\tvalid_0's binary_logloss: 0.15101\n",
            "[37]\tvalid_0's auc: 0.83672\tvalid_0's binary_logloss: 0.150714\n",
            "[38]\tvalid_0's auc: 0.836998\tvalid_0's binary_logloss: 0.150444\n",
            "[39]\tvalid_0's auc: 0.837232\tvalid_0's binary_logloss: 0.150159\n",
            "[40]\tvalid_0's auc: 0.837258\tvalid_0's binary_logloss: 0.149887\n",
            "[41]\tvalid_0's auc: 0.837272\tvalid_0's binary_logloss: 0.149635\n",
            "[42]\tvalid_0's auc: 0.837137\tvalid_0's binary_logloss: 0.149391\n",
            "[43]\tvalid_0's auc: 0.837325\tvalid_0's binary_logloss: 0.149194\n",
            "[44]\tvalid_0's auc: 0.837352\tvalid_0's binary_logloss: 0.148972\n",
            "[45]\tvalid_0's auc: 0.837307\tvalid_0's binary_logloss: 0.148781\n",
            "[46]\tvalid_0's auc: 0.837451\tvalid_0's binary_logloss: 0.148552\n",
            "[47]\tvalid_0's auc: 0.83745\tvalid_0's binary_logloss: 0.14837\n",
            "[48]\tvalid_0's auc: 0.837225\tvalid_0's binary_logloss: 0.148148\n",
            "[49]\tvalid_0's auc: 0.836987\tvalid_0's binary_logloss: 0.147948\n",
            "[50]\tvalid_0's auc: 0.836997\tvalid_0's binary_logloss: 0.147745\n",
            "[51]\tvalid_0's auc: 0.837\tvalid_0's binary_logloss: 0.147536\n",
            "[52]\tvalid_0's auc: 0.837028\tvalid_0's binary_logloss: 0.147369\n",
            "[53]\tvalid_0's auc: 0.836982\tvalid_0's binary_logloss: 0.147169\n",
            "[54]\tvalid_0's auc: 0.836995\tvalid_0's binary_logloss: 0.146965\n",
            "[55]\tvalid_0's auc: 0.83704\tvalid_0's binary_logloss: 0.146761\n",
            "[56]\tvalid_0's auc: 0.837034\tvalid_0's binary_logloss: 0.146569\n",
            "[57]\tvalid_0's auc: 0.837006\tvalid_0's binary_logloss: 0.146376\n",
            "[58]\tvalid_0's auc: 0.837427\tvalid_0's binary_logloss: 0.146172\n",
            "[59]\tvalid_0's auc: 0.837343\tvalid_0's binary_logloss: 0.146007\n",
            "[60]\tvalid_0's auc: 0.837397\tvalid_0's binary_logloss: 0.145827\n",
            "[61]\tvalid_0's auc: 0.837564\tvalid_0's binary_logloss: 0.14565\n",
            "[62]\tvalid_0's auc: 0.83749\tvalid_0's binary_logloss: 0.145474\n",
            "[63]\tvalid_0's auc: 0.837549\tvalid_0's binary_logloss: 0.145298\n",
            "[64]\tvalid_0's auc: 0.837535\tvalid_0's binary_logloss: 0.145119\n",
            "[65]\tvalid_0's auc: 0.837744\tvalid_0's binary_logloss: 0.144978\n",
            "[66]\tvalid_0's auc: 0.837976\tvalid_0's binary_logloss: 0.144804\n",
            "[67]\tvalid_0's auc: 0.838045\tvalid_0's binary_logloss: 0.144641\n",
            "[68]\tvalid_0's auc: 0.838077\tvalid_0's binary_logloss: 0.144521\n",
            "[69]\tvalid_0's auc: 0.837881\tvalid_0's binary_logloss: 0.144408\n",
            "[70]\tvalid_0's auc: 0.837946\tvalid_0's binary_logloss: 0.144254\n",
            "[71]\tvalid_0's auc: 0.838081\tvalid_0's binary_logloss: 0.144104\n",
            "[72]\tvalid_0's auc: 0.838131\tvalid_0's binary_logloss: 0.143964\n",
            "[73]\tvalid_0's auc: 0.83808\tvalid_0's binary_logloss: 0.143858\n",
            "[74]\tvalid_0's auc: 0.838189\tvalid_0's binary_logloss: 0.143728\n",
            "[75]\tvalid_0's auc: 0.838293\tvalid_0's binary_logloss: 0.143609\n",
            "[76]\tvalid_0's auc: 0.838367\tvalid_0's binary_logloss: 0.143461\n",
            "[77]\tvalid_0's auc: 0.83848\tvalid_0's binary_logloss: 0.143313\n",
            "[78]\tvalid_0's auc: 0.838454\tvalid_0's binary_logloss: 0.14316\n",
            "[79]\tvalid_0's auc: 0.838524\tvalid_0's binary_logloss: 0.143026\n",
            "[80]\tvalid_0's auc: 0.838758\tvalid_0's binary_logloss: 0.142893\n",
            "[81]\tvalid_0's auc: 0.838766\tvalid_0's binary_logloss: 0.142797\n",
            "[82]\tvalid_0's auc: 0.838718\tvalid_0's binary_logloss: 0.14271\n",
            "[83]\tvalid_0's auc: 0.838779\tvalid_0's binary_logloss: 0.142565\n",
            "[84]\tvalid_0's auc: 0.838925\tvalid_0's binary_logloss: 0.142424\n",
            "[85]\tvalid_0's auc: 0.838921\tvalid_0's binary_logloss: 0.14231\n",
            "[86]\tvalid_0's auc: 0.839182\tvalid_0's binary_logloss: 0.142177\n",
            "[87]\tvalid_0's auc: 0.839123\tvalid_0's binary_logloss: 0.14206\n",
            "[88]\tvalid_0's auc: 0.839392\tvalid_0's binary_logloss: 0.141931\n",
            "[89]\tvalid_0's auc: 0.839477\tvalid_0's binary_logloss: 0.141815\n",
            "[90]\tvalid_0's auc: 0.839529\tvalid_0's binary_logloss: 0.141711\n",
            "[91]\tvalid_0's auc: 0.839563\tvalid_0's binary_logloss: 0.141584\n",
            "[92]\tvalid_0's auc: 0.839722\tvalid_0's binary_logloss: 0.141479\n",
            "[93]\tvalid_0's auc: 0.839688\tvalid_0's binary_logloss: 0.141383\n",
            "[94]\tvalid_0's auc: 0.840058\tvalid_0's binary_logloss: 0.141272\n",
            "[95]\tvalid_0's auc: 0.83999\tvalid_0's binary_logloss: 0.141178\n",
            "[96]\tvalid_0's auc: 0.84025\tvalid_0's binary_logloss: 0.141071\n",
            "[97]\tvalid_0's auc: 0.840227\tvalid_0's binary_logloss: 0.140961\n",
            "[98]\tvalid_0's auc: 0.840458\tvalid_0's binary_logloss: 0.140851\n",
            "[99]\tvalid_0's auc: 0.840243\tvalid_0's binary_logloss: 0.140751\n",
            "[100]\tvalid_0's auc: 0.840315\tvalid_0's binary_logloss: 0.140648\n",
            "[101]\tvalid_0's auc: 0.840365\tvalid_0's binary_logloss: 0.140558\n",
            "[102]\tvalid_0's auc: 0.840389\tvalid_0's binary_logloss: 0.140475\n",
            "[103]\tvalid_0's auc: 0.840394\tvalid_0's binary_logloss: 0.140402\n",
            "[104]\tvalid_0's auc: 0.840409\tvalid_0's binary_logloss: 0.140331\n",
            "[105]\tvalid_0's auc: 0.840416\tvalid_0's binary_logloss: 0.140251\n",
            "[106]\tvalid_0's auc: 0.840477\tvalid_0's binary_logloss: 0.140155\n",
            "[107]\tvalid_0's auc: 0.840598\tvalid_0's binary_logloss: 0.140065\n",
            "[108]\tvalid_0's auc: 0.840707\tvalid_0's binary_logloss: 0.139967\n",
            "[109]\tvalid_0's auc: 0.840665\tvalid_0's binary_logloss: 0.139904\n",
            "[110]\tvalid_0's auc: 0.840819\tvalid_0's binary_logloss: 0.139816\n",
            "[111]\tvalid_0's auc: 0.840917\tvalid_0's binary_logloss: 0.13972\n",
            "[112]\tvalid_0's auc: 0.840924\tvalid_0's binary_logloss: 0.139641\n",
            "[113]\tvalid_0's auc: 0.841027\tvalid_0's binary_logloss: 0.139572\n",
            "[114]\tvalid_0's auc: 0.841016\tvalid_0's binary_logloss: 0.139504\n",
            "[115]\tvalid_0's auc: 0.840935\tvalid_0's binary_logloss: 0.139444\n",
            "[116]\tvalid_0's auc: 0.840962\tvalid_0's binary_logloss: 0.139376\n",
            "[117]\tvalid_0's auc: 0.841075\tvalid_0's binary_logloss: 0.139297\n",
            "[118]\tvalid_0's auc: 0.841254\tvalid_0's binary_logloss: 0.139227\n",
            "[119]\tvalid_0's auc: 0.841264\tvalid_0's binary_logloss: 0.139154\n",
            "[120]\tvalid_0's auc: 0.841238\tvalid_0's binary_logloss: 0.139086\n",
            "[121]\tvalid_0's auc: 0.841306\tvalid_0's binary_logloss: 0.138993\n",
            "[122]\tvalid_0's auc: 0.841237\tvalid_0's binary_logloss: 0.138934\n",
            "[123]\tvalid_0's auc: 0.841251\tvalid_0's binary_logloss: 0.138856\n",
            "[124]\tvalid_0's auc: 0.841224\tvalid_0's binary_logloss: 0.13878\n",
            "[125]\tvalid_0's auc: 0.841262\tvalid_0's binary_logloss: 0.138713\n",
            "[126]\tvalid_0's auc: 0.841347\tvalid_0's binary_logloss: 0.138654\n",
            "[127]\tvalid_0's auc: 0.841418\tvalid_0's binary_logloss: 0.138582\n",
            "[128]\tvalid_0's auc: 0.841548\tvalid_0's binary_logloss: 0.138537\n",
            "[129]\tvalid_0's auc: 0.841665\tvalid_0's binary_logloss: 0.138472\n",
            "[130]\tvalid_0's auc: 0.841576\tvalid_0's binary_logloss: 0.138416\n",
            "[131]\tvalid_0's auc: 0.841619\tvalid_0's binary_logloss: 0.138363\n",
            "[132]\tvalid_0's auc: 0.841598\tvalid_0's binary_logloss: 0.138309\n",
            "[133]\tvalid_0's auc: 0.841671\tvalid_0's binary_logloss: 0.138244\n",
            "[134]\tvalid_0's auc: 0.841644\tvalid_0's binary_logloss: 0.138204\n",
            "[135]\tvalid_0's auc: 0.841677\tvalid_0's binary_logloss: 0.138154\n",
            "[136]\tvalid_0's auc: 0.841684\tvalid_0's binary_logloss: 0.138106\n",
            "[137]\tvalid_0's auc: 0.841748\tvalid_0's binary_logloss: 0.138042\n",
            "[138]\tvalid_0's auc: 0.841795\tvalid_0's binary_logloss: 0.137986\n",
            "[139]\tvalid_0's auc: 0.841768\tvalid_0's binary_logloss: 0.137945\n",
            "[140]\tvalid_0's auc: 0.841945\tvalid_0's binary_logloss: 0.13789\n",
            "[141]\tvalid_0's auc: 0.841892\tvalid_0's binary_logloss: 0.13784\n",
            "[142]\tvalid_0's auc: 0.841889\tvalid_0's binary_logloss: 0.137808\n",
            "[143]\tvalid_0's auc: 0.841886\tvalid_0's binary_logloss: 0.137752\n",
            "[144]\tvalid_0's auc: 0.841923\tvalid_0's binary_logloss: 0.137694\n",
            "[145]\tvalid_0's auc: 0.84199\tvalid_0's binary_logloss: 0.137646\n",
            "[146]\tvalid_0's auc: 0.842076\tvalid_0's binary_logloss: 0.137603\n",
            "[147]\tvalid_0's auc: 0.84209\tvalid_0's binary_logloss: 0.137552\n",
            "[148]\tvalid_0's auc: 0.842086\tvalid_0's binary_logloss: 0.137503\n",
            "[149]\tvalid_0's auc: 0.842127\tvalid_0's binary_logloss: 0.137446\n",
            "[150]\tvalid_0's auc: 0.842248\tvalid_0's binary_logloss: 0.137398\n",
            "[151]\tvalid_0's auc: 0.84228\tvalid_0's binary_logloss: 0.137351\n",
            "[152]\tvalid_0's auc: 0.842393\tvalid_0's binary_logloss: 0.137297\n",
            "[153]\tvalid_0's auc: 0.842463\tvalid_0's binary_logloss: 0.137248\n",
            "[154]\tvalid_0's auc: 0.842533\tvalid_0's binary_logloss: 0.137207\n",
            "[155]\tvalid_0's auc: 0.84246\tvalid_0's binary_logloss: 0.137168\n",
            "[156]\tvalid_0's auc: 0.842518\tvalid_0's binary_logloss: 0.137106\n",
            "[157]\tvalid_0's auc: 0.842581\tvalid_0's binary_logloss: 0.137053\n",
            "[158]\tvalid_0's auc: 0.842618\tvalid_0's binary_logloss: 0.137009\n",
            "[159]\tvalid_0's auc: 0.842597\tvalid_0's binary_logloss: 0.136969\n",
            "[160]\tvalid_0's auc: 0.842668\tvalid_0's binary_logloss: 0.13692\n",
            "[161]\tvalid_0's auc: 0.842814\tvalid_0's binary_logloss: 0.136878\n",
            "[162]\tvalid_0's auc: 0.842895\tvalid_0's binary_logloss: 0.136838\n",
            "[163]\tvalid_0's auc: 0.842904\tvalid_0's binary_logloss: 0.136793\n",
            "[164]\tvalid_0's auc: 0.842994\tvalid_0's binary_logloss: 0.136753\n",
            "[165]\tvalid_0's auc: 0.842993\tvalid_0's binary_logloss: 0.136705\n",
            "[166]\tvalid_0's auc: 0.84312\tvalid_0's binary_logloss: 0.136666\n",
            "[167]\tvalid_0's auc: 0.843143\tvalid_0's binary_logloss: 0.136623\n",
            "[168]\tvalid_0's auc: 0.84311\tvalid_0's binary_logloss: 0.13659\n",
            "[169]\tvalid_0's auc: 0.843136\tvalid_0's binary_logloss: 0.136556\n",
            "[170]\tvalid_0's auc: 0.843127\tvalid_0's binary_logloss: 0.136525\n",
            "[171]\tvalid_0's auc: 0.843134\tvalid_0's binary_logloss: 0.136483\n",
            "[172]\tvalid_0's auc: 0.843073\tvalid_0's binary_logloss: 0.136456\n",
            "[173]\tvalid_0's auc: 0.843055\tvalid_0's binary_logloss: 0.136426\n",
            "[174]\tvalid_0's auc: 0.843048\tvalid_0's binary_logloss: 0.136377\n",
            "[175]\tvalid_0's auc: 0.843068\tvalid_0's binary_logloss: 0.136342\n",
            "[176]\tvalid_0's auc: 0.843092\tvalid_0's binary_logloss: 0.136309\n",
            "[177]\tvalid_0's auc: 0.84312\tvalid_0's binary_logloss: 0.136275\n",
            "[178]\tvalid_0's auc: 0.843245\tvalid_0's binary_logloss: 0.136229\n",
            "[179]\tvalid_0's auc: 0.8433\tvalid_0's binary_logloss: 0.136191\n",
            "[180]\tvalid_0's auc: 0.843341\tvalid_0's binary_logloss: 0.136144\n",
            "[181]\tvalid_0's auc: 0.843314\tvalid_0's binary_logloss: 0.136118\n",
            "[182]\tvalid_0's auc: 0.843373\tvalid_0's binary_logloss: 0.136086\n",
            "[183]\tvalid_0's auc: 0.843377\tvalid_0's binary_logloss: 0.136051\n",
            "[184]\tvalid_0's auc: 0.843491\tvalid_0's binary_logloss: 0.136014\n",
            "[185]\tvalid_0's auc: 0.843473\tvalid_0's binary_logloss: 0.135991\n",
            "[186]\tvalid_0's auc: 0.843464\tvalid_0's binary_logloss: 0.135957\n",
            "[187]\tvalid_0's auc: 0.843496\tvalid_0's binary_logloss: 0.135932\n",
            "[188]\tvalid_0's auc: 0.843559\tvalid_0's binary_logloss: 0.135909\n",
            "[189]\tvalid_0's auc: 0.843598\tvalid_0's binary_logloss: 0.135868\n",
            "[190]\tvalid_0's auc: 0.843606\tvalid_0's binary_logloss: 0.13583\n",
            "[191]\tvalid_0's auc: 0.843683\tvalid_0's binary_logloss: 0.135793\n",
            "[192]\tvalid_0's auc: 0.843639\tvalid_0's binary_logloss: 0.135763\n",
            "[193]\tvalid_0's auc: 0.843626\tvalid_0's binary_logloss: 0.135739\n",
            "[194]\tvalid_0's auc: 0.843733\tvalid_0's binary_logloss: 0.135708\n",
            "[195]\tvalid_0's auc: 0.843697\tvalid_0's binary_logloss: 0.135688\n",
            "[196]\tvalid_0's auc: 0.843703\tvalid_0's binary_logloss: 0.135662\n",
            "[197]\tvalid_0's auc: 0.843643\tvalid_0's binary_logloss: 0.135631\n",
            "[198]\tvalid_0's auc: 0.843632\tvalid_0's binary_logloss: 0.135602\n",
            "[199]\tvalid_0's auc: 0.843688\tvalid_0's binary_logloss: 0.135569\n",
            "[200]\tvalid_0's auc: 0.843723\tvalid_0's binary_logloss: 0.135549\n",
            "[201]\tvalid_0's auc: 0.843871\tvalid_0's binary_logloss: 0.135515\n",
            "[202]\tvalid_0's auc: 0.843887\tvalid_0's binary_logloss: 0.135487\n",
            "[203]\tvalid_0's auc: 0.843939\tvalid_0's binary_logloss: 0.135459\n",
            "[204]\tvalid_0's auc: 0.844009\tvalid_0's binary_logloss: 0.135431\n",
            "[205]\tvalid_0's auc: 0.843949\tvalid_0's binary_logloss: 0.135408\n",
            "[206]\tvalid_0's auc: 0.843938\tvalid_0's binary_logloss: 0.135381\n",
            "[207]\tvalid_0's auc: 0.843959\tvalid_0's binary_logloss: 0.135362\n",
            "[208]\tvalid_0's auc: 0.844037\tvalid_0's binary_logloss: 0.135336\n",
            "[209]\tvalid_0's auc: 0.844055\tvalid_0's binary_logloss: 0.135309\n",
            "[210]\tvalid_0's auc: 0.84401\tvalid_0's binary_logloss: 0.135293\n",
            "[211]\tvalid_0's auc: 0.844021\tvalid_0's binary_logloss: 0.135273\n",
            "[212]\tvalid_0's auc: 0.844014\tvalid_0's binary_logloss: 0.135253\n",
            "[213]\tvalid_0's auc: 0.843997\tvalid_0's binary_logloss: 0.135226\n",
            "[214]\tvalid_0's auc: 0.843992\tvalid_0's binary_logloss: 0.135209\n",
            "[215]\tvalid_0's auc: 0.843944\tvalid_0's binary_logloss: 0.135182\n",
            "[216]\tvalid_0's auc: 0.843913\tvalid_0's binary_logloss: 0.135169\n",
            "[217]\tvalid_0's auc: 0.843956\tvalid_0's binary_logloss: 0.135141\n",
            "[218]\tvalid_0's auc: 0.843962\tvalid_0's binary_logloss: 0.135117\n",
            "[219]\tvalid_0's auc: 0.843992\tvalid_0's binary_logloss: 0.135102\n",
            "[220]\tvalid_0's auc: 0.844059\tvalid_0's binary_logloss: 0.135077\n",
            "[221]\tvalid_0's auc: 0.844093\tvalid_0's binary_logloss: 0.135053\n",
            "[222]\tvalid_0's auc: 0.844123\tvalid_0's binary_logloss: 0.135033\n",
            "[223]\tvalid_0's auc: 0.844123\tvalid_0's binary_logloss: 0.135011\n",
            "[224]\tvalid_0's auc: 0.844179\tvalid_0's binary_logloss: 0.134996\n",
            "[225]\tvalid_0's auc: 0.844174\tvalid_0's binary_logloss: 0.134977\n",
            "[226]\tvalid_0's auc: 0.844119\tvalid_0's binary_logloss: 0.134963\n",
            "[227]\tvalid_0's auc: 0.844111\tvalid_0's binary_logloss: 0.134943\n",
            "[228]\tvalid_0's auc: 0.844184\tvalid_0's binary_logloss: 0.134925\n",
            "[229]\tvalid_0's auc: 0.844219\tvalid_0's binary_logloss: 0.134906\n",
            "[230]\tvalid_0's auc: 0.844229\tvalid_0's binary_logloss: 0.134882\n",
            "[231]\tvalid_0's auc: 0.844266\tvalid_0's binary_logloss: 0.134863\n",
            "[232]\tvalid_0's auc: 0.844328\tvalid_0's binary_logloss: 0.134849\n",
            "[233]\tvalid_0's auc: 0.844381\tvalid_0's binary_logloss: 0.134826\n",
            "[234]\tvalid_0's auc: 0.844342\tvalid_0's binary_logloss: 0.134811\n",
            "[235]\tvalid_0's auc: 0.844287\tvalid_0's binary_logloss: 0.134805\n",
            "[236]\tvalid_0's auc: 0.844364\tvalid_0's binary_logloss: 0.134782\n",
            "[237]\tvalid_0's auc: 0.844293\tvalid_0's binary_logloss: 0.134773\n",
            "[238]\tvalid_0's auc: 0.844338\tvalid_0's binary_logloss: 0.13475\n",
            "[239]\tvalid_0's auc: 0.844308\tvalid_0's binary_logloss: 0.134739\n",
            "[240]\tvalid_0's auc: 0.844333\tvalid_0's binary_logloss: 0.134719\n",
            "[241]\tvalid_0's auc: 0.844347\tvalid_0's binary_logloss: 0.134706\n",
            "[242]\tvalid_0's auc: 0.844377\tvalid_0's binary_logloss: 0.134687\n",
            "[243]\tvalid_0's auc: 0.844411\tvalid_0's binary_logloss: 0.134672\n",
            "[244]\tvalid_0's auc: 0.844462\tvalid_0's binary_logloss: 0.134645\n",
            "[245]\tvalid_0's auc: 0.844568\tvalid_0's binary_logloss: 0.134628\n",
            "[246]\tvalid_0's auc: 0.844544\tvalid_0's binary_logloss: 0.134612\n",
            "[247]\tvalid_0's auc: 0.844629\tvalid_0's binary_logloss: 0.134591\n",
            "[248]\tvalid_0's auc: 0.844617\tvalid_0's binary_logloss: 0.134578\n",
            "[249]\tvalid_0's auc: 0.844644\tvalid_0's binary_logloss: 0.134556\n",
            "[250]\tvalid_0's auc: 0.844671\tvalid_0's binary_logloss: 0.13454\n",
            "[251]\tvalid_0's auc: 0.844703\tvalid_0's binary_logloss: 0.134524\n",
            "[252]\tvalid_0's auc: 0.844722\tvalid_0's binary_logloss: 0.134507\n",
            "[253]\tvalid_0's auc: 0.844694\tvalid_0's binary_logloss: 0.134494\n",
            "[254]\tvalid_0's auc: 0.844662\tvalid_0's binary_logloss: 0.134481\n",
            "[255]\tvalid_0's auc: 0.844648\tvalid_0's binary_logloss: 0.134476\n",
            "[256]\tvalid_0's auc: 0.844696\tvalid_0's binary_logloss: 0.134454\n",
            "[257]\tvalid_0's auc: 0.844745\tvalid_0's binary_logloss: 0.134435\n",
            "[258]\tvalid_0's auc: 0.844745\tvalid_0's binary_logloss: 0.134419\n",
            "[259]\tvalid_0's auc: 0.844732\tvalid_0's binary_logloss: 0.134407\n",
            "[260]\tvalid_0's auc: 0.84476\tvalid_0's binary_logloss: 0.134392\n",
            "[261]\tvalid_0's auc: 0.844763\tvalid_0's binary_logloss: 0.134378\n",
            "[262]\tvalid_0's auc: 0.844717\tvalid_0's binary_logloss: 0.134371\n",
            "[263]\tvalid_0's auc: 0.844731\tvalid_0's binary_logloss: 0.134352\n",
            "[264]\tvalid_0's auc: 0.844729\tvalid_0's binary_logloss: 0.13434\n",
            "[265]\tvalid_0's auc: 0.844739\tvalid_0's binary_logloss: 0.134323\n",
            "[266]\tvalid_0's auc: 0.844767\tvalid_0's binary_logloss: 0.134309\n",
            "[267]\tvalid_0's auc: 0.844773\tvalid_0's binary_logloss: 0.134295\n",
            "[268]\tvalid_0's auc: 0.844783\tvalid_0's binary_logloss: 0.134275\n",
            "[269]\tvalid_0's auc: 0.844773\tvalid_0's binary_logloss: 0.134264\n",
            "[270]\tvalid_0's auc: 0.844746\tvalid_0's binary_logloss: 0.134256\n",
            "[271]\tvalid_0's auc: 0.84475\tvalid_0's binary_logloss: 0.134239\n",
            "[272]\tvalid_0's auc: 0.844758\tvalid_0's binary_logloss: 0.134224\n",
            "[273]\tvalid_0's auc: 0.844775\tvalid_0's binary_logloss: 0.134213\n",
            "[274]\tvalid_0's auc: 0.844837\tvalid_0's binary_logloss: 0.134204\n",
            "[275]\tvalid_0's auc: 0.844838\tvalid_0's binary_logloss: 0.13419\n",
            "[276]\tvalid_0's auc: 0.844908\tvalid_0's binary_logloss: 0.134171\n",
            "[277]\tvalid_0's auc: 0.844879\tvalid_0's binary_logloss: 0.134165\n",
            "[278]\tvalid_0's auc: 0.844849\tvalid_0's binary_logloss: 0.134168\n",
            "[279]\tvalid_0's auc: 0.844828\tvalid_0's binary_logloss: 0.134163\n",
            "[280]\tvalid_0's auc: 0.844774\tvalid_0's binary_logloss: 0.134155\n",
            "[281]\tvalid_0's auc: 0.844728\tvalid_0's binary_logloss: 0.134149\n",
            "[282]\tvalid_0's auc: 0.844801\tvalid_0's binary_logloss: 0.13413\n",
            "[283]\tvalid_0's auc: 0.844855\tvalid_0's binary_logloss: 0.134119\n",
            "[284]\tvalid_0's auc: 0.844888\tvalid_0's binary_logloss: 0.134104\n",
            "[285]\tvalid_0's auc: 0.844962\tvalid_0's binary_logloss: 0.134082\n",
            "[286]\tvalid_0's auc: 0.845016\tvalid_0's binary_logloss: 0.134065\n",
            "[287]\tvalid_0's auc: 0.845026\tvalid_0's binary_logloss: 0.134055\n",
            "[288]\tvalid_0's auc: 0.845054\tvalid_0's binary_logloss: 0.134042\n",
            "[289]\tvalid_0's auc: 0.844996\tvalid_0's binary_logloss: 0.134039\n",
            "[290]\tvalid_0's auc: 0.845004\tvalid_0's binary_logloss: 0.134032\n",
            "[291]\tvalid_0's auc: 0.845003\tvalid_0's binary_logloss: 0.13403\n",
            "[292]\tvalid_0's auc: 0.84507\tvalid_0's binary_logloss: 0.134012\n",
            "[293]\tvalid_0's auc: 0.844986\tvalid_0's binary_logloss: 0.134009\n",
            "[294]\tvalid_0's auc: 0.844966\tvalid_0's binary_logloss: 0.133999\n",
            "[295]\tvalid_0's auc: 0.844906\tvalid_0's binary_logloss: 0.133996\n",
            "[296]\tvalid_0's auc: 0.844949\tvalid_0's binary_logloss: 0.133989\n",
            "[297]\tvalid_0's auc: 0.844945\tvalid_0's binary_logloss: 0.133982\n",
            "[298]\tvalid_0's auc: 0.844997\tvalid_0's binary_logloss: 0.133966\n",
            "[299]\tvalid_0's auc: 0.844999\tvalid_0's binary_logloss: 0.13396\n",
            "[300]\tvalid_0's auc: 0.845035\tvalid_0's binary_logloss: 0.133945\n",
            "[301]\tvalid_0's auc: 0.845019\tvalid_0's binary_logloss: 0.133937\n",
            "[302]\tvalid_0's auc: 0.845084\tvalid_0's binary_logloss: 0.133922\n",
            "[303]\tvalid_0's auc: 0.845068\tvalid_0's binary_logloss: 0.133913\n",
            "[304]\tvalid_0's auc: 0.845056\tvalid_0's binary_logloss: 0.133912\n",
            "[305]\tvalid_0's auc: 0.845048\tvalid_0's binary_logloss: 0.133902\n",
            "[306]\tvalid_0's auc: 0.845117\tvalid_0's binary_logloss: 0.133886\n",
            "[307]\tvalid_0's auc: 0.845146\tvalid_0's binary_logloss: 0.133883\n",
            "[308]\tvalid_0's auc: 0.845201\tvalid_0's binary_logloss: 0.133867\n",
            "[309]\tvalid_0's auc: 0.845165\tvalid_0's binary_logloss: 0.133865\n",
            "[310]\tvalid_0's auc: 0.845186\tvalid_0's binary_logloss: 0.133853\n",
            "[311]\tvalid_0's auc: 0.845153\tvalid_0's binary_logloss: 0.133846\n",
            "[312]\tvalid_0's auc: 0.845216\tvalid_0's binary_logloss: 0.133828\n",
            "[313]\tvalid_0's auc: 0.845194\tvalid_0's binary_logloss: 0.133823\n",
            "[314]\tvalid_0's auc: 0.845208\tvalid_0's binary_logloss: 0.133814\n",
            "[315]\tvalid_0's auc: 0.84521\tvalid_0's binary_logloss: 0.133818\n",
            "[316]\tvalid_0's auc: 0.845282\tvalid_0's binary_logloss: 0.133801\n",
            "[317]\tvalid_0's auc: 0.845277\tvalid_0's binary_logloss: 0.133786\n",
            "[318]\tvalid_0's auc: 0.845295\tvalid_0's binary_logloss: 0.133776\n",
            "[319]\tvalid_0's auc: 0.845298\tvalid_0's binary_logloss: 0.133769\n",
            "[320]\tvalid_0's auc: 0.845315\tvalid_0's binary_logloss: 0.133761\n",
            "[321]\tvalid_0's auc: 0.84531\tvalid_0's binary_logloss: 0.13376\n",
            "[322]\tvalid_0's auc: 0.845328\tvalid_0's binary_logloss: 0.13375\n",
            "[323]\tvalid_0's auc: 0.845354\tvalid_0's binary_logloss: 0.133741\n",
            "[324]\tvalid_0's auc: 0.845277\tvalid_0's binary_logloss: 0.133742\n",
            "[325]\tvalid_0's auc: 0.845257\tvalid_0's binary_logloss: 0.133736\n",
            "[326]\tvalid_0's auc: 0.845252\tvalid_0's binary_logloss: 0.133732\n",
            "[327]\tvalid_0's auc: 0.845225\tvalid_0's binary_logloss: 0.133727\n",
            "[328]\tvalid_0's auc: 0.845233\tvalid_0's binary_logloss: 0.133728\n",
            "[329]\tvalid_0's auc: 0.845276\tvalid_0's binary_logloss: 0.133726\n",
            "[330]\tvalid_0's auc: 0.84527\tvalid_0's binary_logloss: 0.133717\n",
            "[331]\tvalid_0's auc: 0.845283\tvalid_0's binary_logloss: 0.133706\n",
            "[332]\tvalid_0's auc: 0.845248\tvalid_0's binary_logloss: 0.133704\n",
            "[333]\tvalid_0's auc: 0.845249\tvalid_0's binary_logloss: 0.133701\n",
            "[334]\tvalid_0's auc: 0.845234\tvalid_0's binary_logloss: 0.133693\n",
            "[335]\tvalid_0's auc: 0.845192\tvalid_0's binary_logloss: 0.133693\n",
            "[336]\tvalid_0's auc: 0.8452\tvalid_0's binary_logloss: 0.133685\n",
            "[337]\tvalid_0's auc: 0.845205\tvalid_0's binary_logloss: 0.133677\n",
            "[338]\tvalid_0's auc: 0.845236\tvalid_0's binary_logloss: 0.133668\n",
            "[339]\tvalid_0's auc: 0.845228\tvalid_0's binary_logloss: 0.133665\n",
            "[340]\tvalid_0's auc: 0.845136\tvalid_0's binary_logloss: 0.133662\n",
            "[341]\tvalid_0's auc: 0.845081\tvalid_0's binary_logloss: 0.133656\n",
            "[342]\tvalid_0's auc: 0.845131\tvalid_0's binary_logloss: 0.133638\n",
            "[343]\tvalid_0's auc: 0.845247\tvalid_0's binary_logloss: 0.133638\n",
            "[344]\tvalid_0's auc: 0.845255\tvalid_0's binary_logloss: 0.133632\n",
            "[345]\tvalid_0's auc: 0.845238\tvalid_0's binary_logloss: 0.133623\n",
            "[346]\tvalid_0's auc: 0.845268\tvalid_0's binary_logloss: 0.133623\n",
            "[347]\tvalid_0's auc: 0.845263\tvalid_0's binary_logloss: 0.133621\n",
            "[348]\tvalid_0's auc: 0.84528\tvalid_0's binary_logloss: 0.133612\n",
            "[349]\tvalid_0's auc: 0.845221\tvalid_0's binary_logloss: 0.133611\n",
            "[350]\tvalid_0's auc: 0.845255\tvalid_0's binary_logloss: 0.133596\n",
            "[351]\tvalid_0's auc: 0.845286\tvalid_0's binary_logloss: 0.133589\n",
            "[352]\tvalid_0's auc: 0.845285\tvalid_0's binary_logloss: 0.133584\n",
            "[353]\tvalid_0's auc: 0.8453\tvalid_0's binary_logloss: 0.133584\n",
            "[354]\tvalid_0's auc: 0.84529\tvalid_0's binary_logloss: 0.133585\n",
            "[355]\tvalid_0's auc: 0.845323\tvalid_0's binary_logloss: 0.133573\n",
            "[356]\tvalid_0's auc: 0.845365\tvalid_0's binary_logloss: 0.13356\n",
            "[357]\tvalid_0's auc: 0.84546\tvalid_0's binary_logloss: 0.133553\n",
            "[358]\tvalid_0's auc: 0.84548\tvalid_0's binary_logloss: 0.133538\n",
            "[359]\tvalid_0's auc: 0.845441\tvalid_0's binary_logloss: 0.133536\n",
            "[360]\tvalid_0's auc: 0.845409\tvalid_0's binary_logloss: 0.133538\n",
            "[361]\tvalid_0's auc: 0.845365\tvalid_0's binary_logloss: 0.133539\n",
            "[362]\tvalid_0's auc: 0.845361\tvalid_0's binary_logloss: 0.133538\n",
            "[363]\tvalid_0's auc: 0.845323\tvalid_0's binary_logloss: 0.133543\n",
            "[364]\tvalid_0's auc: 0.845279\tvalid_0's binary_logloss: 0.133542\n",
            "[365]\tvalid_0's auc: 0.845338\tvalid_0's binary_logloss: 0.133529\n",
            "[366]\tvalid_0's auc: 0.845388\tvalid_0's binary_logloss: 0.133526\n",
            "[367]\tvalid_0's auc: 0.845353\tvalid_0's binary_logloss: 0.133525\n",
            "[368]\tvalid_0's auc: 0.845354\tvalid_0's binary_logloss: 0.133517\n",
            "[369]\tvalid_0's auc: 0.845317\tvalid_0's binary_logloss: 0.133518\n",
            "[370]\tvalid_0's auc: 0.845275\tvalid_0's binary_logloss: 0.133513\n",
            "[371]\tvalid_0's auc: 0.845273\tvalid_0's binary_logloss: 0.133511\n",
            "[372]\tvalid_0's auc: 0.845268\tvalid_0's binary_logloss: 0.133506\n",
            "[373]\tvalid_0's auc: 0.845243\tvalid_0's binary_logloss: 0.133511\n",
            "[374]\tvalid_0's auc: 0.845196\tvalid_0's binary_logloss: 0.133517\n",
            "[375]\tvalid_0's auc: 0.845166\tvalid_0's binary_logloss: 0.133518\n",
            "[376]\tvalid_0's auc: 0.845322\tvalid_0's binary_logloss: 0.133503\n",
            "[377]\tvalid_0's auc: 0.845341\tvalid_0's binary_logloss: 0.133497\n",
            "[378]\tvalid_0's auc: 0.845326\tvalid_0's binary_logloss: 0.133497\n",
            "[379]\tvalid_0's auc: 0.84532\tvalid_0's binary_logloss: 0.133501\n",
            "[380]\tvalid_0's auc: 0.845316\tvalid_0's binary_logloss: 0.133489\n",
            "[381]\tvalid_0's auc: 0.845366\tvalid_0's binary_logloss: 0.133481\n",
            "[382]\tvalid_0's auc: 0.845327\tvalid_0's binary_logloss: 0.133483\n",
            "[383]\tvalid_0's auc: 0.845273\tvalid_0's binary_logloss: 0.133485\n",
            "[384]\tvalid_0's auc: 0.845249\tvalid_0's binary_logloss: 0.133481\n",
            "[385]\tvalid_0's auc: 0.845241\tvalid_0's binary_logloss: 0.133475\n",
            "[386]\tvalid_0's auc: 0.845297\tvalid_0's binary_logloss: 0.133468\n",
            "[387]\tvalid_0's auc: 0.845281\tvalid_0's binary_logloss: 0.133464\n",
            "[388]\tvalid_0's auc: 0.845333\tvalid_0's binary_logloss: 0.133461\n",
            "[389]\tvalid_0's auc: 0.845379\tvalid_0's binary_logloss: 0.133454\n",
            "[390]\tvalid_0's auc: 0.845361\tvalid_0's binary_logloss: 0.133456\n",
            "[391]\tvalid_0's auc: 0.845413\tvalid_0's binary_logloss: 0.133451\n",
            "[392]\tvalid_0's auc: 0.845372\tvalid_0's binary_logloss: 0.133446\n",
            "[393]\tvalid_0's auc: 0.845426\tvalid_0's binary_logloss: 0.13344\n",
            "[394]\tvalid_0's auc: 0.845391\tvalid_0's binary_logloss: 0.133439\n",
            "[395]\tvalid_0's auc: 0.84539\tvalid_0's binary_logloss: 0.133436\n",
            "[396]\tvalid_0's auc: 0.845332\tvalid_0's binary_logloss: 0.133437\n",
            "[397]\tvalid_0's auc: 0.84533\tvalid_0's binary_logloss: 0.133438\n",
            "[398]\tvalid_0's auc: 0.845313\tvalid_0's binary_logloss: 0.133434\n",
            "[399]\tvalid_0's auc: 0.8453\tvalid_0's binary_logloss: 0.133436\n",
            "[400]\tvalid_0's auc: 0.845301\tvalid_0's binary_logloss: 0.133441\n",
            "[401]\tvalid_0's auc: 0.845283\tvalid_0's binary_logloss: 0.133445\n",
            "[402]\tvalid_0's auc: 0.845361\tvalid_0's binary_logloss: 0.133434\n",
            "[403]\tvalid_0's auc: 0.845326\tvalid_0's binary_logloss: 0.133426\n",
            "[404]\tvalid_0's auc: 0.845288\tvalid_0's binary_logloss: 0.133421\n",
            "[405]\tvalid_0's auc: 0.845365\tvalid_0's binary_logloss: 0.133412\n",
            "[406]\tvalid_0's auc: 0.845377\tvalid_0's binary_logloss: 0.133407\n",
            "[407]\tvalid_0's auc: 0.8454\tvalid_0's binary_logloss: 0.133402\n",
            "[408]\tvalid_0's auc: 0.845399\tvalid_0's binary_logloss: 0.1334\n",
            "[409]\tvalid_0's auc: 0.845449\tvalid_0's binary_logloss: 0.133393\n",
            "[410]\tvalid_0's auc: 0.845419\tvalid_0's binary_logloss: 0.133391\n",
            "[411]\tvalid_0's auc: 0.845423\tvalid_0's binary_logloss: 0.133386\n",
            "[412]\tvalid_0's auc: 0.845474\tvalid_0's binary_logloss: 0.133383\n",
            "[413]\tvalid_0's auc: 0.845455\tvalid_0's binary_logloss: 0.133381\n",
            "[414]\tvalid_0's auc: 0.845437\tvalid_0's binary_logloss: 0.133375\n",
            "[415]\tvalid_0's auc: 0.845462\tvalid_0's binary_logloss: 0.133369\n",
            "[416]\tvalid_0's auc: 0.845483\tvalid_0's binary_logloss: 0.133368\n",
            "[417]\tvalid_0's auc: 0.845532\tvalid_0's binary_logloss: 0.13336\n",
            "[418]\tvalid_0's auc: 0.845551\tvalid_0's binary_logloss: 0.133358\n",
            "[419]\tvalid_0's auc: 0.845529\tvalid_0's binary_logloss: 0.13335\n",
            "[420]\tvalid_0's auc: 0.845573\tvalid_0's binary_logloss: 0.133346\n",
            "[421]\tvalid_0's auc: 0.845622\tvalid_0's binary_logloss: 0.13334\n",
            "[422]\tvalid_0's auc: 0.845653\tvalid_0's binary_logloss: 0.133339\n",
            "[423]\tvalid_0's auc: 0.845629\tvalid_0's binary_logloss: 0.13334\n",
            "[424]\tvalid_0's auc: 0.845593\tvalid_0's binary_logloss: 0.13334\n",
            "[425]\tvalid_0's auc: 0.845561\tvalid_0's binary_logloss: 0.133336\n",
            "[426]\tvalid_0's auc: 0.845523\tvalid_0's binary_logloss: 0.133337\n",
            "[427]\tvalid_0's auc: 0.845486\tvalid_0's binary_logloss: 0.133339\n",
            "[428]\tvalid_0's auc: 0.845494\tvalid_0's binary_logloss: 0.133342\n",
            "[429]\tvalid_0's auc: 0.845488\tvalid_0's binary_logloss: 0.133344\n",
            "[430]\tvalid_0's auc: 0.845499\tvalid_0's binary_logloss: 0.133346\n",
            "[431]\tvalid_0's auc: 0.845509\tvalid_0's binary_logloss: 0.133344\n",
            "[432]\tvalid_0's auc: 0.845485\tvalid_0's binary_logloss: 0.133342\n",
            "[433]\tvalid_0's auc: 0.845506\tvalid_0's binary_logloss: 0.133337\n",
            "[434]\tvalid_0's auc: 0.845489\tvalid_0's binary_logloss: 0.133342\n",
            "[435]\tvalid_0's auc: 0.845474\tvalid_0's binary_logloss: 0.133344\n",
            "[436]\tvalid_0's auc: 0.845549\tvalid_0's binary_logloss: 0.133334\n",
            "[437]\tvalid_0's auc: 0.845541\tvalid_0's binary_logloss: 0.133333\n",
            "[438]\tvalid_0's auc: 0.845511\tvalid_0's binary_logloss: 0.133331\n",
            "[439]\tvalid_0's auc: 0.845537\tvalid_0's binary_logloss: 0.133325\n",
            "[440]\tvalid_0's auc: 0.84555\tvalid_0's binary_logloss: 0.133324\n",
            "[441]\tvalid_0's auc: 0.845585\tvalid_0's binary_logloss: 0.13332\n",
            "[442]\tvalid_0's auc: 0.845572\tvalid_0's binary_logloss: 0.133321\n",
            "[443]\tvalid_0's auc: 0.845547\tvalid_0's binary_logloss: 0.13332\n",
            "[444]\tvalid_0's auc: 0.845521\tvalid_0's binary_logloss: 0.133324\n",
            "[445]\tvalid_0's auc: 0.845589\tvalid_0's binary_logloss: 0.133321\n",
            "[446]\tvalid_0's auc: 0.845576\tvalid_0's binary_logloss: 0.133314\n",
            "[447]\tvalid_0's auc: 0.845569\tvalid_0's binary_logloss: 0.133313\n",
            "CPU times: user 44.4 s, sys: 216 ms, total: 44.6 s\n",
            "Wall time: 23.4 s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=0.5,\n",
              "               gamma=5, importance_type='split', learning_rate=0.01,\n",
              "               max_depth=7, min_child_samples=20, min_child_weight=0.001,\n",
              "               min_split_gain=0.0, n_estimators=447, n_jobs=-1, nthread=-1,\n",
              "               num_leaves=31, objective='binary', random_state=None,\n",
              "               reg_alpha=0, reg_lambda=0.0, scale_pos_weight=1, silent=True,\n",
              "               subsample=1, subsample_for_bin=200000, subsample_freq=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "OFf2WKQC-V3R",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "outputId": "3280d7b8-b114-483e-8b79-05fc0b6dba52"
      },
      "source": [
        "y_pred = model.predict_proba(X_val_200)[:,1]\n",
        "label=\"LightGBM Classifer(top 250)\"\n",
        "labels.append(label)\n",
        "y_preds.append(y_pred)\n",
        "i+=1\n",
        "result = plot_auc(y_val,y_pred,label)\n",
        "result_df = result_df.append(result)\n",
        "del result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\t\t LightGBM Classifer(top 250) on Normal with Response Encoding dataset \t\t \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "The log loss and auc was found to be 0.133 and 0.846 for LightGBM Classifer(top 250) on Normal with Response Encoding dataset.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F8r0vIjs9i27",
        "colab_type": "text"
      },
      "source": [
        "# Combined"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bfDFKhQ9DajM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 518
        },
        "outputId": "0dd3602c-3185-4d72-a48d-7ed80ea5459e"
      },
      "source": [
        "plot_auc(y_val,y_preds,labels)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FWJqDqYVCUJv",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "outputId": "9b94c1d8-f282-4b89-8e05-7deaf33c9ccf"
      },
      "source": [
        "result_df.sort_values(by='auc',ascending=False)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Model</th>\n",
              "      <th>Dataset</th>\n",
              "      <th>log_loss</th>\n",
              "      <th>auc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>XGBoost Classifer(top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133050</td>\n",
              "      <td>0.845760</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>LightGBM Classifer(top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133313</td>\n",
              "      <td>0.845569</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>LightGBM Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.134520</td>\n",
              "      <td>0.845469</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost Classifer</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.133739</td>\n",
              "      <td>0.845052</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Random Forest Classifier (top 250)</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136440</td>\n",
              "      <td>0.837497</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Random Forest Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.136829</td>\n",
              "      <td>0.836315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Logisic Regression</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.523996</td>\n",
              "      <td>0.823515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Decision Tree Classifier</td>\n",
              "      <td>Normal with Response Encoding</td>\n",
              "      <td>0.149110</td>\n",
              "      <td>0.769599</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                Model  ...       auc\n",
              "6          XGBoost Classifer(top 250)  ...  0.845760\n",
              "7         LightGBM Classifer(top 250)  ...  0.845569\n",
              "4                  LightGBM Classifer  ...  0.845469\n",
              "3                   XGBoost Classifer  ...  0.845052\n",
              "5  Random Forest Classifier (top 250)  ...  0.837497\n",
              "2            Random Forest Classifier  ...  0.836315\n",
              "0                  Logisic Regression  ...  0.823515\n",
              "1            Decision Tree Classifier  ...  0.769599\n",
              "\n",
              "[8 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "lF-jgX2JfCuI"
      },
      "source": [
        "# Saving all the datasets created"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EH0NANEcmKmJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import joblib\n",
        "filename = 'normal_train_re.pkl'\n",
        "X_train.to_pickle(os.path.join(file_path,'best_models',filename))\n",
        "\n",
        "filename = 'normal_val_re.pkl'\n",
        "X_val.to_pickle(os.path.join(file_path,'best_models',filename))\n",
        "\n",
        "filename = 'normal_test_re.pkl'\n",
        "X_test.to_pickle(os.path.join(file_path,'best_models',filename))\n",
        "\n",
        "filename = 'normal_train_re_250.pkl'\n",
        "X_train_200.to_pickle(os.path.join(file_path,'best_models',filename))\n",
        "\n",
        "filename = 'normal_val_re_250.pkl'\n",
        "X_val_200.to_pickle(os.path.join(file_path,'best_models',filename))\n",
        "\n",
        "filename = 'normal_test_re_250.pkl'\n",
        "X_test_200.to_pickle(os.path.join(file_path,'best_models',filename))"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}