catboost.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "7kdWWf5v8JXe"
   },
   "outputs": [],
   "source": [
    "\n",
    "from catboost import CatBoostClassifier\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "otaInzLJ8Sih"
   },
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "X_train = pd.read_csv(\"/content/drive/My Drive/Colab Notebooks/train_Df64byy.csv\")\n",
    "X_test = pd.read_csv(\"/content/drive/My Drive/Colab Notebooks/test_YCcRUnU.csv\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "eRAK0rrvahGj"
   },
   "outputs": [],
   "source": [
    "y = X_train[\"Response\"]\n",
    "X_train = X_train.drop([\"ID\",\"Response\"],axis=1)\n",
    "\n",
    "X_test = X_test.drop([\"ID\"],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "K3iJVDm4bDiv",
    "outputId": "dba3a528-79b1-48b5-d579-b25dc4e13472"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((50882, 12), (21805, 12))"
      ]
     },
     "execution_count": 26,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "u6O3oGQubR0I",
    "outputId": "21b53745-b56b-4b87-d3b9-49002c61864e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(72687, 12)"
      ]
     },
     "execution_count": 27,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data = pd.concat([X_train, X_test]).reset_index(drop=True)\n",
    "all_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "gc9RzeW1icqO",
    "outputId": "e7cbc1cd-8144-4f29-9142-1ad881288878"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['City_Code', 'Region_Code', 'Accomodation_Type', 'Reco_Insurance_Type',\n",
       "       'Upper_Age', 'Lower_Age', 'Is_Spouse', 'Health Indicator',\n",
       "       'Holding_Policy_Duration', 'Holding_Policy_Type', 'Reco_Policy_Cat',\n",
       "       'Reco_Policy_Premium'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 28,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 280
    },
    "id": "eTFEGARiCqry",
    "outputId": "860de9d6-eeb2-4f34-f044-60ba988756a9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "ap_boxplot_bf = plt.subplot()\n",
    "ap_boxplot_bf = sns.boxplot(\n",
    "    x = 'Reco_Policy_Premium',\n",
    "    data = all_data\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "id": "CPLohlVQC4KO",
    "outputId": "24a68415-8752-45bd-f9fb-bddd563d0003"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f1f0ede58d0>"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# sns.catplot(x=\"Response\", y=\"Holding_Policy_Type\", data=all_data)\n",
    "# sns.countplot(x='Region_Code',data=all_data[all_data['City_Code']=='C15'])\n",
    "sns.boxplot(x='Reco_Policy_Cat',y='Reco_Policy_Premium',data=all_data,palette='rainbow')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "J01ClyMTDADs",
    "outputId": "f89d0a41-3617-41de-9db6-59b76beb822a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "City_Code                   0.000000\n",
      "Region_Code                 0.000000\n",
      "Accomodation_Type           0.000000\n",
      "Reco_Insurance_Type         0.000000\n",
      "Upper_Age                   0.000000\n",
      "Lower_Age                   0.000000\n",
      "Is_Spouse                   0.000000\n",
      "Health Indicator           22.999986\n",
      "Holding_Policy_Duration    39.696232\n",
      "Holding_Policy_Type        39.696232\n",
      "Reco_Policy_Cat             0.000000\n",
      "Reco_Policy_Premium         0.000000\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "all_data_na = (all_data.isnull().sum() / len(all_data)) *100\n",
    "print(all_data_na)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "NQ4dyEi1B3so"
   },
   "outputs": [],
   "source": [
    "\n",
    "def custom_rating(dur):\n",
    "    if dur==\"14+\":\n",
    "        return int(5)\n",
    "    elif int(float(dur))==0:\n",
    "        return 0\n",
    "    elif int(float(dur))>=1 and int(float(dur))<4:\n",
    "        return 1\n",
    "    elif int(float(dur))>=4 and int(float(dur))<8:\n",
    "        return 2\n",
    "    elif int(float(dur))>=8 and int(float(dur))<11:\n",
    "        return 3\n",
    "    elif int(float(dur))>=11 and int(float(dur))<=14:\n",
    "        return 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "uG0tEJPxfrYH"
   },
   "outputs": [],
   "source": [
    "#fill null values\n",
    "all_data[\"Health Indicator\"] = all_data[\"Health Indicator\"].fillna(\"None\")\n",
    "all_data[\"Holding_Policy_Duration\"] = all_data[\"Holding_Policy_Duration\"].fillna(0)\n",
    "all_data['Holding_Policy_Duration'] = all_data['Holding_Policy_Duration'].apply(lambda x: custom_rating(x)).astype(str)\n",
    "all_data[\"Holding_Policy_Type\"] = all_data[\"Holding_Policy_Type\"].fillna(\"None\").astype(str)\n",
    "\n",
    "#convert into categorical values\n",
    "all_data[\"Reco_Policy_Cat\"] = all_data[\"Reco_Policy_Cat\"].astype(str)\n",
    "all_data[\"Region_Code\"] = all_data[\"Region_Code\"].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "r7E51BBoDQCG"
   },
   "outputs": [],
   "source": [
    "#Reco policy premium is taken as per person in new column(assuming 2 person in joint)\n",
    "all_data['Reco_Policy_Premium_original'] = all_data['Reco_Policy_Premium']\n",
    "all_data.loc[all_data['Reco_Insurance_Type']=='Joint','Reco_Policy_Premium'] = all_data[all_data['Reco_Insurance_Type']=='Joint']['Reco_Policy_Premium'].apply(lambda x: x/2.0 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "AM29rzNCjboC",
    "outputId": "1b927036-a99e-426c-97f9-dc028074e214"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "City_Code\n",
      "Region_Code\n",
      "Accomodation_Type\n",
      "Reco_Insurance_Type\n",
      "Is_Spouse\n",
      "Health Indicator\n",
      "Holding_Policy_Duration\n",
      "Holding_Policy_Type\n",
      "Reco_Policy_Cat\n"
     ]
    }
   ],
   "source": [
    "\n",
    "cat_columns=[]\n",
    "for col in all_data.select_dtypes('object').columns:\n",
    "  print(col)\n",
    "  cat_columns.append(col)\n",
    "  le = LabelEncoder()\n",
    "  all_data[col] = le.fit_transform(all_data[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "ImfyAIRAbivL"
   },
   "outputs": [],
   "source": [
    "X_test_final = all_data[len(y):]\n",
    "all_data = all_data[:len(y)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DhnBa6bzbh7q",
    "outputId": "e55936d2-568e-48cd-cd39-134b974a071f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((50882, 13), (21805, 12))"
      ]
     },
     "execution_count": 37,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data.shape,X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "TwWJcMV08zwn",
    "outputId": "de5d9863-2c5b-4044-8b6b-723cd07d851b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['City_Code', 'Region_Code', 'Accomodation_Type', 'Reco_Insurance_Type',\n",
       "       'Upper_Age', 'Lower_Age', 'Is_Spouse', 'Health Indicator',\n",
       "       'Holding_Policy_Duration', 'Holding_Policy_Type', 'Reco_Policy_Cat',\n",
       "       'Reco_Policy_Premium', 'Reco_Policy_Premium_original'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 38,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "iOYda0OD9BOw",
    "outputId": "94aef526-f33b-4281-c505-a0fe3d7541d6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(50882, 13) Index(['City_Code', 'Region_Code', 'Accomodation_Type', 'Reco_Insurance_Type',\n",
      "       'Upper_Age', 'Lower_Age', 'Is_Spouse', 'Health Indicator',\n",
      "       'Holding_Policy_Duration', 'Holding_Policy_Type', 'Reco_Policy_Cat',\n",
      "       'Reco_Policy_Premium', 'Reco_Policy_Premium_original'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(all_data.shape,all_data.columns)\n",
    "\n",
    "\n",
    "X_train,X_test,y_train,y_test = train_test_split(all_data,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "CU9UO647n9mk",
    "outputId": "6476e5e8-683b-42e0-aa8a-3f2298257bc4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 1, 2, 3, 6, 7, 8, 9, 10]"
      ]
     },
     "execution_count": 40,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_features_index = [i for i,col in enumerate(all_data.columns) if col in cat_columns]\n",
    "cat_features_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "id": "O0iOuu-C9Er2"
   },
   "outputs": [],
   "source": [
    "# model = CatBoostClassifier(iterations=100,learning_rate=0.1,class_weights=[0.2, 0.8],'eval_metric': 'AUC')\n",
    "model = CatBoostClassifier(iterations=1000,eval_metric= 'AUC',early_stopping_rounds=300,cat_features = cat_features_index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HpNIqJve9Mbc",
    "outputId": "3bfaf160-5f81-460c-c054-6e8e9dc6475f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Learning rate set to 0.050148\n",
      "0:\ttotal: 125ms\tremaining: 2m 4s\n",
      "1:\ttotal: 187ms\tremaining: 1m 33s\n",
      "2:\ttotal: 245ms\tremaining: 1m 21s\n",
      "3:\ttotal: 311ms\tremaining: 1m 17s\n",
      "4:\ttotal: 375ms\tremaining: 1m 14s\n",
      "5:\ttotal: 393ms\tremaining: 1m 5s\n",
      "6:\ttotal: 435ms\tremaining: 1m 1s\n",
      "7:\ttotal: 490ms\tremaining: 1m\n",
      "8:\ttotal: 543ms\tremaining: 59.8s\n",
      "9:\ttotal: 590ms\tremaining: 58.4s\n",
      "10:\ttotal: 609ms\tremaining: 54.8s\n",
      "11:\ttotal: 661ms\tremaining: 54.4s\n",
      "12:\ttotal: 725ms\tremaining: 55.1s\n",
      "13:\ttotal: 777ms\tremaining: 54.7s\n",
      "14:\ttotal: 835ms\tremaining: 54.8s\n",
      "15:\ttotal: 905ms\tremaining: 55.7s\n",
      "16:\ttotal: 957ms\tremaining: 55.3s\n",
      "17:\ttotal: 988ms\tremaining: 53.9s\n",
      "18:\ttotal: 1.04s\tremaining: 53.6s\n",
      "19:\ttotal: 1.09s\tremaining: 53.3s\n",
      "20:\ttotal: 1.11s\tremaining: 51.7s\n",
      "21:\ttotal: 1.16s\tremaining: 51.8s\n",
      "22:\ttotal: 1.24s\tremaining: 52.6s\n",
      "23:\ttotal: 1.29s\tremaining: 52.4s\n",
      "24:\ttotal: 1.35s\tremaining: 52.6s\n",
      "25:\ttotal: 1.41s\tremaining: 52.9s\n",
      "26:\ttotal: 1.47s\tremaining: 52.9s\n",
      "27:\ttotal: 1.52s\tremaining: 52.7s\n",
      "28:\ttotal: 1.56s\tremaining: 52.3s\n",
      "29:\ttotal: 1.63s\tremaining: 52.6s\n",
      "30:\ttotal: 1.68s\tremaining: 52.5s\n",
      "31:\ttotal: 1.75s\tremaining: 52.9s\n",
      "32:\ttotal: 1.8s\tremaining: 52.7s\n",
      "33:\ttotal: 1.86s\tremaining: 52.8s\n",
      "34:\ttotal: 1.91s\tremaining: 52.6s\n",
      "35:\ttotal: 1.97s\tremaining: 52.7s\n",
      "36:\ttotal: 2.02s\tremaining: 52.5s\n",
      "37:\ttotal: 2.07s\tremaining: 52.4s\n",
      "38:\ttotal: 2.12s\tremaining: 52.2s\n",
      "39:\ttotal: 2.14s\tremaining: 51.4s\n",
      "40:\ttotal: 2.19s\tremaining: 51.3s\n",
      "41:\ttotal: 2.24s\tremaining: 51.1s\n",
      "42:\ttotal: 2.3s\tremaining: 51.3s\n",
      "43:\ttotal: 2.36s\tremaining: 51.2s\n",
      "44:\ttotal: 2.4s\tremaining: 51s\n",
      "45:\ttotal: 2.45s\tremaining: 50.9s\n",
      "46:\ttotal: 2.5s\tremaining: 50.8s\n",
      "47:\ttotal: 2.56s\tremaining: 50.7s\n",
      "48:\ttotal: 2.61s\tremaining: 50.6s\n",
      "49:\ttotal: 2.66s\tremaining: 50.5s\n",
      "50:\ttotal: 2.72s\tremaining: 50.6s\n",
      "51:\ttotal: 2.76s\tremaining: 50.4s\n",
      "52:\ttotal: 2.83s\tremaining: 50.6s\n",
      "53:\ttotal: 2.92s\tremaining: 51.2s\n",
      "54:\ttotal: 3s\tremaining: 51.6s\n",
      "55:\ttotal: 3.07s\tremaining: 51.8s\n",
      "56:\ttotal: 3.13s\tremaining: 51.8s\n",
      "57:\ttotal: 3.19s\tremaining: 51.7s\n",
      "58:\ttotal: 3.25s\tremaining: 51.8s\n",
      "59:\ttotal: 3.29s\tremaining: 51.6s\n",
      "60:\ttotal: 3.35s\tremaining: 51.5s\n",
      "61:\ttotal: 3.39s\tremaining: 51.4s\n",
      "62:\ttotal: 3.45s\tremaining: 51.3s\n",
      "63:\ttotal: 3.5s\tremaining: 51.2s\n",
      "64:\ttotal: 3.57s\tremaining: 51.3s\n",
      "65:\ttotal: 3.62s\tremaining: 51.2s\n",
      "66:\ttotal: 3.68s\tremaining: 51.3s\n",
      "67:\ttotal: 3.75s\tremaining: 51.4s\n",
      "68:\ttotal: 3.81s\tremaining: 51.4s\n",
      "69:\ttotal: 3.86s\tremaining: 51.2s\n",
      "70:\ttotal: 3.9s\tremaining: 51.1s\n",
      "71:\ttotal: 3.96s\tremaining: 51s\n",
      "72:\ttotal: 4.03s\tremaining: 51.2s\n",
      "73:\ttotal: 4.08s\tremaining: 51.1s\n",
      "74:\ttotal: 4.16s\tremaining: 51.3s\n",
      "75:\ttotal: 4.21s\tremaining: 51.2s\n",
      "76:\ttotal: 4.26s\tremaining: 51.1s\n",
      "77:\ttotal: 4.32s\tremaining: 51s\n",
      "78:\ttotal: 4.37s\tremaining: 50.9s\n",
      "79:\ttotal: 4.42s\tremaining: 50.8s\n",
      "80:\ttotal: 4.49s\tremaining: 50.9s\n",
      "81:\ttotal: 4.56s\tremaining: 51.1s\n",
      "82:\ttotal: 4.61s\tremaining: 51s\n",
      "83:\ttotal: 4.66s\tremaining: 50.9s\n",
      "84:\ttotal: 4.72s\tremaining: 50.9s\n",
      "85:\ttotal: 4.79s\tremaining: 50.9s\n",
      "86:\ttotal: 4.86s\tremaining: 51s\n",
      "87:\ttotal: 4.92s\tremaining: 51s\n",
      "88:\ttotal: 4.99s\tremaining: 51s\n",
      "89:\ttotal: 5.02s\tremaining: 50.8s\n",
      "90:\ttotal: 5.08s\tremaining: 50.8s\n",
      "91:\ttotal: 5.13s\tremaining: 50.6s\n",
      "92:\ttotal: 5.18s\tremaining: 50.6s\n",
      "93:\ttotal: 5.24s\tremaining: 50.5s\n",
      "94:\ttotal: 5.3s\tremaining: 50.5s\n",
      "95:\ttotal: 5.36s\tremaining: 50.4s\n",
      "96:\ttotal: 5.42s\tremaining: 50.4s\n",
      "97:\ttotal: 5.47s\tremaining: 50.3s\n",
      "98:\ttotal: 5.51s\tremaining: 50.2s\n",
      "99:\ttotal: 5.57s\tremaining: 50.2s\n",
      "100:\ttotal: 5.63s\tremaining: 50.1s\n",
      "101:\ttotal: 5.69s\tremaining: 50.1s\n",
      "102:\ttotal: 5.77s\tremaining: 50.2s\n",
      "103:\ttotal: 5.83s\tremaining: 50.3s\n",
      "104:\ttotal: 5.89s\tremaining: 50.2s\n",
      "105:\ttotal: 5.95s\tremaining: 50.2s\n",
      "106:\ttotal: 6.01s\tremaining: 50.2s\n",
      "107:\ttotal: 6.08s\tremaining: 50.2s\n",
      "108:\ttotal: 6.11s\tremaining: 49.9s\n",
      "109:\ttotal: 6.16s\tremaining: 49.8s\n",
      "110:\ttotal: 6.22s\tremaining: 49.8s\n",
      "111:\ttotal: 6.29s\tremaining: 49.9s\n",
      "112:\ttotal: 6.35s\tremaining: 49.8s\n",
      "113:\ttotal: 6.39s\tremaining: 49.7s\n",
      "114:\ttotal: 6.46s\tremaining: 49.7s\n",
      "115:\ttotal: 6.54s\tremaining: 49.8s\n",
      "116:\ttotal: 6.59s\tremaining: 49.8s\n",
      "117:\ttotal: 6.65s\tremaining: 49.7s\n",
      "118:\ttotal: 6.71s\tremaining: 49.7s\n",
      "119:\ttotal: 6.79s\tremaining: 49.8s\n",
      "120:\ttotal: 6.87s\tremaining: 49.9s\n",
      "121:\ttotal: 6.93s\tremaining: 49.9s\n",
      "122:\ttotal: 7.01s\tremaining: 50s\n",
      "123:\ttotal: 7.08s\tremaining: 50s\n",
      "124:\ttotal: 7.16s\tremaining: 50.1s\n",
      "125:\ttotal: 7.22s\tremaining: 50.1s\n",
      "126:\ttotal: 7.29s\tremaining: 50.1s\n",
      "127:\ttotal: 7.35s\tremaining: 50s\n",
      "128:\ttotal: 7.41s\tremaining: 50s\n",
      "129:\ttotal: 7.47s\tremaining: 50s\n",
      "130:\ttotal: 7.54s\tremaining: 50s\n",
      "131:\ttotal: 7.62s\tremaining: 50.1s\n",
      "132:\ttotal: 7.69s\tremaining: 50.1s\n",
      "133:\ttotal: 7.75s\tremaining: 50.1s\n",
      "134:\ttotal: 7.81s\tremaining: 50s\n",
      "135:\ttotal: 7.89s\tremaining: 50.1s\n",
      "136:\ttotal: 7.92s\tremaining: 49.9s\n",
      "137:\ttotal: 8s\tremaining: 50s\n",
      "138:\ttotal: 8.07s\tremaining: 50s\n",
      "139:\ttotal: 8.12s\tremaining: 49.9s\n",
      "140:\ttotal: 8.18s\tremaining: 49.8s\n",
      "141:\ttotal: 8.26s\tremaining: 49.9s\n",
      "142:\ttotal: 8.32s\tremaining: 49.9s\n",
      "143:\ttotal: 8.39s\tremaining: 49.9s\n",
      "144:\ttotal: 8.46s\tremaining: 49.9s\n",
      "145:\ttotal: 8.53s\tremaining: 49.9s\n",
      "146:\ttotal: 8.61s\tremaining: 50s\n",
      "147:\ttotal: 8.68s\tremaining: 50s\n",
      "148:\ttotal: 8.73s\tremaining: 49.9s\n",
      "149:\ttotal: 8.8s\tremaining: 49.9s\n",
      "150:\ttotal: 8.86s\tremaining: 49.8s\n",
      "151:\ttotal: 8.93s\tremaining: 49.8s\n",
      "152:\ttotal: 9.01s\tremaining: 49.9s\n",
      "153:\ttotal: 9.08s\tremaining: 49.9s\n",
      "154:\ttotal: 9.15s\tremaining: 49.9s\n",
      "155:\ttotal: 9.22s\tremaining: 49.9s\n",
      "156:\ttotal: 9.24s\tremaining: 49.6s\n",
      "157:\ttotal: 9.3s\tremaining: 49.6s\n",
      "158:\ttotal: 9.37s\tremaining: 49.6s\n",
      "159:\ttotal: 9.43s\tremaining: 49.5s\n",
      "160:\ttotal: 9.49s\tremaining: 49.5s\n",
      "161:\ttotal: 9.58s\tremaining: 49.5s\n",
      "162:\ttotal: 9.64s\tremaining: 49.5s\n",
      "163:\ttotal: 9.71s\tremaining: 49.5s\n",
      "164:\ttotal: 9.8s\tremaining: 49.6s\n",
      "165:\ttotal: 9.87s\tremaining: 49.6s\n",
      "166:\ttotal: 9.93s\tremaining: 49.6s\n",
      "167:\ttotal: 9.99s\tremaining: 49.5s\n",
      "168:\ttotal: 10.1s\tremaining: 49.5s\n",
      "169:\ttotal: 10.1s\tremaining: 49.4s\n",
      "170:\ttotal: 10.2s\tremaining: 49.3s\n",
      "171:\ttotal: 10.2s\tremaining: 49.3s\n",
      "172:\ttotal: 10.3s\tremaining: 49.3s\n",
      "173:\ttotal: 10.4s\tremaining: 49.2s\n",
      "174:\ttotal: 10.4s\tremaining: 49.1s\n",
      "175:\ttotal: 10.5s\tremaining: 49.1s\n",
      "176:\ttotal: 10.6s\tremaining: 49.1s\n",
      "177:\ttotal: 10.6s\tremaining: 48.9s\n",
      "178:\ttotal: 10.6s\tremaining: 48.8s\n",
      "179:\ttotal: 10.7s\tremaining: 48.8s\n",
      "180:\ttotal: 10.8s\tremaining: 48.7s\n",
      "181:\ttotal: 10.9s\tremaining: 48.8s\n",
      "182:\ttotal: 10.9s\tremaining: 48.8s\n",
      "183:\ttotal: 11s\tremaining: 48.9s\n",
      "184:\ttotal: 11.1s\tremaining: 48.9s\n",
      "185:\ttotal: 11.2s\tremaining: 48.8s\n",
      "186:\ttotal: 11.2s\tremaining: 48.8s\n",
      "187:\ttotal: 11.3s\tremaining: 48.8s\n",
      "188:\ttotal: 11.4s\tremaining: 48.9s\n",
      "189:\ttotal: 11.4s\tremaining: 48.8s\n",
      "190:\ttotal: 11.5s\tremaining: 48.8s\n",
      "191:\ttotal: 11.6s\tremaining: 48.8s\n",
      "192:\ttotal: 11.7s\tremaining: 48.8s\n",
      "193:\ttotal: 11.7s\tremaining: 48.8s\n",
      "194:\ttotal: 11.9s\tremaining: 48.9s\n",
      "195:\ttotal: 11.9s\tremaining: 48.9s\n",
      "196:\ttotal: 12s\tremaining: 48.9s\n",
      "197:\ttotal: 12.1s\tremaining: 48.9s\n",
      "198:\ttotal: 12.1s\tremaining: 48.9s\n",
      "199:\ttotal: 12.2s\tremaining: 48.9s\n",
      "200:\ttotal: 12.3s\tremaining: 48.9s\n",
      "201:\ttotal: 12.4s\tremaining: 48.9s\n",
      "202:\ttotal: 12.4s\tremaining: 48.8s\n",
      "203:\ttotal: 12.5s\tremaining: 48.8s\n",
      "204:\ttotal: 12.6s\tremaining: 48.7s\n",
      "205:\ttotal: 12.7s\tremaining: 48.8s\n",
      "206:\ttotal: 12.7s\tremaining: 48.7s\n",
      "207:\ttotal: 12.8s\tremaining: 48.8s\n",
      "208:\ttotal: 12.9s\tremaining: 48.8s\n",
      "209:\ttotal: 13s\tremaining: 48.7s\n",
      "210:\ttotal: 13s\tremaining: 48.6s\n",
      "211:\ttotal: 13.1s\tremaining: 48.6s\n",
      "212:\ttotal: 13.2s\tremaining: 48.7s\n",
      "213:\ttotal: 13.2s\tremaining: 48.6s\n",
      "214:\ttotal: 13.3s\tremaining: 48.6s\n",
      "215:\ttotal: 13.4s\tremaining: 48.6s\n",
      "216:\ttotal: 13.5s\tremaining: 48.6s\n",
      "217:\ttotal: 13.5s\tremaining: 48.5s\n",
      "218:\ttotal: 13.6s\tremaining: 48.4s\n",
      "219:\ttotal: 13.7s\tremaining: 48.4s\n",
      "220:\ttotal: 13.7s\tremaining: 48.3s\n",
      "221:\ttotal: 13.8s\tremaining: 48.3s\n",
      "222:\ttotal: 13.8s\tremaining: 48.2s\n",
      "223:\ttotal: 13.9s\tremaining: 48.2s\n",
      "224:\ttotal: 14s\tremaining: 48.2s\n",
      "225:\ttotal: 14s\tremaining: 48.1s\n",
      "226:\ttotal: 14.1s\tremaining: 48.1s\n",
      "227:\ttotal: 14.2s\tremaining: 48.1s\n",
      "228:\ttotal: 14.3s\tremaining: 48.1s\n",
      "229:\ttotal: 14.3s\tremaining: 48s\n",
      "230:\ttotal: 14.4s\tremaining: 48s\n",
      "231:\ttotal: 14.5s\tremaining: 47.9s\n",
      "232:\ttotal: 14.6s\tremaining: 47.9s\n",
      "233:\ttotal: 14.6s\tremaining: 47.9s\n",
      "234:\ttotal: 14.7s\tremaining: 47.9s\n",
      "235:\ttotal: 14.8s\tremaining: 47.9s\n",
      "236:\ttotal: 14.9s\tremaining: 47.8s\n",
      "237:\ttotal: 14.9s\tremaining: 47.8s\n",
      "238:\ttotal: 15s\tremaining: 47.8s\n",
      "239:\ttotal: 15.1s\tremaining: 47.7s\n",
      "240:\ttotal: 15.1s\tremaining: 47.7s\n",
      "241:\ttotal: 15.2s\tremaining: 47.6s\n",
      "242:\ttotal: 15.3s\tremaining: 47.6s\n",
      "243:\ttotal: 15.3s\tremaining: 47.5s\n",
      "244:\ttotal: 15.4s\tremaining: 47.4s\n",
      "245:\ttotal: 15.5s\tremaining: 47.4s\n",
      "246:\ttotal: 15.5s\tremaining: 47.3s\n",
      "247:\ttotal: 15.6s\tremaining: 47.2s\n",
      "248:\ttotal: 15.6s\tremaining: 47.2s\n",
      "249:\ttotal: 15.7s\tremaining: 47.2s\n",
      "250:\ttotal: 15.8s\tremaining: 47.2s\n",
      "251:\ttotal: 15.9s\tremaining: 47.1s\n",
      "252:\ttotal: 16s\tremaining: 47.1s\n",
      "253:\ttotal: 16s\tremaining: 47.1s\n",
      "254:\ttotal: 16.1s\tremaining: 47s\n",
      "255:\ttotal: 16.2s\tremaining: 47s\n",
      "256:\ttotal: 16.2s\tremaining: 46.9s\n",
      "257:\ttotal: 16.3s\tremaining: 46.9s\n",
      "258:\ttotal: 16.4s\tremaining: 46.9s\n",
      "259:\ttotal: 16.4s\tremaining: 46.8s\n",
      "260:\ttotal: 16.5s\tremaining: 46.7s\n",
      "261:\ttotal: 16.6s\tremaining: 46.7s\n",
      "262:\ttotal: 16.6s\tremaining: 46.6s\n",
      "263:\ttotal: 16.7s\tremaining: 46.5s\n",
      "264:\ttotal: 16.8s\tremaining: 46.5s\n",
      "265:\ttotal: 16.8s\tremaining: 46.4s\n",
      "266:\ttotal: 16.9s\tremaining: 46.4s\n",
      "267:\ttotal: 17s\tremaining: 46.4s\n",
      "268:\ttotal: 17s\tremaining: 46.3s\n",
      "269:\ttotal: 17.1s\tremaining: 46.3s\n",
      "270:\ttotal: 17.2s\tremaining: 46.3s\n",
      "271:\ttotal: 17.3s\tremaining: 46.2s\n",
      "272:\ttotal: 17.3s\tremaining: 46.1s\n",
      "273:\ttotal: 17.4s\tremaining: 46.1s\n",
      "274:\ttotal: 17.5s\tremaining: 46.1s\n",
      "275:\ttotal: 17.5s\tremaining: 46s\n",
      "276:\ttotal: 17.6s\tremaining: 46s\n",
      "277:\ttotal: 17.7s\tremaining: 46s\n",
      "278:\ttotal: 17.8s\tremaining: 46s\n",
      "279:\ttotal: 17.9s\tremaining: 45.9s\n",
      "280:\ttotal: 17.9s\tremaining: 45.9s\n",
      "281:\ttotal: 18s\tremaining: 45.8s\n",
      "282:\ttotal: 18.1s\tremaining: 45.8s\n",
      "283:\ttotal: 18.2s\tremaining: 45.8s\n",
      "284:\ttotal: 18.2s\tremaining: 45.8s\n",
      "285:\ttotal: 18.3s\tremaining: 45.7s\n",
      "286:\ttotal: 18.4s\tremaining: 45.7s\n",
      "287:\ttotal: 18.5s\tremaining: 45.7s\n",
      "288:\ttotal: 18.6s\tremaining: 45.7s\n",
      "289:\ttotal: 18.7s\tremaining: 45.7s\n",
      "290:\ttotal: 18.7s\tremaining: 45.6s\n",
      "291:\ttotal: 18.8s\tremaining: 45.6s\n",
      "292:\ttotal: 18.9s\tremaining: 45.6s\n",
      "293:\ttotal: 19s\tremaining: 45.6s\n",
      "294:\ttotal: 19.1s\tremaining: 45.5s\n",
      "295:\ttotal: 19.1s\tremaining: 45.5s\n",
      "296:\ttotal: 19.2s\tremaining: 45.4s\n",
      "297:\ttotal: 19.3s\tremaining: 45.4s\n",
      "298:\ttotal: 19.4s\tremaining: 45.4s\n",
      "299:\ttotal: 19.4s\tremaining: 45.3s\n",
      "300:\ttotal: 19.5s\tremaining: 45.2s\n",
      "301:\ttotal: 19.5s\tremaining: 45.2s\n",
      "302:\ttotal: 19.6s\tremaining: 45.1s\n",
      "303:\ttotal: 19.7s\tremaining: 45.1s\n",
      "304:\ttotal: 19.8s\tremaining: 45.1s\n",
      "305:\ttotal: 19.8s\tremaining: 45s\n",
      "306:\ttotal: 19.9s\tremaining: 44.9s\n",
      "307:\ttotal: 20s\tremaining: 44.9s\n",
      "308:\ttotal: 20.1s\tremaining: 44.9s\n",
      "309:\ttotal: 20.1s\tremaining: 44.8s\n",
      "310:\ttotal: 20.2s\tremaining: 44.8s\n",
      "311:\ttotal: 20.3s\tremaining: 44.7s\n",
      "312:\ttotal: 20.3s\tremaining: 44.7s\n",
      "313:\ttotal: 20.4s\tremaining: 44.6s\n",
      "314:\ttotal: 20.5s\tremaining: 44.6s\n",
      "315:\ttotal: 20.6s\tremaining: 44.5s\n",
      "316:\ttotal: 20.7s\tremaining: 44.5s\n",
      "317:\ttotal: 20.7s\tremaining: 44.5s\n",
      "318:\ttotal: 20.8s\tremaining: 44.4s\n",
      "319:\ttotal: 20.9s\tremaining: 44.4s\n",
      "320:\ttotal: 21s\tremaining: 44.4s\n",
      "321:\ttotal: 21s\tremaining: 44.3s\n",
      "322:\ttotal: 21.1s\tremaining: 44.3s\n",
      "323:\ttotal: 21.2s\tremaining: 44.2s\n",
      "324:\ttotal: 21.3s\tremaining: 44.2s\n",
      "325:\ttotal: 21.3s\tremaining: 44.1s\n",
      "326:\ttotal: 21.4s\tremaining: 44s\n",
      "327:\ttotal: 21.4s\tremaining: 43.9s\n",
      "328:\ttotal: 21.5s\tremaining: 43.9s\n",
      "329:\ttotal: 21.6s\tremaining: 43.8s\n",
      "330:\ttotal: 21.6s\tremaining: 43.7s\n",
      "331:\ttotal: 21.7s\tremaining: 43.7s\n",
      "332:\ttotal: 21.8s\tremaining: 43.6s\n",
      "333:\ttotal: 21.9s\tremaining: 43.6s\n",
      "334:\ttotal: 21.9s\tremaining: 43.6s\n",
      "335:\ttotal: 22s\tremaining: 43.5s\n",
      "336:\ttotal: 22.1s\tremaining: 43.5s\n",
      "337:\ttotal: 22.2s\tremaining: 43.5s\n",
      "338:\ttotal: 22.3s\tremaining: 43.4s\n",
      "339:\ttotal: 22.4s\tremaining: 43.4s\n",
      "340:\ttotal: 22.5s\tremaining: 43.4s\n",
      "341:\ttotal: 22.6s\tremaining: 43.4s\n",
      "342:\ttotal: 22.6s\tremaining: 43.4s\n",
      "343:\ttotal: 22.7s\tremaining: 43.3s\n",
      "344:\ttotal: 22.8s\tremaining: 43.3s\n",
      "345:\ttotal: 22.9s\tremaining: 43.2s\n",
      "346:\ttotal: 22.9s\tremaining: 43.2s\n",
      "347:\ttotal: 23s\tremaining: 43.1s\n",
      "348:\ttotal: 23.1s\tremaining: 43s\n",
      "349:\ttotal: 23.1s\tremaining: 43s\n",
      "350:\ttotal: 23.2s\tremaining: 42.9s\n",
      "351:\ttotal: 23.3s\tremaining: 42.8s\n",
      "352:\ttotal: 23.3s\tremaining: 42.7s\n",
      "353:\ttotal: 23.4s\tremaining: 42.7s\n",
      "354:\ttotal: 23.5s\tremaining: 42.6s\n",
      "355:\ttotal: 23.5s\tremaining: 42.5s\n",
      "356:\ttotal: 23.6s\tremaining: 42.5s\n",
      "357:\ttotal: 23.6s\tremaining: 42.4s\n",
      "358:\ttotal: 23.7s\tremaining: 42.3s\n",
      "359:\ttotal: 23.8s\tremaining: 42.3s\n",
      "360:\ttotal: 23.9s\tremaining: 42.3s\n",
      "361:\ttotal: 24s\tremaining: 42.3s\n",
      "362:\ttotal: 24s\tremaining: 42.2s\n",
      "363:\ttotal: 24.1s\tremaining: 42.1s\n",
      "364:\ttotal: 24.2s\tremaining: 42.1s\n",
      "365:\ttotal: 24.3s\tremaining: 42.1s\n",
      "366:\ttotal: 24.3s\tremaining: 42s\n",
      "367:\ttotal: 24.4s\tremaining: 41.9s\n",
      "368:\ttotal: 24.5s\tremaining: 41.9s\n",
      "369:\ttotal: 24.6s\tremaining: 41.8s\n",
      "370:\ttotal: 24.6s\tremaining: 41.8s\n",
      "371:\ttotal: 24.7s\tremaining: 41.7s\n",
      "372:\ttotal: 24.8s\tremaining: 41.6s\n",
      "373:\ttotal: 24.9s\tremaining: 41.6s\n",
      "374:\ttotal: 24.9s\tremaining: 41.6s\n",
      "375:\ttotal: 25s\tremaining: 41.5s\n",
      "376:\ttotal: 25.1s\tremaining: 41.5s\n",
      "377:\ttotal: 25.2s\tremaining: 41.4s\n",
      "378:\ttotal: 25.3s\tremaining: 41.4s\n",
      "379:\ttotal: 25.3s\tremaining: 41.3s\n",
      "380:\ttotal: 25.4s\tremaining: 41.3s\n",
      "381:\ttotal: 25.5s\tremaining: 41.2s\n",
      "382:\ttotal: 25.6s\tremaining: 41.2s\n",
      "383:\ttotal: 25.6s\tremaining: 41.1s\n",
      "384:\ttotal: 25.7s\tremaining: 41.1s\n",
      "385:\ttotal: 25.8s\tremaining: 41s\n",
      "386:\ttotal: 25.9s\tremaining: 41s\n",
      "387:\ttotal: 25.9s\tremaining: 40.9s\n",
      "388:\ttotal: 26s\tremaining: 40.9s\n",
      "389:\ttotal: 26.1s\tremaining: 40.8s\n",
      "390:\ttotal: 26.2s\tremaining: 40.8s\n",
      "391:\ttotal: 26.3s\tremaining: 40.7s\n",
      "392:\ttotal: 26.3s\tremaining: 40.7s\n",
      "393:\ttotal: 26.4s\tremaining: 40.6s\n",
      "394:\ttotal: 26.5s\tremaining: 40.6s\n",
      "395:\ttotal: 26.6s\tremaining: 40.5s\n",
      "396:\ttotal: 26.6s\tremaining: 40.5s\n",
      "397:\ttotal: 26.7s\tremaining: 40.4s\n",
      "398:\ttotal: 26.8s\tremaining: 40.4s\n",
      "399:\ttotal: 26.8s\tremaining: 40.3s\n",
      "400:\ttotal: 26.9s\tremaining: 40.2s\n",
      "401:\ttotal: 27s\tremaining: 40.1s\n",
      "402:\ttotal: 27.1s\tremaining: 40.1s\n",
      "403:\ttotal: 27.1s\tremaining: 40s\n",
      "404:\ttotal: 27.2s\tremaining: 39.9s\n",
      "405:\ttotal: 27.3s\tremaining: 39.9s\n",
      "406:\ttotal: 27.3s\tremaining: 39.8s\n",
      "407:\ttotal: 27.4s\tremaining: 39.8s\n",
      "408:\ttotal: 27.5s\tremaining: 39.7s\n",
      "409:\ttotal: 27.6s\tremaining: 39.7s\n",
      "410:\ttotal: 27.6s\tremaining: 39.6s\n",
      "411:\ttotal: 27.7s\tremaining: 39.5s\n",
      "412:\ttotal: 27.8s\tremaining: 39.5s\n",
      "413:\ttotal: 27.8s\tremaining: 39.4s\n",
      "414:\ttotal: 27.9s\tremaining: 39.4s\n",
      "415:\ttotal: 28s\tremaining: 39.3s\n",
      "416:\ttotal: 28.1s\tremaining: 39.2s\n",
      "417:\ttotal: 28.1s\tremaining: 39.2s\n",
      "418:\ttotal: 28.2s\tremaining: 39.1s\n",
      "419:\ttotal: 28.3s\tremaining: 39.1s\n",
      "420:\ttotal: 28.4s\tremaining: 39s\n",
      "421:\ttotal: 28.5s\tremaining: 39s\n",
      "422:\ttotal: 28.5s\tremaining: 38.9s\n",
      "423:\ttotal: 28.6s\tremaining: 38.9s\n",
      "424:\ttotal: 28.7s\tremaining: 38.8s\n",
      "425:\ttotal: 28.8s\tremaining: 38.7s\n",
      "426:\ttotal: 28.8s\tremaining: 38.7s\n",
      "427:\ttotal: 28.9s\tremaining: 38.6s\n",
      "428:\ttotal: 29s\tremaining: 38.6s\n",
      "429:\ttotal: 29.1s\tremaining: 38.5s\n",
      "430:\ttotal: 29.2s\tremaining: 38.5s\n",
      "431:\ttotal: 29.2s\tremaining: 38.4s\n",
      "432:\ttotal: 29.3s\tremaining: 38.4s\n",
      "433:\ttotal: 29.4s\tremaining: 38.3s\n",
      "434:\ttotal: 29.4s\tremaining: 38.2s\n",
      "435:\ttotal: 29.5s\tremaining: 38.2s\n",
      "436:\ttotal: 29.6s\tremaining: 38.1s\n",
      "437:\ttotal: 29.7s\tremaining: 38.1s\n",
      "438:\ttotal: 29.7s\tremaining: 38s\n",
      "439:\ttotal: 29.8s\tremaining: 37.9s\n",
      "440:\ttotal: 29.9s\tremaining: 37.9s\n",
      "441:\ttotal: 29.9s\tremaining: 37.8s\n",
      "442:\ttotal: 30s\tremaining: 37.8s\n",
      "443:\ttotal: 30.1s\tremaining: 37.7s\n",
      "444:\ttotal: 30.2s\tremaining: 37.6s\n",
      "445:\ttotal: 30.2s\tremaining: 37.6s\n",
      "446:\ttotal: 30.3s\tremaining: 37.5s\n",
      "447:\ttotal: 30.4s\tremaining: 37.4s\n",
      "448:\ttotal: 30.4s\tremaining: 37.4s\n",
      "449:\ttotal: 30.5s\tremaining: 37.3s\n",
      "450:\ttotal: 30.6s\tremaining: 37.2s\n",
      "451:\ttotal: 30.7s\tremaining: 37.2s\n",
      "452:\ttotal: 30.7s\tremaining: 37.1s\n",
      "453:\ttotal: 30.8s\tremaining: 37s\n",
      "454:\ttotal: 30.9s\tremaining: 37s\n",
      "455:\ttotal: 31s\tremaining: 36.9s\n",
      "456:\ttotal: 31s\tremaining: 36.9s\n",
      "457:\ttotal: 31.1s\tremaining: 36.8s\n",
      "458:\ttotal: 31.2s\tremaining: 36.7s\n",
      "459:\ttotal: 31.2s\tremaining: 36.7s\n",
      "460:\ttotal: 31.3s\tremaining: 36.6s\n",
      "461:\ttotal: 31.3s\tremaining: 36.5s\n",
      "462:\ttotal: 31.4s\tremaining: 36.4s\n",
      "463:\ttotal: 31.5s\tremaining: 36.4s\n",
      "464:\ttotal: 31.6s\tremaining: 36.3s\n",
      "465:\ttotal: 31.6s\tremaining: 36.2s\n",
      "466:\ttotal: 31.7s\tremaining: 36.1s\n",
      "467:\ttotal: 31.7s\tremaining: 36.1s\n",
      "468:\ttotal: 31.8s\tremaining: 36s\n",
      "469:\ttotal: 31.9s\tremaining: 35.9s\n",
      "470:\ttotal: 31.9s\tremaining: 35.9s\n",
      "471:\ttotal: 32s\tremaining: 35.8s\n",
      "472:\ttotal: 32.1s\tremaining: 35.8s\n",
      "473:\ttotal: 32.2s\tremaining: 35.7s\n",
      "474:\ttotal: 32.2s\tremaining: 35.6s\n",
      "475:\ttotal: 32.3s\tremaining: 35.6s\n",
      "476:\ttotal: 32.4s\tremaining: 35.5s\n",
      "477:\ttotal: 32.5s\tremaining: 35.5s\n",
      "478:\ttotal: 32.5s\tremaining: 35.4s\n",
      "479:\ttotal: 32.6s\tremaining: 35.3s\n",
      "480:\ttotal: 32.7s\tremaining: 35.2s\n",
      "481:\ttotal: 32.7s\tremaining: 35.2s\n",
      "482:\ttotal: 32.8s\tremaining: 35.1s\n",
      "483:\ttotal: 32.9s\tremaining: 35s\n",
      "484:\ttotal: 33s\tremaining: 35s\n",
      "485:\ttotal: 33s\tremaining: 34.9s\n",
      "486:\ttotal: 33.1s\tremaining: 34.9s\n",
      "487:\ttotal: 33.2s\tremaining: 34.8s\n",
      "488:\ttotal: 33.3s\tremaining: 34.8s\n",
      "489:\ttotal: 33.3s\tremaining: 34.7s\n",
      "490:\ttotal: 33.4s\tremaining: 34.6s\n",
      "491:\ttotal: 33.5s\tremaining: 34.6s\n",
      "492:\ttotal: 33.6s\tremaining: 34.5s\n",
      "493:\ttotal: 33.6s\tremaining: 34.5s\n",
      "494:\ttotal: 33.7s\tremaining: 34.4s\n",
      "495:\ttotal: 33.8s\tremaining: 34.3s\n",
      "496:\ttotal: 33.8s\tremaining: 34.2s\n",
      "497:\ttotal: 33.9s\tremaining: 34.2s\n",
      "498:\ttotal: 33.9s\tremaining: 34.1s\n",
      "499:\ttotal: 34s\tremaining: 34s\n",
      "500:\ttotal: 34.1s\tremaining: 34s\n",
      "501:\ttotal: 34.2s\tremaining: 33.9s\n",
      "502:\ttotal: 34.3s\tremaining: 33.9s\n",
      "503:\ttotal: 34.3s\tremaining: 33.8s\n",
      "504:\ttotal: 34.4s\tremaining: 33.7s\n",
      "505:\ttotal: 34.5s\tremaining: 33.6s\n",
      "506:\ttotal: 34.5s\tremaining: 33.6s\n",
      "507:\ttotal: 34.6s\tremaining: 33.5s\n",
      "508:\ttotal: 34.7s\tremaining: 33.4s\n",
      "509:\ttotal: 34.7s\tremaining: 33.4s\n",
      "510:\ttotal: 34.8s\tremaining: 33.3s\n",
      "511:\ttotal: 34.9s\tremaining: 33.2s\n",
      "512:\ttotal: 35s\tremaining: 33.2s\n",
      "513:\ttotal: 35s\tremaining: 33.1s\n",
      "514:\ttotal: 35.1s\tremaining: 33.1s\n",
      "515:\ttotal: 35.2s\tremaining: 33s\n",
      "516:\ttotal: 35.3s\tremaining: 33s\n",
      "517:\ttotal: 35.4s\tremaining: 32.9s\n",
      "518:\ttotal: 35.5s\tremaining: 32.9s\n",
      "519:\ttotal: 35.5s\tremaining: 32.8s\n",
      "520:\ttotal: 35.6s\tremaining: 32.7s\n",
      "521:\ttotal: 35.7s\tremaining: 32.7s\n",
      "522:\ttotal: 35.8s\tremaining: 32.6s\n",
      "523:\ttotal: 35.8s\tremaining: 32.6s\n",
      "524:\ttotal: 35.9s\tremaining: 32.5s\n",
      "525:\ttotal: 36s\tremaining: 32.4s\n",
      "526:\ttotal: 36.1s\tremaining: 32.4s\n",
      "527:\ttotal: 36.1s\tremaining: 32.3s\n",
      "528:\ttotal: 36.2s\tremaining: 32.2s\n",
      "529:\ttotal: 36.3s\tremaining: 32.2s\n",
      "530:\ttotal: 36.4s\tremaining: 32.1s\n",
      "531:\ttotal: 36.4s\tremaining: 32.1s\n",
      "532:\ttotal: 36.5s\tremaining: 32s\n",
      "533:\ttotal: 36.6s\tremaining: 31.9s\n",
      "534:\ttotal: 36.6s\tremaining: 31.9s\n",
      "535:\ttotal: 36.7s\tremaining: 31.8s\n",
      "536:\ttotal: 36.8s\tremaining: 31.7s\n",
      "537:\ttotal: 36.9s\tremaining: 31.7s\n",
      "538:\ttotal: 37s\tremaining: 31.6s\n",
      "539:\ttotal: 37s\tremaining: 31.5s\n",
      "540:\ttotal: 37.1s\tremaining: 31.5s\n",
      "541:\ttotal: 37.2s\tremaining: 31.4s\n",
      "542:\ttotal: 37.2s\tremaining: 31.3s\n",
      "543:\ttotal: 37.3s\tremaining: 31.3s\n",
      "544:\ttotal: 37.4s\tremaining: 31.2s\n",
      "545:\ttotal: 37.5s\tremaining: 31.1s\n",
      "546:\ttotal: 37.5s\tremaining: 31.1s\n",
      "547:\ttotal: 37.6s\tremaining: 31s\n",
      "548:\ttotal: 37.7s\tremaining: 31s\n",
      "549:\ttotal: 37.8s\tremaining: 30.9s\n",
      "550:\ttotal: 37.8s\tremaining: 30.8s\n",
      "551:\ttotal: 37.9s\tremaining: 30.8s\n",
      "552:\ttotal: 38s\tremaining: 30.7s\n",
      "553:\ttotal: 38.1s\tremaining: 30.6s\n",
      "554:\ttotal: 38.1s\tremaining: 30.6s\n",
      "555:\ttotal: 38.2s\tremaining: 30.5s\n",
      "556:\ttotal: 38.3s\tremaining: 30.5s\n",
      "557:\ttotal: 38.4s\tremaining: 30.4s\n",
      "558:\ttotal: 38.4s\tremaining: 30.3s\n",
      "559:\ttotal: 38.5s\tremaining: 30.3s\n",
      "560:\ttotal: 38.6s\tremaining: 30.2s\n",
      "561:\ttotal: 38.7s\tremaining: 30.1s\n",
      "562:\ttotal: 38.7s\tremaining: 30.1s\n",
      "563:\ttotal: 38.8s\tremaining: 30s\n",
      "564:\ttotal: 38.9s\tremaining: 29.9s\n",
      "565:\ttotal: 38.9s\tremaining: 29.9s\n",
      "566:\ttotal: 39s\tremaining: 29.8s\n",
      "567:\ttotal: 39.1s\tremaining: 29.7s\n",
      "568:\ttotal: 39.2s\tremaining: 29.7s\n",
      "569:\ttotal: 39.2s\tremaining: 29.6s\n",
      "570:\ttotal: 39.3s\tremaining: 29.5s\n",
      "571:\ttotal: 39.4s\tremaining: 29.5s\n",
      "572:\ttotal: 39.5s\tremaining: 29.4s\n",
      "573:\ttotal: 39.5s\tremaining: 29.3s\n",
      "574:\ttotal: 39.6s\tremaining: 29.3s\n",
      "575:\ttotal: 39.7s\tremaining: 29.2s\n",
      "576:\ttotal: 39.7s\tremaining: 29.1s\n",
      "577:\ttotal: 39.8s\tremaining: 29.1s\n",
      "578:\ttotal: 39.9s\tremaining: 29s\n",
      "579:\ttotal: 40s\tremaining: 28.9s\n",
      "580:\ttotal: 40s\tremaining: 28.9s\n",
      "581:\ttotal: 40.1s\tremaining: 28.8s\n",
      "582:\ttotal: 40.2s\tremaining: 28.8s\n",
      "583:\ttotal: 40.3s\tremaining: 28.7s\n",
      "584:\ttotal: 40.3s\tremaining: 28.6s\n",
      "585:\ttotal: 40.4s\tremaining: 28.6s\n",
      "586:\ttotal: 40.5s\tremaining: 28.5s\n",
      "587:\ttotal: 40.6s\tremaining: 28.4s\n",
      "588:\ttotal: 40.7s\tremaining: 28.4s\n",
      "589:\ttotal: 40.7s\tremaining: 28.3s\n",
      "590:\ttotal: 40.8s\tremaining: 28.3s\n",
      "591:\ttotal: 40.9s\tremaining: 28.2s\n",
      "592:\ttotal: 41s\tremaining: 28.1s\n",
      "593:\ttotal: 41.1s\tremaining: 28.1s\n",
      "594:\ttotal: 41.2s\tremaining: 28s\n",
      "595:\ttotal: 41.2s\tremaining: 28s\n",
      "596:\ttotal: 41.3s\tremaining: 27.9s\n",
      "597:\ttotal: 41.4s\tremaining: 27.8s\n",
      "598:\ttotal: 41.5s\tremaining: 27.8s\n",
      "599:\ttotal: 41.6s\tremaining: 27.7s\n",
      "600:\ttotal: 41.6s\tremaining: 27.6s\n",
      "601:\ttotal: 41.7s\tremaining: 27.6s\n",
      "602:\ttotal: 41.8s\tremaining: 27.5s\n",
      "603:\ttotal: 41.8s\tremaining: 27.4s\n",
      "604:\ttotal: 41.9s\tremaining: 27.4s\n",
      "605:\ttotal: 42s\tremaining: 27.3s\n",
      "606:\ttotal: 42.1s\tremaining: 27.2s\n",
      "607:\ttotal: 42.1s\tremaining: 27.2s\n",
      "608:\ttotal: 42.2s\tremaining: 27.1s\n",
      "609:\ttotal: 42.3s\tremaining: 27s\n",
      "610:\ttotal: 42.4s\tremaining: 27s\n",
      "611:\ttotal: 42.4s\tremaining: 26.9s\n",
      "612:\ttotal: 42.5s\tremaining: 26.8s\n",
      "613:\ttotal: 42.5s\tremaining: 26.7s\n",
      "614:\ttotal: 42.6s\tremaining: 26.7s\n",
      "615:\ttotal: 42.7s\tremaining: 26.6s\n",
      "616:\ttotal: 42.8s\tremaining: 26.5s\n",
      "617:\ttotal: 42.8s\tremaining: 26.5s\n",
      "618:\ttotal: 42.9s\tremaining: 26.4s\n",
      "619:\ttotal: 43s\tremaining: 26.3s\n",
      "620:\ttotal: 43s\tremaining: 26.3s\n",
      "621:\ttotal: 43.1s\tremaining: 26.2s\n",
      "622:\ttotal: 43.2s\tremaining: 26.1s\n",
      "623:\ttotal: 43.3s\tremaining: 26.1s\n",
      "624:\ttotal: 43.3s\tremaining: 26s\n",
      "625:\ttotal: 43.4s\tremaining: 25.9s\n",
      "626:\ttotal: 43.5s\tremaining: 25.9s\n",
      "627:\ttotal: 43.6s\tremaining: 25.8s\n",
      "628:\ttotal: 43.6s\tremaining: 25.7s\n",
      "629:\ttotal: 43.7s\tremaining: 25.7s\n",
      "630:\ttotal: 43.8s\tremaining: 25.6s\n",
      "631:\ttotal: 43.9s\tremaining: 25.5s\n",
      "632:\ttotal: 43.9s\tremaining: 25.5s\n",
      "633:\ttotal: 44s\tremaining: 25.4s\n",
      "634:\ttotal: 44.1s\tremaining: 25.3s\n",
      "635:\ttotal: 44.2s\tremaining: 25.3s\n",
      "636:\ttotal: 44.2s\tremaining: 25.2s\n",
      "637:\ttotal: 44.3s\tremaining: 25.1s\n",
      "638:\ttotal: 44.4s\tremaining: 25.1s\n",
      "639:\ttotal: 44.5s\tremaining: 25s\n",
      "640:\ttotal: 44.5s\tremaining: 24.9s\n",
      "641:\ttotal: 44.6s\tremaining: 24.9s\n",
      "642:\ttotal: 44.6s\tremaining: 24.8s\n",
      "643:\ttotal: 44.7s\tremaining: 24.7s\n",
      "644:\ttotal: 44.8s\tremaining: 24.6s\n",
      "645:\ttotal: 44.9s\tremaining: 24.6s\n",
      "646:\ttotal: 44.9s\tremaining: 24.5s\n",
      "647:\ttotal: 45s\tremaining: 24.4s\n",
      "648:\ttotal: 45.1s\tremaining: 24.4s\n",
      "649:\ttotal: 45.2s\tremaining: 24.3s\n",
      "650:\ttotal: 45.3s\tremaining: 24.3s\n",
      "651:\ttotal: 45.3s\tremaining: 24.2s\n",
      "652:\ttotal: 45.4s\tremaining: 24.1s\n",
      "653:\ttotal: 45.5s\tremaining: 24s\n",
      "654:\ttotal: 45.5s\tremaining: 24s\n",
      "655:\ttotal: 45.6s\tremaining: 23.9s\n",
      "656:\ttotal: 45.7s\tremaining: 23.8s\n",
      "657:\ttotal: 45.7s\tremaining: 23.8s\n",
      "658:\ttotal: 45.8s\tremaining: 23.7s\n",
      "659:\ttotal: 45.9s\tremaining: 23.6s\n",
      "660:\ttotal: 46s\tremaining: 23.6s\n",
      "661:\ttotal: 46s\tremaining: 23.5s\n",
      "662:\ttotal: 46.1s\tremaining: 23.4s\n",
      "663:\ttotal: 46.2s\tremaining: 23.4s\n",
      "664:\ttotal: 46.3s\tremaining: 23.3s\n",
      "665:\ttotal: 46.3s\tremaining: 23.2s\n",
      "666:\ttotal: 46.4s\tremaining: 23.2s\n",
      "667:\ttotal: 46.5s\tremaining: 23.1s\n",
      "668:\ttotal: 46.5s\tremaining: 23s\n",
      "669:\ttotal: 46.6s\tremaining: 23s\n",
      "670:\ttotal: 46.7s\tremaining: 22.9s\n",
      "671:\ttotal: 46.8s\tremaining: 22.8s\n",
      "672:\ttotal: 46.8s\tremaining: 22.8s\n",
      "673:\ttotal: 46.9s\tremaining: 22.7s\n",
      "674:\ttotal: 47s\tremaining: 22.6s\n",
      "675:\ttotal: 47s\tremaining: 22.5s\n",
      "676:\ttotal: 47.1s\tremaining: 22.5s\n",
      "677:\ttotal: 47.2s\tremaining: 22.4s\n",
      "678:\ttotal: 47.2s\tremaining: 22.3s\n",
      "679:\ttotal: 47.3s\tremaining: 22.3s\n",
      "680:\ttotal: 47.4s\tremaining: 22.2s\n",
      "681:\ttotal: 47.5s\tremaining: 22.1s\n",
      "682:\ttotal: 47.5s\tremaining: 22.1s\n",
      "683:\ttotal: 47.6s\tremaining: 22s\n",
      "684:\ttotal: 47.7s\tremaining: 21.9s\n",
      "685:\ttotal: 47.8s\tremaining: 21.9s\n",
      "686:\ttotal: 47.8s\tremaining: 21.8s\n",
      "687:\ttotal: 47.9s\tremaining: 21.7s\n",
      "688:\ttotal: 48s\tremaining: 21.7s\n",
      "689:\ttotal: 48.1s\tremaining: 21.6s\n",
      "690:\ttotal: 48.2s\tremaining: 21.5s\n",
      "691:\ttotal: 48.3s\tremaining: 21.5s\n",
      "692:\ttotal: 48.3s\tremaining: 21.4s\n",
      "693:\ttotal: 48.4s\tremaining: 21.4s\n",
      "694:\ttotal: 48.5s\tremaining: 21.3s\n",
      "695:\ttotal: 48.6s\tremaining: 21.2s\n",
      "696:\ttotal: 48.6s\tremaining: 21.1s\n",
      "697:\ttotal: 48.7s\tremaining: 21.1s\n",
      "698:\ttotal: 48.8s\tremaining: 21s\n",
      "699:\ttotal: 48.8s\tremaining: 20.9s\n",
      "700:\ttotal: 48.9s\tremaining: 20.9s\n",
      "701:\ttotal: 49s\tremaining: 20.8s\n",
      "702:\ttotal: 49.1s\tremaining: 20.7s\n",
      "703:\ttotal: 49.1s\tremaining: 20.7s\n",
      "704:\ttotal: 49.2s\tremaining: 20.6s\n",
      "705:\ttotal: 49.3s\tremaining: 20.5s\n",
      "706:\ttotal: 49.4s\tremaining: 20.5s\n",
      "707:\ttotal: 49.5s\tremaining: 20.4s\n",
      "708:\ttotal: 49.5s\tremaining: 20.3s\n",
      "709:\ttotal: 49.6s\tremaining: 20.3s\n",
      "710:\ttotal: 49.7s\tremaining: 20.2s\n",
      "711:\ttotal: 49.8s\tremaining: 20.1s\n",
      "712:\ttotal: 49.9s\tremaining: 20.1s\n",
      "713:\ttotal: 49.9s\tremaining: 20s\n",
      "714:\ttotal: 50s\tremaining: 19.9s\n",
      "715:\ttotal: 50.1s\tremaining: 19.9s\n",
      "716:\ttotal: 50.2s\tremaining: 19.8s\n",
      "717:\ttotal: 50.2s\tremaining: 19.7s\n",
      "718:\ttotal: 50.3s\tremaining: 19.7s\n",
      "719:\ttotal: 50.4s\tremaining: 19.6s\n",
      "720:\ttotal: 50.5s\tremaining: 19.5s\n",
      "721:\ttotal: 50.6s\tremaining: 19.5s\n",
      "722:\ttotal: 50.6s\tremaining: 19.4s\n",
      "723:\ttotal: 50.7s\tremaining: 19.3s\n",
      "724:\ttotal: 50.8s\tremaining: 19.3s\n",
      "725:\ttotal: 50.9s\tremaining: 19.2s\n",
      "726:\ttotal: 50.9s\tremaining: 19.1s\n",
      "727:\ttotal: 51s\tremaining: 19s\n",
      "728:\ttotal: 51.1s\tremaining: 19s\n",
      "729:\ttotal: 51.1s\tremaining: 18.9s\n",
      "730:\ttotal: 51.2s\tremaining: 18.8s\n",
      "731:\ttotal: 51.3s\tremaining: 18.8s\n",
      "732:\ttotal: 51.4s\tremaining: 18.7s\n",
      "733:\ttotal: 51.4s\tremaining: 18.6s\n",
      "734:\ttotal: 51.5s\tremaining: 18.6s\n",
      "735:\ttotal: 51.6s\tremaining: 18.5s\n",
      "736:\ttotal: 51.6s\tremaining: 18.4s\n",
      "737:\ttotal: 51.7s\tremaining: 18.4s\n",
      "738:\ttotal: 51.8s\tremaining: 18.3s\n",
      "739:\ttotal: 51.9s\tremaining: 18.2s\n",
      "740:\ttotal: 52s\tremaining: 18.2s\n",
      "741:\ttotal: 52s\tremaining: 18.1s\n",
      "742:\ttotal: 52.1s\tremaining: 18s\n",
      "743:\ttotal: 52.2s\tremaining: 17.9s\n",
      "744:\ttotal: 52.2s\tremaining: 17.9s\n",
      "745:\ttotal: 52.3s\tremaining: 17.8s\n",
      "746:\ttotal: 52.4s\tremaining: 17.7s\n",
      "747:\ttotal: 52.4s\tremaining: 17.7s\n",
      "748:\ttotal: 52.5s\tremaining: 17.6s\n",
      "749:\ttotal: 52.6s\tremaining: 17.5s\n",
      "750:\ttotal: 52.7s\tremaining: 17.5s\n",
      "751:\ttotal: 52.7s\tremaining: 17.4s\n",
      "752:\ttotal: 52.8s\tremaining: 17.3s\n",
      "753:\ttotal: 52.9s\tremaining: 17.3s\n",
      "754:\ttotal: 53s\tremaining: 17.2s\n",
      "755:\ttotal: 53s\tremaining: 17.1s\n",
      "756:\ttotal: 53.1s\tremaining: 17.1s\n",
      "757:\ttotal: 53.2s\tremaining: 17s\n",
      "758:\ttotal: 53.3s\tremaining: 16.9s\n",
      "759:\ttotal: 53.4s\tremaining: 16.9s\n",
      "760:\ttotal: 53.5s\tremaining: 16.8s\n",
      "761:\ttotal: 53.6s\tremaining: 16.7s\n",
      "762:\ttotal: 53.6s\tremaining: 16.7s\n",
      "763:\ttotal: 53.7s\tremaining: 16.6s\n",
      "764:\ttotal: 53.8s\tremaining: 16.5s\n",
      "765:\ttotal: 53.9s\tremaining: 16.5s\n",
      "766:\ttotal: 53.9s\tremaining: 16.4s\n",
      "767:\ttotal: 54s\tremaining: 16.3s\n",
      "768:\ttotal: 54.1s\tremaining: 16.2s\n",
      "769:\ttotal: 54.2s\tremaining: 16.2s\n",
      "770:\ttotal: 54.3s\tremaining: 16.1s\n",
      "771:\ttotal: 54.3s\tremaining: 16s\n",
      "772:\ttotal: 54.4s\tremaining: 16s\n",
      "773:\ttotal: 54.5s\tremaining: 15.9s\n",
      "774:\ttotal: 54.6s\tremaining: 15.8s\n",
      "775:\ttotal: 54.6s\tremaining: 15.8s\n",
      "776:\ttotal: 54.7s\tremaining: 15.7s\n",
      "777:\ttotal: 54.8s\tremaining: 15.6s\n",
      "778:\ttotal: 54.8s\tremaining: 15.6s\n",
      "779:\ttotal: 54.9s\tremaining: 15.5s\n",
      "780:\ttotal: 55s\tremaining: 15.4s\n",
      "781:\ttotal: 55.1s\tremaining: 15.4s\n",
      "782:\ttotal: 55.2s\tremaining: 15.3s\n",
      "783:\ttotal: 55.2s\tremaining: 15.2s\n",
      "784:\ttotal: 55.3s\tremaining: 15.1s\n",
      "785:\ttotal: 55.4s\tremaining: 15.1s\n",
      "786:\ttotal: 55.5s\tremaining: 15s\n",
      "787:\ttotal: 55.5s\tremaining: 14.9s\n",
      "788:\ttotal: 55.6s\tremaining: 14.9s\n",
      "789:\ttotal: 55.7s\tremaining: 14.8s\n",
      "790:\ttotal: 55.8s\tremaining: 14.7s\n",
      "791:\ttotal: 55.9s\tremaining: 14.7s\n",
      "792:\ttotal: 56s\tremaining: 14.6s\n",
      "793:\ttotal: 56s\tremaining: 14.5s\n",
      "794:\ttotal: 56.1s\tremaining: 14.5s\n",
      "795:\ttotal: 56.2s\tremaining: 14.4s\n",
      "796:\ttotal: 56.3s\tremaining: 14.3s\n",
      "797:\ttotal: 56.3s\tremaining: 14.3s\n",
      "798:\ttotal: 56.4s\tremaining: 14.2s\n",
      "799:\ttotal: 56.5s\tremaining: 14.1s\n",
      "800:\ttotal: 56.6s\tremaining: 14.1s\n",
      "801:\ttotal: 56.7s\tremaining: 14s\n",
      "802:\ttotal: 56.7s\tremaining: 13.9s\n",
      "803:\ttotal: 56.8s\tremaining: 13.8s\n",
      "804:\ttotal: 56.9s\tremaining: 13.8s\n",
      "805:\ttotal: 57s\tremaining: 13.7s\n",
      "806:\ttotal: 57s\tremaining: 13.6s\n",
      "807:\ttotal: 57.1s\tremaining: 13.6s\n",
      "808:\ttotal: 57.2s\tremaining: 13.5s\n",
      "809:\ttotal: 57.3s\tremaining: 13.4s\n",
      "810:\ttotal: 57.3s\tremaining: 13.4s\n",
      "811:\ttotal: 57.4s\tremaining: 13.3s\n",
      "812:\ttotal: 57.5s\tremaining: 13.2s\n",
      "813:\ttotal: 57.6s\tremaining: 13.2s\n",
      "814:\ttotal: 57.7s\tremaining: 13.1s\n",
      "815:\ttotal: 57.7s\tremaining: 13s\n",
      "816:\ttotal: 57.8s\tremaining: 12.9s\n",
      "817:\ttotal: 57.9s\tremaining: 12.9s\n",
      "818:\ttotal: 57.9s\tremaining: 12.8s\n",
      "819:\ttotal: 58s\tremaining: 12.7s\n",
      "820:\ttotal: 58.1s\tremaining: 12.7s\n",
      "821:\ttotal: 58.2s\tremaining: 12.6s\n",
      "822:\ttotal: 58.3s\tremaining: 12.5s\n",
      "823:\ttotal: 58.3s\tremaining: 12.5s\n",
      "824:\ttotal: 58.4s\tremaining: 12.4s\n",
      "825:\ttotal: 58.5s\tremaining: 12.3s\n",
      "826:\ttotal: 58.6s\tremaining: 12.3s\n",
      "827:\ttotal: 58.6s\tremaining: 12.2s\n",
      "828:\ttotal: 58.7s\tremaining: 12.1s\n",
      "829:\ttotal: 58.8s\tremaining: 12s\n",
      "830:\ttotal: 58.8s\tremaining: 12s\n",
      "831:\ttotal: 58.9s\tremaining: 11.9s\n",
      "832:\ttotal: 59s\tremaining: 11.8s\n",
      "833:\ttotal: 59.1s\tremaining: 11.8s\n",
      "834:\ttotal: 59.1s\tremaining: 11.7s\n",
      "835:\ttotal: 59.2s\tremaining: 11.6s\n",
      "836:\ttotal: 59.3s\tremaining: 11.5s\n",
      "837:\ttotal: 59.4s\tremaining: 11.5s\n",
      "838:\ttotal: 59.5s\tremaining: 11.4s\n",
      "839:\ttotal: 59.5s\tremaining: 11.3s\n",
      "840:\ttotal: 59.6s\tremaining: 11.3s\n",
      "841:\ttotal: 59.7s\tremaining: 11.2s\n",
      "842:\ttotal: 59.7s\tremaining: 11.1s\n",
      "843:\ttotal: 59.8s\tremaining: 11s\n",
      "844:\ttotal: 59.9s\tremaining: 11s\n",
      "845:\ttotal: 59.9s\tremaining: 10.9s\n",
      "846:\ttotal: 1m\tremaining: 10.8s\n",
      "847:\ttotal: 1m\tremaining: 10.8s\n",
      "848:\ttotal: 1m\tremaining: 10.7s\n",
      "849:\ttotal: 1m\tremaining: 10.6s\n",
      "850:\ttotal: 1m\tremaining: 10.6s\n",
      "851:\ttotal: 1m\tremaining: 10.5s\n",
      "852:\ttotal: 1m\tremaining: 10.4s\n",
      "853:\ttotal: 1m\tremaining: 10.3s\n",
      "854:\ttotal: 1m\tremaining: 10.3s\n",
      "855:\ttotal: 1m\tremaining: 10.2s\n",
      "856:\ttotal: 1m\tremaining: 10.1s\n",
      "857:\ttotal: 1m\tremaining: 10.1s\n",
      "858:\ttotal: 1m\tremaining: 10s\n",
      "859:\ttotal: 1m\tremaining: 9.93s\n",
      "860:\ttotal: 1m 1s\tremaining: 9.86s\n",
      "861:\ttotal: 1m 1s\tremaining: 9.79s\n",
      "862:\ttotal: 1m 1s\tremaining: 9.72s\n",
      "863:\ttotal: 1m 1s\tremaining: 9.64s\n",
      "864:\ttotal: 1m 1s\tremaining: 9.57s\n",
      "865:\ttotal: 1m 1s\tremaining: 9.51s\n",
      "866:\ttotal: 1m 1s\tremaining: 9.44s\n",
      "867:\ttotal: 1m 1s\tremaining: 9.37s\n",
      "868:\ttotal: 1m 1s\tremaining: 9.29s\n",
      "869:\ttotal: 1m 1s\tremaining: 9.22s\n",
      "870:\ttotal: 1m 1s\tremaining: 9.15s\n",
      "871:\ttotal: 1m 1s\tremaining: 9.09s\n",
      "872:\ttotal: 1m 1s\tremaining: 9.02s\n",
      "873:\ttotal: 1m 2s\tremaining: 8.95s\n",
      "874:\ttotal: 1m 2s\tremaining: 8.88s\n",
      "875:\ttotal: 1m 2s\tremaining: 8.8s\n",
      "876:\ttotal: 1m 2s\tremaining: 8.73s\n",
      "877:\ttotal: 1m 2s\tremaining: 8.66s\n",
      "878:\ttotal: 1m 2s\tremaining: 8.59s\n",
      "879:\ttotal: 1m 2s\tremaining: 8.52s\n",
      "880:\ttotal: 1m 2s\tremaining: 8.45s\n",
      "881:\ttotal: 1m 2s\tremaining: 8.38s\n",
      "882:\ttotal: 1m 2s\tremaining: 8.31s\n",
      "883:\ttotal: 1m 2s\tremaining: 8.24s\n",
      "884:\ttotal: 1m 2s\tremaining: 8.17s\n",
      "885:\ttotal: 1m 2s\tremaining: 8.1s\n",
      "886:\ttotal: 1m 3s\tremaining: 8.03s\n",
      "887:\ttotal: 1m 3s\tremaining: 7.96s\n",
      "888:\ttotal: 1m 3s\tremaining: 7.89s\n",
      "889:\ttotal: 1m 3s\tremaining: 7.82s\n",
      "890:\ttotal: 1m 3s\tremaining: 7.75s\n",
      "891:\ttotal: 1m 3s\tremaining: 7.68s\n",
      "892:\ttotal: 1m 3s\tremaining: 7.61s\n",
      "893:\ttotal: 1m 3s\tremaining: 7.54s\n",
      "894:\ttotal: 1m 3s\tremaining: 7.47s\n",
      "895:\ttotal: 1m 3s\tremaining: 7.4s\n",
      "896:\ttotal: 1m 3s\tremaining: 7.33s\n",
      "897:\ttotal: 1m 3s\tremaining: 7.26s\n",
      "898:\ttotal: 1m 3s\tremaining: 7.18s\n",
      "899:\ttotal: 1m 4s\tremaining: 7.11s\n",
      "900:\ttotal: 1m 4s\tremaining: 7.04s\n",
      "901:\ttotal: 1m 4s\tremaining: 6.97s\n",
      "902:\ttotal: 1m 4s\tremaining: 6.9s\n",
      "903:\ttotal: 1m 4s\tremaining: 6.83s\n",
      "904:\ttotal: 1m 4s\tremaining: 6.76s\n",
      "905:\ttotal: 1m 4s\tremaining: 6.69s\n",
      "906:\ttotal: 1m 4s\tremaining: 6.62s\n",
      "907:\ttotal: 1m 4s\tremaining: 6.55s\n",
      "908:\ttotal: 1m 4s\tremaining: 6.48s\n",
      "909:\ttotal: 1m 4s\tremaining: 6.41s\n",
      "910:\ttotal: 1m 4s\tremaining: 6.33s\n",
      "911:\ttotal: 1m 4s\tremaining: 6.26s\n",
      "912:\ttotal: 1m 5s\tremaining: 6.19s\n",
      "913:\ttotal: 1m 5s\tremaining: 6.12s\n",
      "914:\ttotal: 1m 5s\tremaining: 6.05s\n",
      "915:\ttotal: 1m 5s\tremaining: 5.98s\n",
      "916:\ttotal: 1m 5s\tremaining: 5.91s\n",
      "917:\ttotal: 1m 5s\tremaining: 5.84s\n",
      "918:\ttotal: 1m 5s\tremaining: 5.77s\n",
      "919:\ttotal: 1m 5s\tremaining: 5.7s\n",
      "920:\ttotal: 1m 5s\tremaining: 5.63s\n",
      "921:\ttotal: 1m 5s\tremaining: 5.56s\n",
      "922:\ttotal: 1m 5s\tremaining: 5.49s\n",
      "923:\ttotal: 1m 5s\tremaining: 5.42s\n",
      "924:\ttotal: 1m 5s\tremaining: 5.34s\n",
      "925:\ttotal: 1m 5s\tremaining: 5.27s\n",
      "926:\ttotal: 1m 6s\tremaining: 5.2s\n",
      "927:\ttotal: 1m 6s\tremaining: 5.13s\n",
      "928:\ttotal: 1m 6s\tremaining: 5.06s\n",
      "929:\ttotal: 1m 6s\tremaining: 4.99s\n",
      "930:\ttotal: 1m 6s\tremaining: 4.92s\n",
      "931:\ttotal: 1m 6s\tremaining: 4.85s\n",
      "932:\ttotal: 1m 6s\tremaining: 4.78s\n",
      "933:\ttotal: 1m 6s\tremaining: 4.71s\n",
      "934:\ttotal: 1m 6s\tremaining: 4.64s\n",
      "935:\ttotal: 1m 6s\tremaining: 4.57s\n",
      "936:\ttotal: 1m 6s\tremaining: 4.5s\n",
      "937:\ttotal: 1m 6s\tremaining: 4.43s\n",
      "938:\ttotal: 1m 7s\tremaining: 4.36s\n",
      "939:\ttotal: 1m 7s\tremaining: 4.28s\n",
      "940:\ttotal: 1m 7s\tremaining: 4.21s\n",
      "941:\ttotal: 1m 7s\tremaining: 4.14s\n",
      "942:\ttotal: 1m 7s\tremaining: 4.07s\n",
      "943:\ttotal: 1m 7s\tremaining: 4s\n",
      "944:\ttotal: 1m 7s\tremaining: 3.93s\n",
      "945:\ttotal: 1m 7s\tremaining: 3.86s\n",
      "946:\ttotal: 1m 7s\tremaining: 3.79s\n",
      "947:\ttotal: 1m 7s\tremaining: 3.71s\n",
      "948:\ttotal: 1m 7s\tremaining: 3.64s\n",
      "949:\ttotal: 1m 7s\tremaining: 3.57s\n",
      "950:\ttotal: 1m 7s\tremaining: 3.5s\n",
      "951:\ttotal: 1m 8s\tremaining: 3.43s\n",
      "952:\ttotal: 1m 8s\tremaining: 3.36s\n",
      "953:\ttotal: 1m 8s\tremaining: 3.29s\n",
      "954:\ttotal: 1m 8s\tremaining: 3.21s\n",
      "955:\ttotal: 1m 8s\tremaining: 3.15s\n",
      "956:\ttotal: 1m 8s\tremaining: 3.07s\n",
      "957:\ttotal: 1m 8s\tremaining: 3s\n",
      "958:\ttotal: 1m 8s\tremaining: 2.93s\n",
      "959:\ttotal: 1m 8s\tremaining: 2.86s\n",
      "960:\ttotal: 1m 8s\tremaining: 2.79s\n",
      "961:\ttotal: 1m 8s\tremaining: 2.72s\n",
      "962:\ttotal: 1m 8s\tremaining: 2.65s\n",
      "963:\ttotal: 1m 8s\tremaining: 2.57s\n",
      "964:\ttotal: 1m 9s\tremaining: 2.5s\n",
      "965:\ttotal: 1m 9s\tremaining: 2.43s\n",
      "966:\ttotal: 1m 9s\tremaining: 2.36s\n",
      "967:\ttotal: 1m 9s\tremaining: 2.29s\n",
      "968:\ttotal: 1m 9s\tremaining: 2.22s\n",
      "969:\ttotal: 1m 9s\tremaining: 2.15s\n",
      "970:\ttotal: 1m 9s\tremaining: 2.07s\n",
      "971:\ttotal: 1m 9s\tremaining: 2s\n",
      "972:\ttotal: 1m 9s\tremaining: 1.93s\n",
      "973:\ttotal: 1m 9s\tremaining: 1.86s\n",
      "974:\ttotal: 1m 9s\tremaining: 1.79s\n",
      "975:\ttotal: 1m 9s\tremaining: 1.72s\n",
      "976:\ttotal: 1m 9s\tremaining: 1.65s\n",
      "977:\ttotal: 1m 9s\tremaining: 1.57s\n",
      "978:\ttotal: 1m 10s\tremaining: 1.5s\n",
      "979:\ttotal: 1m 10s\tremaining: 1.43s\n",
      "980:\ttotal: 1m 10s\tremaining: 1.36s\n",
      "981:\ttotal: 1m 10s\tremaining: 1.29s\n",
      "982:\ttotal: 1m 10s\tremaining: 1.22s\n",
      "983:\ttotal: 1m 10s\tremaining: 1.15s\n",
      "984:\ttotal: 1m 10s\tremaining: 1.07s\n",
      "985:\ttotal: 1m 10s\tremaining: 1s\n",
      "986:\ttotal: 1m 10s\tremaining: 932ms\n",
      "987:\ttotal: 1m 10s\tremaining: 860ms\n",
      "988:\ttotal: 1m 10s\tremaining: 789ms\n",
      "989:\ttotal: 1m 10s\tremaining: 717ms\n",
      "990:\ttotal: 1m 11s\tremaining: 645ms\n",
      "991:\ttotal: 1m 11s\tremaining: 574ms\n",
      "992:\ttotal: 1m 11s\tremaining: 502ms\n",
      "993:\ttotal: 1m 11s\tremaining: 430ms\n",
      "994:\ttotal: 1m 11s\tremaining: 359ms\n",
      "995:\ttotal: 1m 11s\tremaining: 287ms\n",
      "996:\ttotal: 1m 11s\tremaining: 215ms\n",
      "997:\ttotal: 1m 11s\tremaining: 144ms\n",
      "998:\ttotal: 1m 11s\tremaining: 71.8ms\n",
      "999:\ttotal: 1m 11s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x7f1f03567290>"
      ]
     },
     "execution_count": 42,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train,y_train)\n",
    "# model.fit(X_train,y_train['Response'],cat_features = categorical_var)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IUXlzg869Tv8",
    "outputId": "4752fea5-9ed0-4350-9b9e-5ac49afc5ebb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8034683512763128\n"
     ]
    }
   ],
   "source": [
    "\n",
    "ypred1 = model.predict_proba(X_test)\n",
    "print(roc_auc_score(y_test,ypred1[:,1] ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 452
    },
    "id": "Qn-FfHDB94Tf",
    "outputId": "15669737-e81a-4567-e52c-72b73e300f0c"
   },
   "outputs": [
    {
     "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>names</th>\n",
       "      <th>feat_imp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>City_Code</td>\n",
       "      <td>11.575661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Region_Code</td>\n",
       "      <td>23.339155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Accomodation_Type</td>\n",
       "      <td>0.586341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Reco_Insurance_Type</td>\n",
       "      <td>0.201680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Upper_Age</td>\n",
       "      <td>2.587616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Lower_Age</td>\n",
       "      <td>3.086536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Is_Spouse</td>\n",
       "      <td>0.320086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Health Indicator</td>\n",
       "      <td>5.029470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Holding_Policy_Duration</td>\n",
       "      <td>4.265320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Holding_Policy_Type</td>\n",
       "      <td>6.006698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Reco_Policy_Cat</td>\n",
       "      <td>35.389919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Reco_Policy_Premium</td>\n",
       "      <td>3.414772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Reco_Policy_Premium_original</td>\n",
       "      <td>4.196746</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           names   feat_imp\n",
       "0                      City_Code  11.575661\n",
       "1                    Region_Code  23.339155\n",
       "2              Accomodation_Type   0.586341\n",
       "3            Reco_Insurance_Type   0.201680\n",
       "4                      Upper_Age   2.587616\n",
       "5                      Lower_Age   3.086536\n",
       "6                      Is_Spouse   0.320086\n",
       "7               Health Indicator   5.029470\n",
       "8        Holding_Policy_Duration   4.265320\n",
       "9            Holding_Policy_Type   6.006698\n",
       "10               Reco_Policy_Cat  35.389919\n",
       "11           Reco_Policy_Premium   3.414772\n",
       "12  Reco_Policy_Premium_original   4.196746"
      ]
     },
     "execution_count": 52,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = all_data.columns\n",
    "# feat_final = sorted(zip(map(lambda x: round(x, 4), model.feature_importances_),names ), reverse=False)\n",
    "# feat_final\n",
    "df = pd.DataFrame()\n",
    "df['names'] = names\n",
    "df['feat_imp'] = model.feature_importances_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 441
    },
    "id": "UsUz2-XjGUfm",
    "outputId": "64d05b2d-d9b0-484a-86db-70e06d3bfdca"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6)) \n",
    "sns.barplot(x=\"names\", y=\"feat_imp\", data=df.sort_values(by=\"feat_imp\", ascending=False))\n",
    "plt.title(\"CatBoostClassifier's Feature Importance\")\n",
    "plt.xticks(rotation='vertical')\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "D6FtTLxSIS7a"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "catboost.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}