Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ad260319",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from statsmodels.stats.outliers_influence import variance_inflation_factor\n",
    "from mlxtend.feature_selection import SequentialFeatureSelector as sfs\n",
    "from sklearn.model_selection import train_test_split\n",
    "import statsmodels.api as sm\n",
    "from sklearn.metrics import accuracy_score,cohen_kappa_score,classification_report,f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5fa7616b",
   "metadata": {},
   "outputs": [],
   "source": [
    "base = pd.read_csv('entire_dataset.csv')\n",
    "df = base.copy(deep = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e4ecd3ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1309, 9)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5c0fc452",
   "metadata": {},
   "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>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Survived  Pclass     Sex   Age  SibSp  Parch     Fare Cabin Embarked\n",
       "0         0       3    male  22.0      1      0   7.2500   NaN        S\n",
       "1         1       1  female  38.0      1      0  71.2833   C85        C\n",
       "2         1       3  female  26.0      0      0   7.9250   NaN        S\n",
       "3         1       1  female  35.0      1      0  53.1000  C123        S\n",
       "4         0       3    male  35.0      0      0   8.0500   NaN        S"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "32330d36",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Null value treatment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bfdcc4ab",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Survived     0.000000\n",
       "Pclass       0.000000\n",
       "Sex          0.000000\n",
       "Age         20.091673\n",
       "SibSp        0.000000\n",
       "Parch        0.000000\n",
       "Fare         0.076394\n",
       "Cabin       77.463713\n",
       "Embarked     0.152788\n",
       "dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(df.isnull().sum() / len(df))*100"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12786dea",
   "metadata": {},
   "source": [
    "Inference : The column 'age' and 'cabin' are missing from the dataset. As the dataset, Cabin has more than 77% of null values. We can go ahead and drop the column. whereas the column 'Age' can be imputed based on the median proportions based on gender sub class - Male and Female because of 20% data is null. Dropping off the rows for the columns 'Fare','Embarekd' have null values are less than 1%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e8790d78",
   "metadata": {},
   "outputs": [],
   "source": [
    "# conditional Filling of the column \"age\" based on the gender column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ab55ed11",
   "metadata": {},
   "outputs": [],
   "source": [
    "median = df.groupby(['Sex'])['Age'].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5fad876d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df.set_index('Sex', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cc89f738",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Age'].fillna(median, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8b3535ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "96c52a8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop off a column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ccac5503",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(columns=['Cabin'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b3196cf7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop off a row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7f65d86d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.dropna(subset=['Fare','Embarked'], axis = 0, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "aadfbd84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sex         0\n",
       "Survived    0\n",
       "Pclass      0\n",
       "Age         0\n",
       "SibSp       0\n",
       "Parch       0\n",
       "Fare        0\n",
       "Embarked    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ed8edbaa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Categorical Imputation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1f4e3fab",
   "metadata": {},
   "outputs": [],
   "source": [
    "obj = df.select_dtypes(object)\n",
    "num = df.select_dtypes(np.number)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "20a3c574",
   "metadata": {},
   "outputs": [],
   "source": [
    "obj_encoded = pd.get_dummies(data = obj, drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "54bb2d12",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_enc = pd.concat([obj_encoded,num], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "19963f12",
   "metadata": {
    "scrolled": true
   },
   "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>Sex_male</th>\n",
       "      <th>Embarked_Q</th>\n",
       "      <th>Embarked_S</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1304</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1305</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>39.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>108.9000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1306</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>38.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1307</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1308</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>28.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22.3583</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1306 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Sex_male  Embarked_Q  Embarked_S  Survived  Pclass   Age  SibSp  Parch  \\\n",
       "0            1           0           1         0       3  22.0      1      0   \n",
       "1            0           0           0         1       1  38.0      1      0   \n",
       "2            0           0           1         1       3  26.0      0      0   \n",
       "3            0           0           1         1       1  35.0      1      0   \n",
       "4            1           0           1         0       3  35.0      0      0   \n",
       "...        ...         ...         ...       ...     ...   ...    ...    ...   \n",
       "1304         1           0           1         0       3  28.0      0      0   \n",
       "1305         0           0           0         1       1  39.0      0      0   \n",
       "1306         1           0           1         0       3  38.5      0      0   \n",
       "1307         1           0           1         0       3  28.0      0      0   \n",
       "1308         1           0           0         0       3  28.0      1      1   \n",
       "\n",
       "          Fare  \n",
       "0       7.2500  \n",
       "1      71.2833  \n",
       "2       7.9250  \n",
       "3      53.1000  \n",
       "4       8.0500  \n",
       "...        ...  \n",
       "1304    8.0500  \n",
       "1305  108.9000  \n",
       "1306    7.2500  \n",
       "1307    8.0500  \n",
       "1308   22.3583  \n",
       "\n",
       "[1306 rows x 9 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_enc"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bae3020d",
   "metadata": {},
   "source": [
    "Inference : From the dataset, as we have only 2 categorical variables – ‘Embarked’, ‘Gender’ and both are nominal dtype. Therefore, we can go ahead and one hot encoding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "1df6ac67",
   "metadata": {},
   "outputs": [],
   "source": [
    " # Target variable understanding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d37f12f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    62.33\n",
       "1    37.67\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df['Survived'].value_counts(normalize=True) * 100 ,2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ea8b4cc",
   "metadata": {},
   "source": [
    "Inference : From the dataset, we have 0 of 62.33% and 1 of 37.67% and proportions of two subclass are still good for the prediction and there no is no of using the smote technique."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "5549cb1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9a0b11cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Correlation understanding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "2c28675c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10,5))\n",
    "sns.heatmap(df_enc.corr(), annot = True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75ad787c",
   "metadata": {},
   "source": [
    "Inference : \n",
    "- We can see that variables like Embarked Q, EMbarked_S are moderate co-relate each other.\n",
    "- The variables like Pclass and Fare are moderate co-relate each other.\n",
    "- The variable - 'Sex_male' is highly co-relate with the target variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "93d3d705",
   "metadata": {},
   "outputs": [],
   "source": [
    "# multicolinearity checkup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "29b3191b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age           4.303873\n",
       "Embarked_S    3.334285\n",
       "Sex_male      2.663264\n",
       "Fare          1.688328\n",
       "Parch         1.472560\n",
       "SibSp         1.463704\n",
       "Embarked_Q    1.258468\n",
       "dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([variance_inflation_factor(df_enc.drop(columns = ['Survived','Pclass'],axis =1).values, i) for i in range(df_enc.drop(columns = ['Survived','Pclass'], axis =1).shape[1])], index=df_enc.drop(columns = ['Survived','Pclass'],axis =1).columns).sort_values(ascending= False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "bd89dc3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pclass        6.522742\n",
       "Age           4.719108\n",
       "Embarked_S    4.449452\n",
       "Sex_male      3.051311\n",
       "Fare          1.768283\n",
       "Embarked_Q    1.648108\n",
       "Parch         1.530793\n",
       "SibSp         1.504324\n",
       "dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series([variance_inflation_factor(df_enc.drop(columns = ['Survived'],axis =1).values, i) for i in range(df_enc.drop(columns = ['Survived'], axis =1).shape[1])], index=df_enc.drop(columns = ['Survived'],axis =1).columns).sort_values(ascending= False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2ac1eda",
   "metadata": {},
   "source": [
    "Inference : From the above analysis, we could observe that the variable 'Pclass' had the VIF value of 6.5 and after dropping it where all remaining variables have VIF thershold lesser than 5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "99e1bbcd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Forward feature engineering:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "61eac741",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "LR = LogisticRegression()\n",
    "\n",
    "sfs1 = sfs(LR, k_features='best', forward=True, verbose=2, scoring='f1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "03fcca96",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:43] Features: 1/8 -- score: 0.7984769767754061[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:    0.2s finished\n",
      "\n",
      "[2022-05-06 21:16:43] Features: 2/8 -- score: 0.798847996708831[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:43] Features: 3/8 -- score: 0.7998735846089656[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.3s finished\n",
      "\n",
      "[2022-05-06 21:16:44] Features: 4/8 -- score: 0.7998735846089656[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    0.3s finished\n",
      "\n",
      "[2022-05-06 21:16:44] Features: 5/8 -- score: 0.7998735846089656[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.2s finished\n",
      "\n",
      "[2022-05-06 21:16:44] Features: 6/8 -- score: 0.7978118899584385[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.2s finished\n",
      "\n",
      "[2022-05-06 21:16:44] Features: 7/8 -- score: 0.7934978440175391[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:45] Features: 8/8 -- score: 0.7940591566254838"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SequentialFeatureSelector(estimator=LogisticRegression(), k_features='best',\n",
       "                          scoring='f1', verbose=2)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs1.fit(df_enc.drop(columns = ['Survived'],axis =1), df_enc['Survived'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "69b89a66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7998735846089656"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs1.k_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "89ce0854",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('Sex_male', 'Age', 'SibSp')"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs1.k_feature_names_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "abb3fbf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Backward feature engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "351b74c0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "LR = LogisticRegression()\n",
    "\n",
    "sfs2 = sfs(LR, k_features='best', forward=False, verbose=2, scoring='f1')  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "bc9cf09f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.8s finished\n",
      "\n",
      "[2022-05-06 21:16:46] Features: 7/1 -- score: 0.7978118899584385[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n",
      "[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:    0.7s finished\n",
      "\n",
      "[2022-05-06 21:16:46] Features: 6/1 -- score: 0.7978118899584385[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:46] Features: 5/1 -- score: 0.798847996708831[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:47] Features: 4/1 -- score: 0.798847996708831[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:47] Features: 3/1 -- score: 0.798847996708831[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.1s finished\n",
      "\n",
      "[2022-05-06 21:16:47] Features: 2/1 -- score: 0.798847996708831[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.0s finished\n",
      "\n",
      "[2022-05-06 21:16:47] Features: 1/1 -- score: 0.7984769767754061"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SequentialFeatureSelector(estimator=LogisticRegression(), forward=False,\n",
       "                          k_features='best', scoring='f1', verbose=2)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs2.fit(df_enc.drop(columns = ['Survived'],axis =1), df_enc['Survived'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1f6d63b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.798847996708831"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs2.k_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "02eb84e1",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('Sex_male', 'Embarked_Q', 'Embarked_S', 'SibSp', 'Parch')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sfs2.k_feature_names_"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b24b9153",
   "metadata": {},
   "source": [
    "Inference : From the forward and backward feature engineering that features we can see that features 'Sex_male', 'Embarked_Q, 'Embarked_S','SibSp','Parch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "007f6724",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "201ca3f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/kumaransekar/opt/anaconda3/lib/python3.9/site-packages/statsmodels/tsa/tsatools.py:142: FutureWarning: In a future version of pandas all arguments of concat except for the argument 'objs' will be keyword-only\n",
      "  x = pd.concat(x[::order], 1)\n"
     ]
    }
   ],
   "source": [
    "x  = df_enc.drop(columns=['Survived'], axis =1)\n",
    "Y = df_enc['Survived']\n",
    "X = sm.add_constant(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "716f67cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,Y_train,Y_test = train_test_split(X,Y,test_size=0.3, stratify = None,random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "67464da2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(914, 9)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "e0901d1f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(392, 9)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "b7eee1ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(914,)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "c77bc383",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(392,)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "313d5b60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    61.816193\n",
       "1    38.183807\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(Y_train.value_counts() / len(Y_train))*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "a46b42c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    63.520408\n",
       "1    36.479592\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(Y_test.value_counts() / len(Y_test))*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "64b16294",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Base Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "895b179e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "00e33bef",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.381386\n",
      "         Iterations 7\n"
     ]
    }
   ],
   "source": [
    "LR_model = sm.Logit(Y_train,X_train).fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "1c783159",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Logit Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>       <td>Survived</td>     <th>  No. Observations:  </th>   <td>   914</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                 <td>Logit</td>      <th>  Df Residuals:      </th>   <td>   905</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                 <td>MLE</td>       <th>  Df Model:          </th>   <td>     8</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 06 May 2022</td> <th>  Pseudo R-squ.:     </th>   <td>0.4265</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>21:16:47</td>     <th>  Log-Likelihood:    </th>  <td> -348.59</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>converged:</th>             <td>True</td>       <th>  LL-Null:           </th>  <td> -607.77</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>     <td>nonrobust</td>    <th>  LLR p-value:       </th> <td>8.045e-107</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "       <td></td>         <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>      <td>    5.3136</td> <td>    0.599</td> <td>    8.865</td> <td> 0.000</td> <td>    4.139</td> <td>    6.488</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sex_male</th>   <td>   -3.5605</td> <td>    0.224</td> <td>  -15.897</td> <td> 0.000</td> <td>   -3.999</td> <td>   -3.121</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Embarked_Q</th> <td>    0.2594</td> <td>    0.415</td> <td>    0.625</td> <td> 0.532</td> <td>   -0.554</td> <td>    1.073</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Embarked_S</th> <td>   -0.1536</td> <td>    0.259</td> <td>   -0.593</td> <td> 0.553</td> <td>   -0.661</td> <td>    0.354</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Pclass</th>     <td>   -1.0986</td> <td>    0.156</td> <td>   -7.033</td> <td> 0.000</td> <td>   -1.405</td> <td>   -0.792</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Age</th>        <td>   -0.0370</td> <td>    0.008</td> <td>   -4.394</td> <td> 0.000</td> <td>   -0.054</td> <td>   -0.021</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>SibSp</th>      <td>   -0.2643</td> <td>    0.105</td> <td>   -2.526</td> <td> 0.012</td> <td>   -0.469</td> <td>   -0.059</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Parch</th>      <td>   -0.0385</td> <td>    0.107</td> <td>   -0.360</td> <td> 0.719</td> <td>   -0.248</td> <td>    0.171</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Fare</th>       <td>    0.0021</td> <td>    0.002</td> <td>    0.963</td> <td> 0.335</td> <td>   -0.002</td> <td>    0.006</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                           Logit Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:               Survived   No. Observations:                  914\n",
       "Model:                          Logit   Df Residuals:                      905\n",
       "Method:                           MLE   Df Model:                            8\n",
       "Date:                Fri, 06 May 2022   Pseudo R-squ.:                  0.4265\n",
       "Time:                        21:16:47   Log-Likelihood:                -348.59\n",
       "converged:                       True   LL-Null:                       -607.77\n",
       "Covariance Type:            nonrobust   LLR p-value:                8.045e-107\n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          5.3136      0.599      8.865      0.000       4.139       6.488\n",
       "Sex_male      -3.5605      0.224    -15.897      0.000      -3.999      -3.121\n",
       "Embarked_Q     0.2594      0.415      0.625      0.532      -0.554       1.073\n",
       "Embarked_S    -0.1536      0.259     -0.593      0.553      -0.661       0.354\n",
       "Pclass        -1.0986      0.156     -7.033      0.000      -1.405      -0.792\n",
       "Age           -0.0370      0.008     -4.394      0.000      -0.054      -0.021\n",
       "SibSp         -0.2643      0.105     -2.526      0.012      -0.469      -0.059\n",
       "Parch         -0.0385      0.107     -0.360      0.719      -0.248       0.171\n",
       "Fare           0.0021      0.002      0.963      0.335      -0.002       0.006\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LR_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "8a657992",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_LR = LR_model.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "8712bf8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_LR = LR_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "1d9ee6f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_LR_class = [0 if val<0.5 else 1 for val in y_pred_test_LR]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "775675e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_LR_class = [0 if val<0.5 else 1 for val in y_pred_train_LR]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "2571d064",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8435448577680525"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LR_train_accuarcy = accuracy_score(Y_train,y_pred_train_LR_class)\n",
    "LR_train_accuarcy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "273b8005",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8852040816326531"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LR_test_accuarcy = accuracy_score(Y_test,y_pred_test_LR_class)\n",
    "LR_test_accuarcy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "3ec7c4ab",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.748172681589767"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cohen_kappa_score_LR = cohen_kappa_score(Y_test,y_pred_test_LR_class)\n",
    "cohen_kappa_score_LR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "3703ebf5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "64db7c62",
   "metadata": {},
   "outputs": [],
   "source": [
    "Report_Card = pd.DataFrame(columns= ['Model','Train_Score','Test_Score','Cohen_kappa_score'], index = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "9a7c0999",
   "metadata": {},
   "outputs": [],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'Logit_regression', \n",
    "                    'Train_Score' : round(LR_train_accuarcy,2),\n",
    "                    'Test_Score': round(LR_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_LR,2)\n",
    "                   }, ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "319e1620",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0  Logit_regression         0.84        0.89               0.75"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "7374c7d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# KNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "9525d42f",
   "metadata": {},
   "outputs": [],
   "source": [
    "X  = df_enc.drop(columns=['Survived'], axis =1)\n",
    "Y = df_enc['Survived']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "49ebf5f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,Y_train,Y_test = train_test_split(X,Y,test_size=0.3, stratify = None,random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "3b176960",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn = KNeighborsClassifier()\n",
    "knn.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "2f4b3332",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_KNN = knn.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "331a0e59",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_KNN = knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "dd40863c",
   "metadata": {},
   "outputs": [],
   "source": [
    "KNN_train_accuarcy = accuracy_score(Y_train,y_pred_train_KNN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "8b614011",
   "metadata": {},
   "outputs": [],
   "source": [
    "KNN_test_accuarcy = accuracy_score(Y_test,y_pred_test_KNN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "701da7b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_KNN = cohen_kappa_score(Y_test,y_pred_test_KNN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "86f46e70",
   "metadata": {},
   "outputs": [],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'KNN', \n",
    "                    'Train_Score' : round(KNN_train_accuarcy,2),\n",
    "                    'Test_Score': round(KNN_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_KNN,2)\n",
    "                   }, ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "de995b28",
   "metadata": {
    "scrolled": true
   },
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0  Logit_regression         0.84        0.89               0.75\n",
       "1               KNN         0.81        0.70               0.34"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "3eca97b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "68dbf2bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "DT = DecisionTreeClassifier()\n",
    "DT.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "8841ed82",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_DT = DT.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "708aa2c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_DT = DT.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "c8321814",
   "metadata": {},
   "outputs": [],
   "source": [
    "DT_train_accuarcy = accuracy_score(Y_train,y_pred_train_DT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "8720aa97",
   "metadata": {},
   "outputs": [],
   "source": [
    "DT_test_accuarcy = accuracy_score(Y_test,y_pred_test_DT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "e935b9fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_DT = cohen_kappa_score(Y_test,y_pred_test_DT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "00866077",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision_tree</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0  Logit_regression         0.84        0.89               0.75\n",
       "1               KNN         0.81        0.70               0.34\n",
       "2     Decision_tree         0.98        0.81               0.59"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'Decision_tree', \n",
    "                    'Train_Score' : round(DT_train_accuarcy,2),\n",
    "                    'Test_Score': round(DT_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_DT,2)\n",
    "                   }, ignore_index=True)\n",
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "d35648f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Random_Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "7f9e4125",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(random_state=10)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "RF = RandomForestClassifier(random_state=10)\n",
    "RF.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "d6007b6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_RF = RF.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "d3bf2e9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_RF = RF.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "184b335f",
   "metadata": {},
   "outputs": [],
   "source": [
    "RF_train_accuarcy = accuracy_score(Y_train,y_pred_train_RF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "b8a11cfa",
   "metadata": {},
   "outputs": [],
   "source": [
    "RF_test_accuarcy = accuracy_score(Y_test,y_pred_test_RF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "8fd9ed71",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_RF = cohen_kappa_score(Y_test,y_pred_test_RF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "5203092b",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision_tree</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0  Logit_regression         0.84        0.89               0.75\n",
       "1               KNN         0.81        0.70               0.34\n",
       "2     Decision_tree         0.98        0.81               0.59\n",
       "3     Random_Forest         0.98        0.83               0.64"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'Random_Forest', \n",
    "                    'Train_Score' : round(RF_train_accuarcy,2),\n",
    "                    'Test_Score': round(RF_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_RF,2)\n",
    "                   }, ignore_index=True)\n",
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "eebbc241",
   "metadata": {},
   "outputs": [],
   "source": [
    "# AdaBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "e8694a16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AdaBoostClassifier()"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "ABC = AdaBoostClassifier()\n",
    "ABC.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "ff0f60ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_ABC = ABC.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "6701dcc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_ABC = ABC.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "8c430761",
   "metadata": {},
   "outputs": [],
   "source": [
    "ABC_train_accuarcy = accuracy_score(Y_train,y_pred_train_ABC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "0af3d4c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "ABC_test_accuarcy = accuracy_score(Y_test,y_pred_test_ABC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "8386e4ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_ABC = cohen_kappa_score(Y_test,y_pred_test_ABC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "42d5b3c2",
   "metadata": {
    "scrolled": true
   },
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision_tree</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.71</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0    Logit_regression         0.84        0.89               0.75\n",
       "1                 KNN         0.81        0.70               0.34\n",
       "2       Decision_tree         0.98        0.81               0.59\n",
       "3       Random_Forest         0.98        0.83               0.64\n",
       "4  AdaBoostClassifier         0.86        0.87               0.71"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'AdaBoostClassifier', \n",
    "                    'Train_Score' : round(ABC_train_accuarcy,2),\n",
    "                    'Test_Score': round(ABC_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_ABC,2)\n",
    "                   }, ignore_index=True)\n",
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "c229abac",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Gradient Boosting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "c58dfd30",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier()"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "GBC = GradientBoostingClassifier()\n",
    "GBC.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "d9afab4f",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_GBC = GBC.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "c4bfb15b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_GBC = GBC.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "0b89d73b",
   "metadata": {},
   "outputs": [],
   "source": [
    "GBC_train_accuarcy = accuracy_score(Y_train,y_pred_train_GBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "2d8cd886",
   "metadata": {},
   "outputs": [],
   "source": [
    "GBC_test_accuarcy = accuracy_score(Y_test,y_pred_test_GBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "391d2b40",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_GBC = cohen_kappa_score(Y_test,y_pred_test_GBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "61087ca9",
   "metadata": {
    "scrolled": true
   },
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision_tree</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GradientBoostingClassifier</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0            Logit_regression         0.84        0.89               0.75\n",
       "1                         KNN         0.81        0.70               0.34\n",
       "2               Decision_tree         0.98        0.81               0.59\n",
       "3               Random_Forest         0.98        0.83               0.64\n",
       "4          AdaBoostClassifier         0.86        0.87               0.71\n",
       "5  GradientBoostingClassifier         0.90        0.88               0.74"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'GradientBoostingClassifier', \n",
    "                    'Train_Score' : round(GBC_train_accuarcy,2),\n",
    "                    'Test_Score': round(GBC_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_GBC,2)\n",
    "                   }, ignore_index=True)\n",
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "df2c8f31",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Catboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "15fa1c43",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Learning rate set to 0.009914\n",
      "0:\tlearn: 0.6855950\ttotal: 65.3ms\tremaining: 1m 5s\n",
      "1:\tlearn: 0.6777298\ttotal: 66ms\tremaining: 32.9s\n",
      "2:\tlearn: 0.6703737\ttotal: 66.5ms\tremaining: 22.1s\n",
      "3:\tlearn: 0.6629707\ttotal: 67.1ms\tremaining: 16.7s\n",
      "4:\tlearn: 0.6558085\ttotal: 67.6ms\tremaining: 13.4s\n",
      "5:\tlearn: 0.6507981\ttotal: 67.8ms\tremaining: 11.2s\n",
      "6:\tlearn: 0.6445628\ttotal: 68.2ms\tremaining: 9.67s\n",
      "7:\tlearn: 0.6379237\ttotal: 68.6ms\tremaining: 8.51s\n",
      "8:\tlearn: 0.6316493\ttotal: 69.2ms\tremaining: 7.62s\n",
      "9:\tlearn: 0.6257914\ttotal: 69.6ms\tremaining: 6.89s\n",
      "10:\tlearn: 0.6194062\ttotal: 70.1ms\tremaining: 6.3s\n",
      "11:\tlearn: 0.6125312\ttotal: 70.8ms\tremaining: 5.83s\n",
      "12:\tlearn: 0.6062214\ttotal: 71.3ms\tremaining: 5.41s\n",
      "13:\tlearn: 0.6005299\ttotal: 71.8ms\tremaining: 5.06s\n",
      "14:\tlearn: 0.5937849\ttotal: 72.4ms\tremaining: 4.76s\n",
      "15:\tlearn: 0.5882969\ttotal: 73ms\tremaining: 4.49s\n",
      "16:\tlearn: 0.5831933\ttotal: 73.5ms\tremaining: 4.25s\n",
      "17:\tlearn: 0.5776039\ttotal: 74ms\tremaining: 4.04s\n",
      "18:\tlearn: 0.5721222\ttotal: 74.5ms\tremaining: 3.85s\n",
      "19:\tlearn: 0.5664450\ttotal: 75.1ms\tremaining: 3.68s\n",
      "20:\tlearn: 0.5617071\ttotal: 75.6ms\tremaining: 3.52s\n",
      "21:\tlearn: 0.5567224\ttotal: 76.1ms\tremaining: 3.38s\n",
      "22:\tlearn: 0.5523741\ttotal: 76.6ms\tremaining: 3.25s\n",
      "23:\tlearn: 0.5483345\ttotal: 77ms\tremaining: 3.13s\n",
      "24:\tlearn: 0.5434666\ttotal: 77.7ms\tremaining: 3.03s\n",
      "25:\tlearn: 0.5390838\ttotal: 78.5ms\tremaining: 2.94s\n",
      "26:\tlearn: 0.5350662\ttotal: 79ms\tremaining: 2.85s\n",
      "27:\tlearn: 0.5319343\ttotal: 79.4ms\tremaining: 2.76s\n",
      "28:\tlearn: 0.5287010\ttotal: 80ms\tremaining: 2.68s\n",
      "29:\tlearn: 0.5248499\ttotal: 80.5ms\tremaining: 2.6s\n",
      "30:\tlearn: 0.5202375\ttotal: 81.2ms\tremaining: 2.54s\n",
      "31:\tlearn: 0.5172710\ttotal: 82ms\tremaining: 2.48s\n",
      "32:\tlearn: 0.5141284\ttotal: 82.4ms\tremaining: 2.41s\n",
      "33:\tlearn: 0.5115176\ttotal: 82.8ms\tremaining: 2.35s\n",
      "34:\tlearn: 0.5084150\ttotal: 83.5ms\tremaining: 2.3s\n",
      "35:\tlearn: 0.5051125\ttotal: 84.1ms\tremaining: 2.25s\n",
      "36:\tlearn: 0.5014851\ttotal: 84.7ms\tremaining: 2.2s\n",
      "37:\tlearn: 0.4979923\ttotal: 85.4ms\tremaining: 2.16s\n",
      "38:\tlearn: 0.4944271\ttotal: 86ms\tremaining: 2.12s\n",
      "39:\tlearn: 0.4910918\ttotal: 86.6ms\tremaining: 2.08s\n",
      "40:\tlearn: 0.4890781\ttotal: 86.9ms\tremaining: 2.03s\n",
      "41:\tlearn: 0.4857127\ttotal: 87.6ms\tremaining: 2s\n",
      "42:\tlearn: 0.4824263\ttotal: 88.4ms\tremaining: 1.97s\n",
      "43:\tlearn: 0.4795991\ttotal: 89.1ms\tremaining: 1.93s\n",
      "44:\tlearn: 0.4768906\ttotal: 89.6ms\tremaining: 1.9s\n",
      "45:\tlearn: 0.4739929\ttotal: 90.3ms\tremaining: 1.87s\n",
      "46:\tlearn: 0.4706983\ttotal: 90.9ms\tremaining: 1.84s\n",
      "47:\tlearn: 0.4681925\ttotal: 91.6ms\tremaining: 1.82s\n",
      "48:\tlearn: 0.4652281\ttotal: 92.4ms\tremaining: 1.79s\n",
      "49:\tlearn: 0.4624495\ttotal: 93.1ms\tremaining: 1.77s\n",
      "50:\tlearn: 0.4599337\ttotal: 93.6ms\tremaining: 1.74s\n",
      "51:\tlearn: 0.4577292\ttotal: 94.2ms\tremaining: 1.72s\n",
      "52:\tlearn: 0.4553173\ttotal: 95ms\tremaining: 1.7s\n",
      "53:\tlearn: 0.4533092\ttotal: 95.4ms\tremaining: 1.67s\n",
      "54:\tlearn: 0.4508818\ttotal: 96ms\tremaining: 1.65s\n",
      "55:\tlearn: 0.4491523\ttotal: 96.7ms\tremaining: 1.63s\n",
      "56:\tlearn: 0.4465449\ttotal: 97.4ms\tremaining: 1.61s\n",
      "57:\tlearn: 0.4450784\ttotal: 97.9ms\tremaining: 1.59s\n",
      "58:\tlearn: 0.4433357\ttotal: 98.7ms\tremaining: 1.57s\n",
      "59:\tlearn: 0.4409154\ttotal: 99.4ms\tremaining: 1.56s\n",
      "60:\tlearn: 0.4387715\ttotal: 100ms\tremaining: 1.54s\n",
      "61:\tlearn: 0.4370839\ttotal: 100ms\tremaining: 1.52s\n",
      "62:\tlearn: 0.4358959\ttotal: 101ms\tremaining: 1.5s\n",
      "63:\tlearn: 0.4339623\ttotal: 101ms\tremaining: 1.48s\n",
      "64:\tlearn: 0.4326435\ttotal: 102ms\tremaining: 1.46s\n",
      "65:\tlearn: 0.4306481\ttotal: 102ms\tremaining: 1.45s\n",
      "66:\tlearn: 0.4290996\ttotal: 103ms\tremaining: 1.43s\n",
      "67:\tlearn: 0.4279990\ttotal: 103ms\tremaining: 1.42s\n",
      "68:\tlearn: 0.4265007\ttotal: 104ms\tremaining: 1.4s\n",
      "69:\tlearn: 0.4250783\ttotal: 105ms\tremaining: 1.39s\n",
      "70:\tlearn: 0.4231258\ttotal: 105ms\tremaining: 1.38s\n",
      "71:\tlearn: 0.4217666\ttotal: 106ms\tremaining: 1.36s\n",
      "72:\tlearn: 0.4202040\ttotal: 107ms\tremaining: 1.35s\n",
      "73:\tlearn: 0.4186766\ttotal: 107ms\tremaining: 1.34s\n",
      "74:\tlearn: 0.4177374\ttotal: 108ms\tremaining: 1.33s\n",
      "75:\tlearn: 0.4159712\ttotal: 108ms\tremaining: 1.32s\n",
      "76:\tlearn: 0.4144372\ttotal: 109ms\tremaining: 1.31s\n",
      "77:\tlearn: 0.4130430\ttotal: 110ms\tremaining: 1.3s\n",
      "78:\tlearn: 0.4117566\ttotal: 111ms\tremaining: 1.29s\n",
      "79:\tlearn: 0.4102010\ttotal: 111ms\tremaining: 1.28s\n",
      "80:\tlearn: 0.4090463\ttotal: 112ms\tremaining: 1.27s\n",
      "81:\tlearn: 0.4079430\ttotal: 113ms\tremaining: 1.26s\n",
      "82:\tlearn: 0.4065957\ttotal: 113ms\tremaining: 1.25s\n",
      "83:\tlearn: 0.4052350\ttotal: 114ms\tremaining: 1.24s\n",
      "84:\tlearn: 0.4039880\ttotal: 115ms\tremaining: 1.23s\n",
      "85:\tlearn: 0.4027509\ttotal: 115ms\tremaining: 1.22s\n",
      "86:\tlearn: 0.4017608\ttotal: 116ms\tremaining: 1.21s\n",
      "87:\tlearn: 0.4007069\ttotal: 116ms\tremaining: 1.2s\n",
      "88:\tlearn: 0.3995870\ttotal: 117ms\tremaining: 1.2s\n",
      "89:\tlearn: 0.3981251\ttotal: 118ms\tremaining: 1.19s\n",
      "90:\tlearn: 0.3970192\ttotal: 118ms\tremaining: 1.18s\n",
      "91:\tlearn: 0.3958366\ttotal: 119ms\tremaining: 1.17s\n",
      "92:\tlearn: 0.3946847\ttotal: 120ms\tremaining: 1.17s\n",
      "93:\tlearn: 0.3933629\ttotal: 120ms\tremaining: 1.16s\n",
      "94:\tlearn: 0.3922199\ttotal: 121ms\tremaining: 1.15s\n",
      "95:\tlearn: 0.3912114\ttotal: 122ms\tremaining: 1.15s\n",
      "96:\tlearn: 0.3902880\ttotal: 122ms\tremaining: 1.14s\n",
      "97:\tlearn: 0.3893674\ttotal: 123ms\tremaining: 1.13s\n",
      "98:\tlearn: 0.3884017\ttotal: 123ms\tremaining: 1.12s\n",
      "99:\tlearn: 0.3872422\ttotal: 124ms\tremaining: 1.12s\n",
      "100:\tlearn: 0.3861153\ttotal: 125ms\tremaining: 1.11s\n",
      "101:\tlearn: 0.3851450\ttotal: 125ms\tremaining: 1.1s\n",
      "102:\tlearn: 0.3842362\ttotal: 126ms\tremaining: 1.1s\n",
      "103:\tlearn: 0.3833879\ttotal: 126ms\tremaining: 1.09s\n",
      "104:\tlearn: 0.3824470\ttotal: 127ms\tremaining: 1.08s\n",
      "105:\tlearn: 0.3816113\ttotal: 128ms\tremaining: 1.08s\n",
      "106:\tlearn: 0.3804686\ttotal: 128ms\tremaining: 1.07s\n",
      "107:\tlearn: 0.3794146\ttotal: 129ms\tremaining: 1.06s\n",
      "108:\tlearn: 0.3788437\ttotal: 130ms\tremaining: 1.06s\n",
      "109:\tlearn: 0.3777697\ttotal: 130ms\tremaining: 1.05s\n",
      "110:\tlearn: 0.3768135\ttotal: 131ms\tremaining: 1.05s\n",
      "111:\tlearn: 0.3759040\ttotal: 131ms\tremaining: 1.04s\n",
      "112:\tlearn: 0.3750592\ttotal: 132ms\tremaining: 1.04s\n",
      "113:\tlearn: 0.3741316\ttotal: 133ms\tremaining: 1.03s\n",
      "114:\tlearn: 0.3732474\ttotal: 134ms\tremaining: 1.03s\n",
      "115:\tlearn: 0.3721872\ttotal: 134ms\tremaining: 1.02s\n",
      "116:\tlearn: 0.3715020\ttotal: 135ms\tremaining: 1.02s\n",
      "117:\tlearn: 0.3711817\ttotal: 135ms\tremaining: 1.01s\n",
      "118:\tlearn: 0.3705087\ttotal: 136ms\tremaining: 1.01s\n",
      "119:\tlearn: 0.3698495\ttotal: 137ms\tremaining: 1s\n",
      "120:\tlearn: 0.3691269\ttotal: 137ms\tremaining: 998ms\n",
      "121:\tlearn: 0.3686711\ttotal: 138ms\tremaining: 993ms\n",
      "122:\tlearn: 0.3677862\ttotal: 139ms\tremaining: 989ms\n",
      "123:\tlearn: 0.3670461\ttotal: 139ms\tremaining: 985ms\n",
      "124:\tlearn: 0.3665637\ttotal: 140ms\tremaining: 981ms\n",
      "125:\tlearn: 0.3659721\ttotal: 141ms\tremaining: 977ms\n",
      "126:\tlearn: 0.3652498\ttotal: 142ms\tremaining: 973ms\n",
      "127:\tlearn: 0.3647579\ttotal: 142ms\tremaining: 969ms\n",
      "128:\tlearn: 0.3644388\ttotal: 143ms\tremaining: 963ms\n",
      "129:\tlearn: 0.3638670\ttotal: 143ms\tremaining: 958ms\n",
      "130:\tlearn: 0.3633299\ttotal: 144ms\tremaining: 955ms\n",
      "131:\tlearn: 0.3627912\ttotal: 145ms\tremaining: 951ms\n",
      "132:\tlearn: 0.3622891\ttotal: 145ms\tremaining: 947ms\n",
      "133:\tlearn: 0.3618470\ttotal: 146ms\tremaining: 942ms\n",
      "134:\tlearn: 0.3613650\ttotal: 146ms\tremaining: 938ms\n",
      "135:\tlearn: 0.3609514\ttotal: 147ms\tremaining: 934ms\n",
      "136:\tlearn: 0.3606136\ttotal: 148ms\tremaining: 930ms\n",
      "137:\tlearn: 0.3603091\ttotal: 148ms\tremaining: 926ms\n",
      "138:\tlearn: 0.3599509\ttotal: 149ms\tremaining: 922ms\n",
      "139:\tlearn: 0.3594694\ttotal: 149ms\tremaining: 918ms\n",
      "140:\tlearn: 0.3590246\ttotal: 150ms\tremaining: 915ms\n",
      "141:\tlearn: 0.3585632\ttotal: 151ms\tremaining: 910ms\n",
      "142:\tlearn: 0.3581888\ttotal: 151ms\tremaining: 906ms\n",
      "143:\tlearn: 0.3579096\ttotal: 151ms\tremaining: 901ms\n",
      "144:\tlearn: 0.3574023\ttotal: 152ms\tremaining: 897ms\n",
      "145:\tlearn: 0.3571142\ttotal: 153ms\tremaining: 897ms\n",
      "146:\tlearn: 0.3567660\ttotal: 154ms\tremaining: 894ms\n",
      "147:\tlearn: 0.3561645\ttotal: 155ms\tremaining: 890ms\n",
      "148:\tlearn: 0.3559493\ttotal: 155ms\tremaining: 885ms\n",
      "149:\tlearn: 0.3556563\ttotal: 156ms\tremaining: 881ms\n",
      "150:\tlearn: 0.3549022\ttotal: 156ms\tremaining: 878ms\n",
      "151:\tlearn: 0.3544545\ttotal: 157ms\tremaining: 875ms\n",
      "152:\tlearn: 0.3538731\ttotal: 157ms\tremaining: 872ms\n",
      "153:\tlearn: 0.3534909\ttotal: 158ms\tremaining: 868ms\n",
      "154:\tlearn: 0.3530741\ttotal: 159ms\tremaining: 865ms\n",
      "155:\tlearn: 0.3526620\ttotal: 159ms\tremaining: 862ms\n",
      "156:\tlearn: 0.3523145\ttotal: 160ms\tremaining: 859ms\n",
      "157:\tlearn: 0.3521216\ttotal: 161ms\tremaining: 857ms\n",
      "158:\tlearn: 0.3516099\ttotal: 161ms\tremaining: 854ms\n",
      "159:\tlearn: 0.3511556\ttotal: 162ms\tremaining: 851ms\n",
      "160:\tlearn: 0.3509050\ttotal: 163ms\tremaining: 849ms\n",
      "161:\tlearn: 0.3504419\ttotal: 164ms\tremaining: 846ms\n",
      "162:\tlearn: 0.3501017\ttotal: 164ms\tremaining: 843ms\n",
      "163:\tlearn: 0.3496267\ttotal: 165ms\tremaining: 841ms\n",
      "164:\tlearn: 0.3491997\ttotal: 166ms\tremaining: 838ms\n",
      "165:\tlearn: 0.3486739\ttotal: 166ms\tremaining: 835ms\n",
      "166:\tlearn: 0.3482389\ttotal: 167ms\tremaining: 833ms\n",
      "167:\tlearn: 0.3478808\ttotal: 168ms\tremaining: 830ms\n",
      "168:\tlearn: 0.3475665\ttotal: 168ms\tremaining: 828ms\n",
      "169:\tlearn: 0.3472341\ttotal: 169ms\tremaining: 825ms\n",
      "170:\tlearn: 0.3467604\ttotal: 170ms\tremaining: 822ms\n",
      "171:\tlearn: 0.3463127\ttotal: 170ms\tremaining: 820ms\n",
      "172:\tlearn: 0.3460005\ttotal: 171ms\tremaining: 817ms\n",
      "173:\tlearn: 0.3457248\ttotal: 171ms\tremaining: 814ms\n",
      "174:\tlearn: 0.3454861\ttotal: 172ms\tremaining: 811ms\n",
      "175:\tlearn: 0.3451675\ttotal: 173ms\tremaining: 808ms\n",
      "176:\tlearn: 0.3448043\ttotal: 173ms\tremaining: 806ms\n",
      "177:\tlearn: 0.3446531\ttotal: 174ms\tremaining: 803ms\n",
      "178:\tlearn: 0.3444035\ttotal: 174ms\tremaining: 800ms\n",
      "179:\tlearn: 0.3443442\ttotal: 175ms\tremaining: 797ms\n",
      "180:\tlearn: 0.3438612\ttotal: 176ms\tremaining: 794ms\n",
      "181:\tlearn: 0.3436406\ttotal: 176ms\tremaining: 792ms\n",
      "182:\tlearn: 0.3431936\ttotal: 177ms\tremaining: 790ms\n",
      "183:\tlearn: 0.3428703\ttotal: 178ms\tremaining: 787ms\n",
      "184:\tlearn: 0.3425282\ttotal: 178ms\tremaining: 785ms\n",
      "185:\tlearn: 0.3420893\ttotal: 179ms\tremaining: 784ms\n",
      "186:\tlearn: 0.3418384\ttotal: 180ms\tremaining: 782ms\n",
      "187:\tlearn: 0.3413891\ttotal: 180ms\tremaining: 779ms\n",
      "188:\tlearn: 0.3410830\ttotal: 181ms\tremaining: 776ms\n",
      "189:\tlearn: 0.3410166\ttotal: 181ms\tremaining: 772ms\n",
      "190:\tlearn: 0.3407860\ttotal: 182ms\tremaining: 770ms\n",
      "191:\tlearn: 0.3405684\ttotal: 182ms\tremaining: 768ms\n",
      "192:\tlearn: 0.3401695\ttotal: 183ms\tremaining: 765ms\n",
      "193:\tlearn: 0.3398268\ttotal: 184ms\tremaining: 763ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "194:\tlearn: 0.3395393\ttotal: 185ms\tremaining: 762ms\n",
      "195:\tlearn: 0.3391449\ttotal: 185ms\tremaining: 760ms\n",
      "196:\tlearn: 0.3389483\ttotal: 186ms\tremaining: 758ms\n",
      "197:\tlearn: 0.3386158\ttotal: 187ms\tremaining: 756ms\n",
      "198:\tlearn: 0.3382408\ttotal: 187ms\tremaining: 754ms\n",
      "199:\tlearn: 0.3378603\ttotal: 188ms\tremaining: 752ms\n",
      "200:\tlearn: 0.3376011\ttotal: 189ms\tremaining: 750ms\n",
      "201:\tlearn: 0.3374282\ttotal: 189ms\tremaining: 748ms\n",
      "202:\tlearn: 0.3372757\ttotal: 190ms\tremaining: 746ms\n",
      "203:\tlearn: 0.3369380\ttotal: 191ms\tremaining: 743ms\n",
      "204:\tlearn: 0.3367745\ttotal: 191ms\tremaining: 741ms\n",
      "205:\tlearn: 0.3366656\ttotal: 192ms\tremaining: 739ms\n",
      "206:\tlearn: 0.3365703\ttotal: 192ms\tremaining: 737ms\n",
      "207:\tlearn: 0.3363591\ttotal: 193ms\tremaining: 735ms\n",
      "208:\tlearn: 0.3361954\ttotal: 194ms\tremaining: 733ms\n",
      "209:\tlearn: 0.3359532\ttotal: 194ms\tremaining: 731ms\n",
      "210:\tlearn: 0.3356916\ttotal: 195ms\tremaining: 729ms\n",
      "211:\tlearn: 0.3352643\ttotal: 196ms\tremaining: 727ms\n",
      "212:\tlearn: 0.3349926\ttotal: 196ms\tremaining: 725ms\n",
      "213:\tlearn: 0.3347437\ttotal: 197ms\tremaining: 724ms\n",
      "214:\tlearn: 0.3344774\ttotal: 198ms\tremaining: 722ms\n",
      "215:\tlearn: 0.3342274\ttotal: 198ms\tremaining: 720ms\n",
      "216:\tlearn: 0.3339454\ttotal: 199ms\tremaining: 718ms\n",
      "217:\tlearn: 0.3337248\ttotal: 200ms\tremaining: 717ms\n",
      "218:\tlearn: 0.3335262\ttotal: 201ms\tremaining: 715ms\n",
      "219:\tlearn: 0.3332873\ttotal: 201ms\tremaining: 713ms\n",
      "220:\tlearn: 0.3329887\ttotal: 202ms\tremaining: 711ms\n",
      "221:\tlearn: 0.3326698\ttotal: 203ms\tremaining: 710ms\n",
      "222:\tlearn: 0.3325756\ttotal: 204ms\tremaining: 709ms\n",
      "223:\tlearn: 0.3323900\ttotal: 204ms\tremaining: 707ms\n",
      "224:\tlearn: 0.3323443\ttotal: 205ms\tremaining: 705ms\n",
      "225:\tlearn: 0.3321920\ttotal: 205ms\tremaining: 704ms\n",
      "226:\tlearn: 0.3321434\ttotal: 206ms\tremaining: 701ms\n",
      "227:\tlearn: 0.3320082\ttotal: 207ms\tremaining: 700ms\n",
      "228:\tlearn: 0.3316863\ttotal: 207ms\tremaining: 698ms\n",
      "229:\tlearn: 0.3315543\ttotal: 208ms\tremaining: 697ms\n",
      "230:\tlearn: 0.3314230\ttotal: 209ms\tremaining: 695ms\n",
      "231:\tlearn: 0.3313917\ttotal: 209ms\tremaining: 692ms\n",
      "232:\tlearn: 0.3312790\ttotal: 210ms\tremaining: 690ms\n",
      "233:\tlearn: 0.3310783\ttotal: 210ms\tremaining: 688ms\n",
      "234:\tlearn: 0.3308225\ttotal: 211ms\tremaining: 688ms\n",
      "235:\tlearn: 0.3306997\ttotal: 212ms\tremaining: 686ms\n",
      "236:\tlearn: 0.3304126\ttotal: 213ms\tremaining: 685ms\n",
      "237:\tlearn: 0.3302882\ttotal: 213ms\tremaining: 683ms\n",
      "238:\tlearn: 0.3300984\ttotal: 214ms\tremaining: 682ms\n",
      "239:\tlearn: 0.3299692\ttotal: 215ms\tremaining: 680ms\n",
      "240:\tlearn: 0.3297601\ttotal: 216ms\tremaining: 679ms\n",
      "241:\tlearn: 0.3296622\ttotal: 216ms\tremaining: 677ms\n",
      "242:\tlearn: 0.3295317\ttotal: 217ms\tremaining: 676ms\n",
      "243:\tlearn: 0.3293484\ttotal: 218ms\tremaining: 674ms\n",
      "244:\tlearn: 0.3290929\ttotal: 218ms\tremaining: 673ms\n",
      "245:\tlearn: 0.3288864\ttotal: 219ms\tremaining: 671ms\n",
      "246:\tlearn: 0.3287244\ttotal: 220ms\tremaining: 669ms\n",
      "247:\tlearn: 0.3286443\ttotal: 220ms\tremaining: 667ms\n",
      "248:\tlearn: 0.3284599\ttotal: 221ms\tremaining: 666ms\n",
      "249:\tlearn: 0.3283536\ttotal: 222ms\tremaining: 665ms\n",
      "250:\tlearn: 0.3280107\ttotal: 222ms\tremaining: 663ms\n",
      "251:\tlearn: 0.3276583\ttotal: 223ms\tremaining: 662ms\n",
      "252:\tlearn: 0.3274567\ttotal: 224ms\tremaining: 661ms\n",
      "253:\tlearn: 0.3273186\ttotal: 224ms\tremaining: 659ms\n",
      "254:\tlearn: 0.3271119\ttotal: 225ms\tremaining: 657ms\n",
      "255:\tlearn: 0.3269645\ttotal: 225ms\tremaining: 655ms\n",
      "256:\tlearn: 0.3268268\ttotal: 226ms\tremaining: 654ms\n",
      "257:\tlearn: 0.3266060\ttotal: 227ms\tremaining: 653ms\n",
      "258:\tlearn: 0.3264786\ttotal: 228ms\tremaining: 651ms\n",
      "259:\tlearn: 0.3263261\ttotal: 228ms\tremaining: 649ms\n",
      "260:\tlearn: 0.3259551\ttotal: 229ms\tremaining: 648ms\n",
      "261:\tlearn: 0.3256807\ttotal: 229ms\tremaining: 646ms\n",
      "262:\tlearn: 0.3255080\ttotal: 230ms\tremaining: 645ms\n",
      "263:\tlearn: 0.3252371\ttotal: 231ms\tremaining: 643ms\n",
      "264:\tlearn: 0.3251318\ttotal: 231ms\tremaining: 642ms\n",
      "265:\tlearn: 0.3249323\ttotal: 232ms\tremaining: 640ms\n",
      "266:\tlearn: 0.3248553\ttotal: 233ms\tremaining: 639ms\n",
      "267:\tlearn: 0.3246765\ttotal: 233ms\tremaining: 637ms\n",
      "268:\tlearn: 0.3245596\ttotal: 234ms\tremaining: 636ms\n",
      "269:\tlearn: 0.3243783\ttotal: 235ms\tremaining: 634ms\n",
      "270:\tlearn: 0.3243511\ttotal: 235ms\tremaining: 632ms\n",
      "271:\tlearn: 0.3241529\ttotal: 236ms\tremaining: 631ms\n",
      "272:\tlearn: 0.3240755\ttotal: 236ms\tremaining: 629ms\n",
      "273:\tlearn: 0.3239010\ttotal: 237ms\tremaining: 628ms\n",
      "274:\tlearn: 0.3238050\ttotal: 237ms\tremaining: 626ms\n",
      "275:\tlearn: 0.3236518\ttotal: 238ms\tremaining: 625ms\n",
      "276:\tlearn: 0.3235911\ttotal: 239ms\tremaining: 624ms\n",
      "277:\tlearn: 0.3234570\ttotal: 240ms\tremaining: 623ms\n",
      "278:\tlearn: 0.3233565\ttotal: 241ms\tremaining: 622ms\n",
      "279:\tlearn: 0.3232357\ttotal: 241ms\tremaining: 620ms\n",
      "280:\tlearn: 0.3230368\ttotal: 242ms\tremaining: 619ms\n",
      "281:\tlearn: 0.3229441\ttotal: 243ms\tremaining: 618ms\n",
      "282:\tlearn: 0.3228735\ttotal: 243ms\tremaining: 617ms\n",
      "283:\tlearn: 0.3225691\ttotal: 244ms\tremaining: 615ms\n",
      "284:\tlearn: 0.3222853\ttotal: 245ms\tremaining: 614ms\n",
      "285:\tlearn: 0.3219777\ttotal: 246ms\tremaining: 613ms\n",
      "286:\tlearn: 0.3216815\ttotal: 246ms\tremaining: 612ms\n",
      "287:\tlearn: 0.3215964\ttotal: 247ms\tremaining: 610ms\n",
      "288:\tlearn: 0.3214448\ttotal: 247ms\tremaining: 609ms\n",
      "289:\tlearn: 0.3212467\ttotal: 248ms\tremaining: 607ms\n",
      "290:\tlearn: 0.3210852\ttotal: 249ms\tremaining: 606ms\n",
      "291:\tlearn: 0.3209610\ttotal: 250ms\tremaining: 605ms\n",
      "292:\tlearn: 0.3208251\ttotal: 250ms\tremaining: 604ms\n",
      "293:\tlearn: 0.3208078\ttotal: 251ms\tremaining: 602ms\n",
      "294:\tlearn: 0.3205497\ttotal: 251ms\tremaining: 601ms\n",
      "295:\tlearn: 0.3203302\ttotal: 252ms\tremaining: 599ms\n",
      "296:\tlearn: 0.3202661\ttotal: 253ms\tremaining: 598ms\n",
      "297:\tlearn: 0.3200853\ttotal: 253ms\tremaining: 597ms\n",
      "298:\tlearn: 0.3198449\ttotal: 254ms\tremaining: 595ms\n",
      "299:\tlearn: 0.3196678\ttotal: 254ms\tremaining: 594ms\n",
      "300:\tlearn: 0.3195523\ttotal: 255ms\tremaining: 592ms\n",
      "301:\tlearn: 0.3193099\ttotal: 256ms\tremaining: 591ms\n",
      "302:\tlearn: 0.3192932\ttotal: 256ms\tremaining: 590ms\n",
      "303:\tlearn: 0.3191969\ttotal: 257ms\tremaining: 588ms\n",
      "304:\tlearn: 0.3190631\ttotal: 258ms\tremaining: 587ms\n",
      "305:\tlearn: 0.3188760\ttotal: 258ms\tremaining: 586ms\n",
      "306:\tlearn: 0.3187615\ttotal: 259ms\tremaining: 585ms\n",
      "307:\tlearn: 0.3186680\ttotal: 260ms\tremaining: 583ms\n",
      "308:\tlearn: 0.3184896\ttotal: 260ms\tremaining: 582ms\n",
      "309:\tlearn: 0.3182425\ttotal: 261ms\tremaining: 581ms\n",
      "310:\tlearn: 0.3180936\ttotal: 262ms\tremaining: 580ms\n",
      "311:\tlearn: 0.3179603\ttotal: 263ms\tremaining: 579ms\n",
      "312:\tlearn: 0.3177037\ttotal: 263ms\tremaining: 578ms\n",
      "313:\tlearn: 0.3175681\ttotal: 264ms\tremaining: 577ms\n",
      "314:\tlearn: 0.3173988\ttotal: 265ms\tremaining: 576ms\n",
      "315:\tlearn: 0.3173231\ttotal: 266ms\tremaining: 575ms\n",
      "316:\tlearn: 0.3170934\ttotal: 267ms\tremaining: 574ms\n",
      "317:\tlearn: 0.3169449\ttotal: 267ms\tremaining: 573ms\n",
      "318:\tlearn: 0.3167510\ttotal: 268ms\tremaining: 572ms\n",
      "319:\tlearn: 0.3166652\ttotal: 268ms\tremaining: 571ms\n",
      "320:\tlearn: 0.3164799\ttotal: 269ms\tremaining: 569ms\n",
      "321:\tlearn: 0.3162518\ttotal: 270ms\tremaining: 568ms\n",
      "322:\tlearn: 0.3161865\ttotal: 270ms\tremaining: 567ms\n",
      "323:\tlearn: 0.3160081\ttotal: 271ms\tremaining: 566ms\n",
      "324:\tlearn: 0.3159265\ttotal: 272ms\tremaining: 565ms\n",
      "325:\tlearn: 0.3156668\ttotal: 273ms\tremaining: 564ms\n",
      "326:\tlearn: 0.3154471\ttotal: 273ms\tremaining: 563ms\n",
      "327:\tlearn: 0.3154181\ttotal: 274ms\tremaining: 561ms\n",
      "328:\tlearn: 0.3151216\ttotal: 275ms\tremaining: 560ms\n",
      "329:\tlearn: 0.3149131\ttotal: 275ms\tremaining: 559ms\n",
      "330:\tlearn: 0.3147600\ttotal: 276ms\tremaining: 558ms\n",
      "331:\tlearn: 0.3146551\ttotal: 277ms\tremaining: 557ms\n",
      "332:\tlearn: 0.3145514\ttotal: 277ms\tremaining: 556ms\n",
      "333:\tlearn: 0.3144373\ttotal: 278ms\tremaining: 555ms\n",
      "334:\tlearn: 0.3143576\ttotal: 279ms\tremaining: 553ms\n",
      "335:\tlearn: 0.3142108\ttotal: 279ms\tremaining: 552ms\n",
      "336:\tlearn: 0.3139999\ttotal: 280ms\tremaining: 551ms\n",
      "337:\tlearn: 0.3137811\ttotal: 281ms\tremaining: 550ms\n",
      "338:\tlearn: 0.3137139\ttotal: 281ms\tremaining: 549ms\n",
      "339:\tlearn: 0.3136237\ttotal: 282ms\tremaining: 548ms\n",
      "340:\tlearn: 0.3136053\ttotal: 283ms\tremaining: 546ms\n",
      "341:\tlearn: 0.3133537\ttotal: 283ms\tremaining: 545ms\n",
      "342:\tlearn: 0.3132286\ttotal: 284ms\tremaining: 544ms\n",
      "343:\tlearn: 0.3131433\ttotal: 285ms\tremaining: 543ms\n",
      "344:\tlearn: 0.3129721\ttotal: 285ms\tremaining: 542ms\n",
      "345:\tlearn: 0.3128574\ttotal: 286ms\tremaining: 541ms\n",
      "346:\tlearn: 0.3127687\ttotal: 287ms\tremaining: 540ms\n",
      "347:\tlearn: 0.3125943\ttotal: 288ms\tremaining: 539ms\n",
      "348:\tlearn: 0.3125383\ttotal: 288ms\tremaining: 538ms\n",
      "349:\tlearn: 0.3124608\ttotal: 289ms\tremaining: 536ms\n",
      "350:\tlearn: 0.3122189\ttotal: 289ms\tremaining: 535ms\n",
      "351:\tlearn: 0.3121273\ttotal: 290ms\tremaining: 534ms\n",
      "352:\tlearn: 0.3120136\ttotal: 291ms\tremaining: 533ms\n",
      "353:\tlearn: 0.3119396\ttotal: 291ms\tremaining: 532ms\n",
      "354:\tlearn: 0.3118104\ttotal: 292ms\tremaining: 531ms\n",
      "355:\tlearn: 0.3116765\ttotal: 293ms\tremaining: 530ms\n",
      "356:\tlearn: 0.3116398\ttotal: 294ms\tremaining: 529ms\n",
      "357:\tlearn: 0.3115027\ttotal: 294ms\tremaining: 528ms\n",
      "358:\tlearn: 0.3114889\ttotal: 295ms\tremaining: 526ms\n",
      "359:\tlearn: 0.3113754\ttotal: 295ms\tremaining: 525ms\n",
      "360:\tlearn: 0.3112157\ttotal: 296ms\tremaining: 524ms\n",
      "361:\tlearn: 0.3110862\ttotal: 296ms\tremaining: 523ms\n",
      "362:\tlearn: 0.3110505\ttotal: 297ms\tremaining: 521ms\n",
      "363:\tlearn: 0.3110355\ttotal: 297ms\tremaining: 520ms\n",
      "364:\tlearn: 0.3108755\ttotal: 298ms\tremaining: 519ms\n",
      "365:\tlearn: 0.3107355\ttotal: 299ms\tremaining: 517ms\n",
      "366:\tlearn: 0.3106663\ttotal: 300ms\tremaining: 517ms\n",
      "367:\tlearn: 0.3105493\ttotal: 300ms\tremaining: 516ms\n",
      "368:\tlearn: 0.3105143\ttotal: 301ms\tremaining: 514ms\n",
      "369:\tlearn: 0.3104171\ttotal: 302ms\tremaining: 513ms\n",
      "370:\tlearn: 0.3103152\ttotal: 302ms\tremaining: 512ms\n",
      "371:\tlearn: 0.3101556\ttotal: 303ms\tremaining: 512ms\n",
      "372:\tlearn: 0.3100980\ttotal: 304ms\tremaining: 511ms\n",
      "373:\tlearn: 0.3100173\ttotal: 304ms\tremaining: 509ms\n",
      "374:\tlearn: 0.3099683\ttotal: 305ms\tremaining: 508ms\n",
      "375:\tlearn: 0.3098487\ttotal: 305ms\tremaining: 507ms\n",
      "376:\tlearn: 0.3096833\ttotal: 306ms\tremaining: 506ms\n",
      "377:\tlearn: 0.3096428\ttotal: 307ms\tremaining: 504ms\n",
      "378:\tlearn: 0.3095686\ttotal: 307ms\tremaining: 503ms\n",
      "379:\tlearn: 0.3094117\ttotal: 308ms\tremaining: 502ms\n",
      "380:\tlearn: 0.3093361\ttotal: 309ms\tremaining: 501ms\n",
      "381:\tlearn: 0.3092003\ttotal: 309ms\tremaining: 500ms\n",
      "382:\tlearn: 0.3091545\ttotal: 310ms\tremaining: 499ms\n",
      "383:\tlearn: 0.3090386\ttotal: 311ms\tremaining: 498ms\n",
      "384:\tlearn: 0.3088773\ttotal: 311ms\tremaining: 497ms\n",
      "385:\tlearn: 0.3087700\ttotal: 312ms\tremaining: 496ms\n",
      "386:\tlearn: 0.3086046\ttotal: 313ms\tremaining: 495ms\n",
      "387:\tlearn: 0.3084305\ttotal: 313ms\tremaining: 494ms\n",
      "388:\tlearn: 0.3083155\ttotal: 314ms\tremaining: 493ms\n",
      "389:\tlearn: 0.3081776\ttotal: 315ms\tremaining: 492ms\n",
      "390:\tlearn: 0.3080922\ttotal: 315ms\tremaining: 491ms\n",
      "391:\tlearn: 0.3078331\ttotal: 316ms\tremaining: 490ms\n",
      "392:\tlearn: 0.3077257\ttotal: 317ms\tremaining: 489ms\n",
      "393:\tlearn: 0.3075916\ttotal: 317ms\tremaining: 488ms\n",
      "394:\tlearn: 0.3074658\ttotal: 318ms\tremaining: 487ms\n",
      "395:\tlearn: 0.3072967\ttotal: 319ms\tremaining: 486ms\n",
      "396:\tlearn: 0.3071550\ttotal: 319ms\tremaining: 485ms\n",
      "397:\tlearn: 0.3069790\ttotal: 320ms\tremaining: 484ms\n",
      "398:\tlearn: 0.3068960\ttotal: 321ms\tremaining: 483ms\n",
      "399:\tlearn: 0.3068487\ttotal: 322ms\tremaining: 483ms\n",
      "400:\tlearn: 0.3067470\ttotal: 322ms\tremaining: 482ms\n",
      "401:\tlearn: 0.3066929\ttotal: 323ms\tremaining: 481ms\n",
      "402:\tlearn: 0.3066360\ttotal: 324ms\tremaining: 480ms\n",
      "403:\tlearn: 0.3066072\ttotal: 324ms\tremaining: 478ms\n",
      "404:\tlearn: 0.3065381\ttotal: 325ms\tremaining: 478ms\n",
      "405:\tlearn: 0.3064650\ttotal: 326ms\tremaining: 477ms\n",
      "406:\tlearn: 0.3063176\ttotal: 326ms\tremaining: 475ms\n",
      "407:\tlearn: 0.3061803\ttotal: 327ms\tremaining: 474ms\n",
      "408:\tlearn: 0.3060195\ttotal: 328ms\tremaining: 473ms\n",
      "409:\tlearn: 0.3059397\ttotal: 328ms\tremaining: 472ms\n",
      "410:\tlearn: 0.3057390\ttotal: 329ms\tremaining: 472ms\n",
      "411:\tlearn: 0.3056412\ttotal: 330ms\tremaining: 471ms\n",
      "412:\tlearn: 0.3056387\ttotal: 330ms\tremaining: 469ms\n",
      "413:\tlearn: 0.3056272\ttotal: 330ms\tremaining: 468ms\n",
      "414:\tlearn: 0.3055369\ttotal: 331ms\tremaining: 467ms\n",
      "415:\tlearn: 0.3053949\ttotal: 332ms\tremaining: 466ms\n",
      "416:\tlearn: 0.3053295\ttotal: 332ms\tremaining: 465ms\n",
      "417:\tlearn: 0.3052226\ttotal: 333ms\tremaining: 464ms\n",
      "418:\tlearn: 0.3051299\ttotal: 334ms\tremaining: 463ms\n",
      "419:\tlearn: 0.3050166\ttotal: 335ms\tremaining: 462ms\n",
      "420:\tlearn: 0.3049376\ttotal: 335ms\tremaining: 461ms\n",
      "421:\tlearn: 0.3049230\ttotal: 336ms\tremaining: 460ms\n",
      "422:\tlearn: 0.3047377\ttotal: 336ms\tremaining: 459ms\n",
      "423:\tlearn: 0.3047115\ttotal: 337ms\tremaining: 458ms\n",
      "424:\tlearn: 0.3045693\ttotal: 338ms\tremaining: 457ms\n",
      "425:\tlearn: 0.3045106\ttotal: 339ms\tremaining: 456ms\n",
      "426:\tlearn: 0.3044387\ttotal: 339ms\tremaining: 455ms\n",
      "427:\tlearn: 0.3043124\ttotal: 340ms\tremaining: 454ms\n",
      "428:\tlearn: 0.3042053\ttotal: 340ms\tremaining: 453ms\n",
      "429:\tlearn: 0.3040660\ttotal: 341ms\tremaining: 452ms\n",
      "430:\tlearn: 0.3039634\ttotal: 342ms\tremaining: 451ms\n",
      "431:\tlearn: 0.3038319\ttotal: 343ms\tremaining: 450ms\n",
      "432:\tlearn: 0.3037166\ttotal: 343ms\tremaining: 449ms\n",
      "433:\tlearn: 0.3035446\ttotal: 344ms\tremaining: 448ms\n",
      "434:\tlearn: 0.3033403\ttotal: 344ms\tremaining: 447ms\n",
      "435:\tlearn: 0.3033223\ttotal: 345ms\tremaining: 446ms\n",
      "436:\tlearn: 0.3032637\ttotal: 345ms\tremaining: 445ms\n",
      "437:\tlearn: 0.3031924\ttotal: 346ms\tremaining: 444ms\n",
      "438:\tlearn: 0.3031050\ttotal: 347ms\tremaining: 443ms\n",
      "439:\tlearn: 0.3029648\ttotal: 347ms\tremaining: 442ms\n",
      "440:\tlearn: 0.3029372\ttotal: 348ms\tremaining: 441ms\n",
      "441:\tlearn: 0.3028499\ttotal: 349ms\tremaining: 440ms\n",
      "442:\tlearn: 0.3028054\ttotal: 350ms\tremaining: 439ms\n",
      "443:\tlearn: 0.3026852\ttotal: 350ms\tremaining: 439ms\n",
      "444:\tlearn: 0.3026321\ttotal: 351ms\tremaining: 438ms\n",
      "445:\tlearn: 0.3025259\ttotal: 352ms\tremaining: 437ms\n",
      "446:\tlearn: 0.3024816\ttotal: 352ms\tremaining: 436ms\n",
      "447:\tlearn: 0.3024469\ttotal: 353ms\tremaining: 435ms\n",
      "448:\tlearn: 0.3023854\ttotal: 354ms\tremaining: 434ms\n",
      "449:\tlearn: 0.3023286\ttotal: 355ms\tremaining: 434ms\n",
      "450:\tlearn: 0.3021733\ttotal: 355ms\tremaining: 433ms\n",
      "451:\tlearn: 0.3021282\ttotal: 356ms\tremaining: 432ms\n",
      "452:\tlearn: 0.3020004\ttotal: 357ms\tremaining: 431ms\n",
      "453:\tlearn: 0.3018618\ttotal: 357ms\tremaining: 430ms\n",
      "454:\tlearn: 0.3016664\ttotal: 358ms\tremaining: 429ms\n",
      "455:\tlearn: 0.3016314\ttotal: 359ms\tremaining: 428ms\n",
      "456:\tlearn: 0.3015352\ttotal: 359ms\tremaining: 427ms\n",
      "457:\tlearn: 0.3014125\ttotal: 360ms\tremaining: 426ms\n",
      "458:\tlearn: 0.3013452\ttotal: 361ms\tremaining: 425ms\n",
      "459:\tlearn: 0.3012631\ttotal: 362ms\tremaining: 425ms\n",
      "460:\tlearn: 0.3012480\ttotal: 362ms\tremaining: 423ms\n",
      "461:\tlearn: 0.3010790\ttotal: 363ms\tremaining: 422ms\n",
      "462:\tlearn: 0.3010136\ttotal: 363ms\tremaining: 421ms\n",
      "463:\tlearn: 0.3009883\ttotal: 364ms\tremaining: 420ms\n",
      "464:\tlearn: 0.3008822\ttotal: 365ms\tremaining: 419ms\n",
      "465:\tlearn: 0.3007485\ttotal: 365ms\tremaining: 418ms\n",
      "466:\tlearn: 0.3007113\ttotal: 366ms\tremaining: 418ms\n",
      "467:\tlearn: 0.3005752\ttotal: 367ms\tremaining: 417ms\n",
      "468:\tlearn: 0.3005212\ttotal: 367ms\tremaining: 416ms\n",
      "469:\tlearn: 0.3004786\ttotal: 368ms\tremaining: 415ms\n",
      "470:\tlearn: 0.3004299\ttotal: 369ms\tremaining: 414ms\n",
      "471:\tlearn: 0.3002557\ttotal: 369ms\tremaining: 413ms\n",
      "472:\tlearn: 0.3001743\ttotal: 370ms\tremaining: 412ms\n",
      "473:\tlearn: 0.3001024\ttotal: 371ms\tremaining: 411ms\n",
      "474:\tlearn: 0.3000167\ttotal: 371ms\tremaining: 411ms\n",
      "475:\tlearn: 0.2999233\ttotal: 372ms\tremaining: 410ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "476:\tlearn: 0.2998345\ttotal: 373ms\tremaining: 409ms\n",
      "477:\tlearn: 0.2998271\ttotal: 374ms\tremaining: 408ms\n",
      "478:\tlearn: 0.2997114\ttotal: 374ms\tremaining: 407ms\n",
      "479:\tlearn: 0.2995644\ttotal: 375ms\tremaining: 406ms\n",
      "480:\tlearn: 0.2995326\ttotal: 376ms\tremaining: 405ms\n",
      "481:\tlearn: 0.2994377\ttotal: 376ms\tremaining: 405ms\n",
      "482:\tlearn: 0.2993259\ttotal: 377ms\tremaining: 404ms\n",
      "483:\tlearn: 0.2991913\ttotal: 378ms\tremaining: 403ms\n",
      "484:\tlearn: 0.2991423\ttotal: 378ms\tremaining: 402ms\n",
      "485:\tlearn: 0.2990916\ttotal: 379ms\tremaining: 401ms\n",
      "486:\tlearn: 0.2990536\ttotal: 380ms\tremaining: 400ms\n",
      "487:\tlearn: 0.2988918\ttotal: 380ms\tremaining: 399ms\n",
      "488:\tlearn: 0.2987334\ttotal: 381ms\tremaining: 398ms\n",
      "489:\tlearn: 0.2986421\ttotal: 382ms\tremaining: 397ms\n",
      "490:\tlearn: 0.2985271\ttotal: 383ms\tremaining: 397ms\n",
      "491:\tlearn: 0.2983529\ttotal: 383ms\tremaining: 396ms\n",
      "492:\tlearn: 0.2981919\ttotal: 384ms\tremaining: 395ms\n",
      "493:\tlearn: 0.2981287\ttotal: 385ms\tremaining: 394ms\n",
      "494:\tlearn: 0.2980367\ttotal: 385ms\tremaining: 393ms\n",
      "495:\tlearn: 0.2978904\ttotal: 386ms\tremaining: 392ms\n",
      "496:\tlearn: 0.2978692\ttotal: 387ms\tremaining: 391ms\n",
      "497:\tlearn: 0.2978137\ttotal: 387ms\tremaining: 390ms\n",
      "498:\tlearn: 0.2977541\ttotal: 388ms\tremaining: 390ms\n",
      "499:\tlearn: 0.2976783\ttotal: 389ms\tremaining: 389ms\n",
      "500:\tlearn: 0.2975813\ttotal: 389ms\tremaining: 388ms\n",
      "501:\tlearn: 0.2975801\ttotal: 390ms\tremaining: 387ms\n",
      "502:\tlearn: 0.2975080\ttotal: 391ms\tremaining: 386ms\n",
      "503:\tlearn: 0.2973636\ttotal: 391ms\tremaining: 385ms\n",
      "504:\tlearn: 0.2973169\ttotal: 392ms\tremaining: 384ms\n",
      "505:\tlearn: 0.2972512\ttotal: 392ms\tremaining: 383ms\n",
      "506:\tlearn: 0.2972101\ttotal: 393ms\tremaining: 382ms\n",
      "507:\tlearn: 0.2971715\ttotal: 394ms\tremaining: 381ms\n",
      "508:\tlearn: 0.2970286\ttotal: 394ms\tremaining: 380ms\n",
      "509:\tlearn: 0.2968706\ttotal: 395ms\tremaining: 380ms\n",
      "510:\tlearn: 0.2967711\ttotal: 396ms\tremaining: 379ms\n",
      "511:\tlearn: 0.2965208\ttotal: 397ms\tremaining: 378ms\n",
      "512:\tlearn: 0.2964265\ttotal: 397ms\tremaining: 377ms\n",
      "513:\tlearn: 0.2963903\ttotal: 398ms\tremaining: 376ms\n",
      "514:\tlearn: 0.2962203\ttotal: 399ms\tremaining: 376ms\n",
      "515:\tlearn: 0.2961176\ttotal: 400ms\tremaining: 375ms\n",
      "516:\tlearn: 0.2960869\ttotal: 400ms\tremaining: 374ms\n",
      "517:\tlearn: 0.2958471\ttotal: 401ms\tremaining: 373ms\n",
      "518:\tlearn: 0.2957404\ttotal: 402ms\tremaining: 373ms\n",
      "519:\tlearn: 0.2956630\ttotal: 403ms\tremaining: 372ms\n",
      "520:\tlearn: 0.2955146\ttotal: 404ms\tremaining: 371ms\n",
      "521:\tlearn: 0.2954315\ttotal: 404ms\tremaining: 370ms\n",
      "522:\tlearn: 0.2953980\ttotal: 405ms\tremaining: 369ms\n",
      "523:\tlearn: 0.2952739\ttotal: 405ms\tremaining: 368ms\n",
      "524:\tlearn: 0.2951884\ttotal: 406ms\tremaining: 367ms\n",
      "525:\tlearn: 0.2951598\ttotal: 406ms\tremaining: 366ms\n",
      "526:\tlearn: 0.2949863\ttotal: 407ms\tremaining: 365ms\n",
      "527:\tlearn: 0.2949260\ttotal: 408ms\tremaining: 364ms\n",
      "528:\tlearn: 0.2948302\ttotal: 408ms\tremaining: 364ms\n",
      "529:\tlearn: 0.2947989\ttotal: 409ms\tremaining: 363ms\n",
      "530:\tlearn: 0.2947915\ttotal: 409ms\tremaining: 361ms\n",
      "531:\tlearn: 0.2947354\ttotal: 410ms\tremaining: 360ms\n",
      "532:\tlearn: 0.2946585\ttotal: 410ms\tremaining: 359ms\n",
      "533:\tlearn: 0.2946306\ttotal: 411ms\tremaining: 359ms\n",
      "534:\tlearn: 0.2945709\ttotal: 412ms\tremaining: 358ms\n",
      "535:\tlearn: 0.2943823\ttotal: 412ms\tremaining: 357ms\n",
      "536:\tlearn: 0.2942754\ttotal: 413ms\tremaining: 356ms\n",
      "537:\tlearn: 0.2941745\ttotal: 413ms\tremaining: 355ms\n",
      "538:\tlearn: 0.2941307\ttotal: 414ms\tremaining: 354ms\n",
      "539:\tlearn: 0.2939651\ttotal: 415ms\tremaining: 353ms\n",
      "540:\tlearn: 0.2937643\ttotal: 416ms\tremaining: 353ms\n",
      "541:\tlearn: 0.2936948\ttotal: 416ms\tremaining: 352ms\n",
      "542:\tlearn: 0.2935374\ttotal: 417ms\tremaining: 351ms\n",
      "543:\tlearn: 0.2934588\ttotal: 417ms\tremaining: 350ms\n",
      "544:\tlearn: 0.2934208\ttotal: 418ms\tremaining: 349ms\n",
      "545:\tlearn: 0.2933521\ttotal: 419ms\tremaining: 348ms\n",
      "546:\tlearn: 0.2933391\ttotal: 419ms\tremaining: 347ms\n",
      "547:\tlearn: 0.2932114\ttotal: 420ms\tremaining: 346ms\n",
      "548:\tlearn: 0.2931349\ttotal: 421ms\tremaining: 346ms\n",
      "549:\tlearn: 0.2930340\ttotal: 421ms\tremaining: 345ms\n",
      "550:\tlearn: 0.2929340\ttotal: 422ms\tremaining: 344ms\n",
      "551:\tlearn: 0.2929320\ttotal: 423ms\tremaining: 343ms\n",
      "552:\tlearn: 0.2928502\ttotal: 423ms\tremaining: 342ms\n",
      "553:\tlearn: 0.2928263\ttotal: 424ms\tremaining: 341ms\n",
      "554:\tlearn: 0.2927712\ttotal: 424ms\tremaining: 340ms\n",
      "555:\tlearn: 0.2927612\ttotal: 425ms\tremaining: 339ms\n",
      "556:\tlearn: 0.2927141\ttotal: 426ms\tremaining: 339ms\n",
      "557:\tlearn: 0.2926053\ttotal: 426ms\tremaining: 338ms\n",
      "558:\tlearn: 0.2925179\ttotal: 427ms\tremaining: 337ms\n",
      "559:\tlearn: 0.2924485\ttotal: 428ms\tremaining: 336ms\n",
      "560:\tlearn: 0.2923035\ttotal: 428ms\tremaining: 335ms\n",
      "561:\tlearn: 0.2922375\ttotal: 429ms\tremaining: 334ms\n",
      "562:\tlearn: 0.2921089\ttotal: 430ms\tremaining: 334ms\n",
      "563:\tlearn: 0.2920028\ttotal: 430ms\tremaining: 333ms\n",
      "564:\tlearn: 0.2919404\ttotal: 431ms\tremaining: 332ms\n",
      "565:\tlearn: 0.2918296\ttotal: 432ms\tremaining: 331ms\n",
      "566:\tlearn: 0.2917658\ttotal: 433ms\tremaining: 330ms\n",
      "567:\tlearn: 0.2916990\ttotal: 433ms\tremaining: 330ms\n",
      "568:\tlearn: 0.2916432\ttotal: 434ms\tremaining: 329ms\n",
      "569:\tlearn: 0.2915451\ttotal: 435ms\tremaining: 328ms\n",
      "570:\tlearn: 0.2914654\ttotal: 436ms\tremaining: 327ms\n",
      "571:\tlearn: 0.2913860\ttotal: 436ms\tremaining: 326ms\n",
      "572:\tlearn: 0.2913399\ttotal: 437ms\tremaining: 326ms\n",
      "573:\tlearn: 0.2911945\ttotal: 438ms\tremaining: 325ms\n",
      "574:\tlearn: 0.2910754\ttotal: 438ms\tremaining: 324ms\n",
      "575:\tlearn: 0.2910244\ttotal: 439ms\tremaining: 323ms\n",
      "576:\tlearn: 0.2909147\ttotal: 440ms\tremaining: 322ms\n",
      "577:\tlearn: 0.2909024\ttotal: 440ms\tremaining: 321ms\n",
      "578:\tlearn: 0.2908390\ttotal: 441ms\tremaining: 321ms\n",
      "579:\tlearn: 0.2906691\ttotal: 442ms\tremaining: 320ms\n",
      "580:\tlearn: 0.2905840\ttotal: 442ms\tremaining: 319ms\n",
      "581:\tlearn: 0.2905587\ttotal: 443ms\tremaining: 318ms\n",
      "582:\tlearn: 0.2904524\ttotal: 444ms\tremaining: 317ms\n",
      "583:\tlearn: 0.2904120\ttotal: 444ms\tremaining: 317ms\n",
      "584:\tlearn: 0.2903216\ttotal: 445ms\tremaining: 316ms\n",
      "585:\tlearn: 0.2901981\ttotal: 446ms\tremaining: 315ms\n",
      "586:\tlearn: 0.2901690\ttotal: 447ms\tremaining: 314ms\n",
      "587:\tlearn: 0.2900740\ttotal: 447ms\tremaining: 313ms\n",
      "588:\tlearn: 0.2899626\ttotal: 448ms\tremaining: 313ms\n",
      "589:\tlearn: 0.2899037\ttotal: 449ms\tremaining: 312ms\n",
      "590:\tlearn: 0.2898315\ttotal: 450ms\tremaining: 311ms\n",
      "591:\tlearn: 0.2897774\ttotal: 450ms\tremaining: 310ms\n",
      "592:\tlearn: 0.2897397\ttotal: 451ms\tremaining: 310ms\n",
      "593:\tlearn: 0.2895564\ttotal: 452ms\tremaining: 309ms\n",
      "594:\tlearn: 0.2895347\ttotal: 453ms\tremaining: 308ms\n",
      "595:\tlearn: 0.2892850\ttotal: 453ms\tremaining: 307ms\n",
      "596:\tlearn: 0.2891882\ttotal: 454ms\tremaining: 306ms\n",
      "597:\tlearn: 0.2891287\ttotal: 454ms\tremaining: 306ms\n",
      "598:\tlearn: 0.2891124\ttotal: 455ms\tremaining: 305ms\n",
      "599:\tlearn: 0.2890251\ttotal: 456ms\tremaining: 304ms\n",
      "600:\tlearn: 0.2888324\ttotal: 456ms\tremaining: 303ms\n",
      "601:\tlearn: 0.2888110\ttotal: 457ms\tremaining: 302ms\n",
      "602:\tlearn: 0.2886499\ttotal: 458ms\tremaining: 301ms\n",
      "603:\tlearn: 0.2885804\ttotal: 458ms\tremaining: 300ms\n",
      "604:\tlearn: 0.2885330\ttotal: 459ms\tremaining: 300ms\n",
      "605:\tlearn: 0.2885142\ttotal: 460ms\tremaining: 299ms\n",
      "606:\tlearn: 0.2884650\ttotal: 460ms\tremaining: 298ms\n",
      "607:\tlearn: 0.2883093\ttotal: 461ms\tremaining: 297ms\n",
      "608:\tlearn: 0.2882588\ttotal: 462ms\tremaining: 297ms\n",
      "609:\tlearn: 0.2881734\ttotal: 463ms\tremaining: 296ms\n",
      "610:\tlearn: 0.2881408\ttotal: 464ms\tremaining: 295ms\n",
      "611:\tlearn: 0.2880588\ttotal: 465ms\tremaining: 295ms\n",
      "612:\tlearn: 0.2879189\ttotal: 465ms\tremaining: 294ms\n",
      "613:\tlearn: 0.2878145\ttotal: 466ms\tremaining: 293ms\n",
      "614:\tlearn: 0.2877511\ttotal: 467ms\tremaining: 292ms\n",
      "615:\tlearn: 0.2876260\ttotal: 467ms\tremaining: 291ms\n",
      "616:\tlearn: 0.2875027\ttotal: 468ms\tremaining: 290ms\n",
      "617:\tlearn: 0.2874798\ttotal: 469ms\tremaining: 290ms\n",
      "618:\tlearn: 0.2874180\ttotal: 469ms\tremaining: 289ms\n",
      "619:\tlearn: 0.2872769\ttotal: 470ms\tremaining: 288ms\n",
      "620:\tlearn: 0.2871404\ttotal: 471ms\tremaining: 287ms\n",
      "621:\tlearn: 0.2870768\ttotal: 471ms\tremaining: 286ms\n",
      "622:\tlearn: 0.2869609\ttotal: 472ms\tremaining: 286ms\n",
      "623:\tlearn: 0.2868526\ttotal: 473ms\tremaining: 285ms\n",
      "624:\tlearn: 0.2867878\ttotal: 473ms\tremaining: 284ms\n",
      "625:\tlearn: 0.2866940\ttotal: 474ms\tremaining: 283ms\n",
      "626:\tlearn: 0.2864967\ttotal: 475ms\tremaining: 282ms\n",
      "627:\tlearn: 0.2864054\ttotal: 475ms\tremaining: 282ms\n",
      "628:\tlearn: 0.2861993\ttotal: 476ms\tremaining: 281ms\n",
      "629:\tlearn: 0.2860803\ttotal: 477ms\tremaining: 280ms\n",
      "630:\tlearn: 0.2860395\ttotal: 478ms\tremaining: 279ms\n",
      "631:\tlearn: 0.2859883\ttotal: 479ms\tremaining: 279ms\n",
      "632:\tlearn: 0.2857750\ttotal: 480ms\tremaining: 278ms\n",
      "633:\tlearn: 0.2857447\ttotal: 481ms\tremaining: 277ms\n",
      "634:\tlearn: 0.2856249\ttotal: 481ms\tremaining: 277ms\n",
      "635:\tlearn: 0.2855292\ttotal: 482ms\tremaining: 276ms\n",
      "636:\tlearn: 0.2854102\ttotal: 483ms\tremaining: 275ms\n",
      "637:\tlearn: 0.2852850\ttotal: 483ms\tremaining: 274ms\n",
      "638:\tlearn: 0.2852562\ttotal: 484ms\tremaining: 273ms\n",
      "639:\tlearn: 0.2851559\ttotal: 485ms\tremaining: 273ms\n",
      "640:\tlearn: 0.2850821\ttotal: 486ms\tremaining: 272ms\n",
      "641:\tlearn: 0.2850441\ttotal: 486ms\tremaining: 271ms\n",
      "642:\tlearn: 0.2849465\ttotal: 487ms\tremaining: 270ms\n",
      "643:\tlearn: 0.2848423\ttotal: 488ms\tremaining: 270ms\n",
      "644:\tlearn: 0.2847601\ttotal: 488ms\tremaining: 269ms\n",
      "645:\tlearn: 0.2846886\ttotal: 489ms\tremaining: 268ms\n",
      "646:\tlearn: 0.2846586\ttotal: 490ms\tremaining: 267ms\n",
      "647:\tlearn: 0.2846095\ttotal: 490ms\tremaining: 266ms\n",
      "648:\tlearn: 0.2845257\ttotal: 491ms\tremaining: 265ms\n",
      "649:\tlearn: 0.2843564\ttotal: 491ms\tremaining: 265ms\n",
      "650:\tlearn: 0.2843047\ttotal: 492ms\tremaining: 264ms\n",
      "651:\tlearn: 0.2841992\ttotal: 493ms\tremaining: 263ms\n",
      "652:\tlearn: 0.2841439\ttotal: 493ms\tremaining: 262ms\n",
      "653:\tlearn: 0.2840274\ttotal: 494ms\tremaining: 261ms\n",
      "654:\tlearn: 0.2839023\ttotal: 495ms\tremaining: 261ms\n",
      "655:\tlearn: 0.2838756\ttotal: 495ms\tremaining: 260ms\n",
      "656:\tlearn: 0.2838098\ttotal: 496ms\tremaining: 259ms\n",
      "657:\tlearn: 0.2837297\ttotal: 497ms\tremaining: 258ms\n",
      "658:\tlearn: 0.2836451\ttotal: 497ms\tremaining: 257ms\n",
      "659:\tlearn: 0.2835966\ttotal: 498ms\tremaining: 256ms\n",
      "660:\tlearn: 0.2834964\ttotal: 499ms\tremaining: 256ms\n",
      "661:\tlearn: 0.2834000\ttotal: 499ms\tremaining: 255ms\n",
      "662:\tlearn: 0.2833772\ttotal: 500ms\tremaining: 254ms\n",
      "663:\tlearn: 0.2833365\ttotal: 501ms\tremaining: 253ms\n",
      "664:\tlearn: 0.2832356\ttotal: 501ms\tremaining: 253ms\n",
      "665:\tlearn: 0.2831404\ttotal: 502ms\tremaining: 252ms\n",
      "666:\tlearn: 0.2831080\ttotal: 502ms\tremaining: 251ms\n",
      "667:\tlearn: 0.2829767\ttotal: 503ms\tremaining: 250ms\n",
      "668:\tlearn: 0.2828975\ttotal: 504ms\tremaining: 249ms\n",
      "669:\tlearn: 0.2828471\ttotal: 504ms\tremaining: 248ms\n",
      "670:\tlearn: 0.2827138\ttotal: 505ms\tremaining: 248ms\n",
      "671:\tlearn: 0.2825859\ttotal: 506ms\tremaining: 247ms\n",
      "672:\tlearn: 0.2824283\ttotal: 507ms\tremaining: 246ms\n",
      "673:\tlearn: 0.2823215\ttotal: 508ms\tremaining: 246ms\n",
      "674:\tlearn: 0.2822788\ttotal: 508ms\tremaining: 245ms\n",
      "675:\tlearn: 0.2822052\ttotal: 509ms\tremaining: 244ms\n",
      "676:\tlearn: 0.2821302\ttotal: 510ms\tremaining: 243ms\n",
      "677:\tlearn: 0.2820622\ttotal: 511ms\tremaining: 243ms\n",
      "678:\tlearn: 0.2819229\ttotal: 511ms\tremaining: 242ms\n",
      "679:\tlearn: 0.2818514\ttotal: 512ms\tremaining: 241ms\n",
      "680:\tlearn: 0.2817215\ttotal: 513ms\tremaining: 240ms\n",
      "681:\tlearn: 0.2816415\ttotal: 513ms\tremaining: 239ms\n",
      "682:\tlearn: 0.2814518\ttotal: 514ms\tremaining: 239ms\n",
      "683:\tlearn: 0.2814262\ttotal: 515ms\tremaining: 238ms\n",
      "684:\tlearn: 0.2813233\ttotal: 516ms\tremaining: 237ms\n",
      "685:\tlearn: 0.2812211\ttotal: 516ms\tremaining: 236ms\n",
      "686:\tlearn: 0.2811775\ttotal: 517ms\tremaining: 235ms\n",
      "687:\tlearn: 0.2811535\ttotal: 518ms\tremaining: 235ms\n",
      "688:\tlearn: 0.2810884\ttotal: 518ms\tremaining: 234ms\n",
      "689:\tlearn: 0.2810521\ttotal: 519ms\tremaining: 233ms\n",
      "690:\tlearn: 0.2810106\ttotal: 519ms\tremaining: 232ms\n",
      "691:\tlearn: 0.2809752\ttotal: 520ms\tremaining: 231ms\n",
      "692:\tlearn: 0.2808574\ttotal: 521ms\tremaining: 231ms\n",
      "693:\tlearn: 0.2807253\ttotal: 521ms\tremaining: 230ms\n",
      "694:\tlearn: 0.2805782\ttotal: 522ms\tremaining: 229ms\n",
      "695:\tlearn: 0.2805214\ttotal: 523ms\tremaining: 229ms\n",
      "696:\tlearn: 0.2804590\ttotal: 524ms\tremaining: 228ms\n",
      "697:\tlearn: 0.2803820\ttotal: 525ms\tremaining: 227ms\n",
      "698:\tlearn: 0.2802891\ttotal: 525ms\tremaining: 226ms\n",
      "699:\tlearn: 0.2802271\ttotal: 526ms\tremaining: 225ms\n",
      "700:\tlearn: 0.2800950\ttotal: 527ms\tremaining: 225ms\n",
      "701:\tlearn: 0.2800122\ttotal: 528ms\tremaining: 224ms\n",
      "702:\tlearn: 0.2798917\ttotal: 529ms\tremaining: 223ms\n",
      "703:\tlearn: 0.2798784\ttotal: 529ms\tremaining: 223ms\n",
      "704:\tlearn: 0.2798132\ttotal: 531ms\tremaining: 222ms\n",
      "705:\tlearn: 0.2797421\ttotal: 531ms\tremaining: 221ms\n",
      "706:\tlearn: 0.2797032\ttotal: 532ms\tremaining: 221ms\n",
      "707:\tlearn: 0.2795401\ttotal: 533ms\tremaining: 220ms\n",
      "708:\tlearn: 0.2794733\ttotal: 534ms\tremaining: 219ms\n",
      "709:\tlearn: 0.2793387\ttotal: 535ms\tremaining: 218ms\n",
      "710:\tlearn: 0.2792920\ttotal: 535ms\tremaining: 218ms\n",
      "711:\tlearn: 0.2792216\ttotal: 536ms\tremaining: 217ms\n",
      "712:\tlearn: 0.2791423\ttotal: 537ms\tremaining: 216ms\n",
      "713:\tlearn: 0.2791010\ttotal: 537ms\tremaining: 215ms\n",
      "714:\tlearn: 0.2790349\ttotal: 538ms\tremaining: 214ms\n",
      "715:\tlearn: 0.2789959\ttotal: 539ms\tremaining: 214ms\n",
      "716:\tlearn: 0.2788788\ttotal: 539ms\tremaining: 213ms\n",
      "717:\tlearn: 0.2787016\ttotal: 540ms\tremaining: 212ms\n",
      "718:\tlearn: 0.2785189\ttotal: 541ms\tremaining: 211ms\n",
      "719:\tlearn: 0.2784795\ttotal: 542ms\tremaining: 211ms\n",
      "720:\tlearn: 0.2783877\ttotal: 543ms\tremaining: 210ms\n",
      "721:\tlearn: 0.2783233\ttotal: 543ms\tremaining: 209ms\n",
      "722:\tlearn: 0.2782368\ttotal: 544ms\tremaining: 208ms\n",
      "723:\tlearn: 0.2781641\ttotal: 545ms\tremaining: 208ms\n",
      "724:\tlearn: 0.2779030\ttotal: 545ms\tremaining: 207ms\n",
      "725:\tlearn: 0.2777175\ttotal: 546ms\tremaining: 206ms\n",
      "726:\tlearn: 0.2776609\ttotal: 547ms\tremaining: 205ms\n",
      "727:\tlearn: 0.2775266\ttotal: 547ms\tremaining: 204ms\n",
      "728:\tlearn: 0.2773759\ttotal: 548ms\tremaining: 204ms\n",
      "729:\tlearn: 0.2773004\ttotal: 548ms\tremaining: 203ms\n",
      "730:\tlearn: 0.2772195\ttotal: 549ms\tremaining: 202ms\n",
      "731:\tlearn: 0.2772155\ttotal: 550ms\tremaining: 201ms\n",
      "732:\tlearn: 0.2771522\ttotal: 550ms\tremaining: 200ms\n",
      "733:\tlearn: 0.2770813\ttotal: 551ms\tremaining: 200ms\n",
      "734:\tlearn: 0.2770303\ttotal: 551ms\tremaining: 199ms\n",
      "735:\tlearn: 0.2768800\ttotal: 552ms\tremaining: 198ms\n",
      "736:\tlearn: 0.2767999\ttotal: 553ms\tremaining: 197ms\n",
      "737:\tlearn: 0.2767432\ttotal: 554ms\tremaining: 197ms\n",
      "738:\tlearn: 0.2766449\ttotal: 554ms\tremaining: 196ms\n",
      "739:\tlearn: 0.2764700\ttotal: 555ms\tremaining: 195ms\n",
      "740:\tlearn: 0.2764094\ttotal: 556ms\tremaining: 194ms\n",
      "741:\tlearn: 0.2763283\ttotal: 557ms\tremaining: 194ms\n",
      "742:\tlearn: 0.2762780\ttotal: 557ms\tremaining: 193ms\n",
      "743:\tlearn: 0.2762288\ttotal: 558ms\tremaining: 192ms\n",
      "744:\tlearn: 0.2761015\ttotal: 559ms\tremaining: 191ms\n",
      "745:\tlearn: 0.2760169\ttotal: 559ms\tremaining: 190ms\n",
      "746:\tlearn: 0.2759613\ttotal: 560ms\tremaining: 190ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "747:\tlearn: 0.2758222\ttotal: 561ms\tremaining: 189ms\n",
      "748:\tlearn: 0.2757200\ttotal: 562ms\tremaining: 188ms\n",
      "749:\tlearn: 0.2755574\ttotal: 563ms\tremaining: 188ms\n",
      "750:\tlearn: 0.2754857\ttotal: 564ms\tremaining: 187ms\n",
      "751:\tlearn: 0.2753963\ttotal: 565ms\tremaining: 186ms\n",
      "752:\tlearn: 0.2753460\ttotal: 565ms\tremaining: 185ms\n",
      "753:\tlearn: 0.2752656\ttotal: 566ms\tremaining: 185ms\n",
      "754:\tlearn: 0.2751611\ttotal: 567ms\tremaining: 184ms\n",
      "755:\tlearn: 0.2750688\ttotal: 567ms\tremaining: 183ms\n",
      "756:\tlearn: 0.2749330\ttotal: 568ms\tremaining: 182ms\n",
      "757:\tlearn: 0.2748557\ttotal: 569ms\tremaining: 182ms\n",
      "758:\tlearn: 0.2747443\ttotal: 569ms\tremaining: 181ms\n",
      "759:\tlearn: 0.2746507\ttotal: 570ms\tremaining: 180ms\n",
      "760:\tlearn: 0.2745336\ttotal: 571ms\tremaining: 179ms\n",
      "761:\tlearn: 0.2745304\ttotal: 571ms\tremaining: 178ms\n",
      "762:\tlearn: 0.2744388\ttotal: 572ms\tremaining: 178ms\n",
      "763:\tlearn: 0.2743361\ttotal: 573ms\tremaining: 177ms\n",
      "764:\tlearn: 0.2741750\ttotal: 574ms\tremaining: 176ms\n",
      "765:\tlearn: 0.2741400\ttotal: 575ms\tremaining: 176ms\n",
      "766:\tlearn: 0.2740660\ttotal: 575ms\tremaining: 175ms\n",
      "767:\tlearn: 0.2739928\ttotal: 576ms\tremaining: 174ms\n",
      "768:\tlearn: 0.2739226\ttotal: 577ms\tremaining: 173ms\n",
      "769:\tlearn: 0.2738043\ttotal: 578ms\tremaining: 173ms\n",
      "770:\tlearn: 0.2736254\ttotal: 578ms\tremaining: 172ms\n",
      "771:\tlearn: 0.2735373\ttotal: 579ms\tremaining: 171ms\n",
      "772:\tlearn: 0.2734390\ttotal: 580ms\tremaining: 170ms\n",
      "773:\tlearn: 0.2733257\ttotal: 581ms\tremaining: 170ms\n",
      "774:\tlearn: 0.2732193\ttotal: 581ms\tremaining: 169ms\n",
      "775:\tlearn: 0.2731380\ttotal: 582ms\tremaining: 168ms\n",
      "776:\tlearn: 0.2730012\ttotal: 582ms\tremaining: 167ms\n",
      "777:\tlearn: 0.2728662\ttotal: 583ms\tremaining: 166ms\n",
      "778:\tlearn: 0.2726997\ttotal: 584ms\tremaining: 166ms\n",
      "779:\tlearn: 0.2725780\ttotal: 585ms\tremaining: 165ms\n",
      "780:\tlearn: 0.2725145\ttotal: 585ms\tremaining: 164ms\n",
      "781:\tlearn: 0.2723433\ttotal: 586ms\tremaining: 163ms\n",
      "782:\tlearn: 0.2721883\ttotal: 587ms\tremaining: 163ms\n",
      "783:\tlearn: 0.2721794\ttotal: 587ms\tremaining: 162ms\n",
      "784:\tlearn: 0.2721545\ttotal: 588ms\tremaining: 161ms\n",
      "785:\tlearn: 0.2720836\ttotal: 589ms\tremaining: 160ms\n",
      "786:\tlearn: 0.2719498\ttotal: 589ms\tremaining: 160ms\n",
      "787:\tlearn: 0.2718394\ttotal: 590ms\tremaining: 159ms\n",
      "788:\tlearn: 0.2717162\ttotal: 591ms\tremaining: 158ms\n",
      "789:\tlearn: 0.2716422\ttotal: 592ms\tremaining: 157ms\n",
      "790:\tlearn: 0.2715845\ttotal: 592ms\tremaining: 157ms\n",
      "791:\tlearn: 0.2715248\ttotal: 593ms\tremaining: 156ms\n",
      "792:\tlearn: 0.2714787\ttotal: 594ms\tremaining: 155ms\n",
      "793:\tlearn: 0.2713926\ttotal: 594ms\tremaining: 154ms\n",
      "794:\tlearn: 0.2713037\ttotal: 595ms\tremaining: 153ms\n",
      "795:\tlearn: 0.2712280\ttotal: 596ms\tremaining: 153ms\n",
      "796:\tlearn: 0.2710743\ttotal: 597ms\tremaining: 152ms\n",
      "797:\tlearn: 0.2709982\ttotal: 597ms\tremaining: 151ms\n",
      "798:\tlearn: 0.2709216\ttotal: 598ms\tremaining: 150ms\n",
      "799:\tlearn: 0.2708469\ttotal: 599ms\tremaining: 150ms\n",
      "800:\tlearn: 0.2707729\ttotal: 599ms\tremaining: 149ms\n",
      "801:\tlearn: 0.2706519\ttotal: 600ms\tremaining: 148ms\n",
      "802:\tlearn: 0.2705958\ttotal: 601ms\tremaining: 147ms\n",
      "803:\tlearn: 0.2705541\ttotal: 601ms\tremaining: 147ms\n",
      "804:\tlearn: 0.2704778\ttotal: 602ms\tremaining: 146ms\n",
      "805:\tlearn: 0.2703997\ttotal: 603ms\tremaining: 145ms\n",
      "806:\tlearn: 0.2703596\ttotal: 603ms\tremaining: 144ms\n",
      "807:\tlearn: 0.2702303\ttotal: 604ms\tremaining: 143ms\n",
      "808:\tlearn: 0.2701626\ttotal: 605ms\tremaining: 143ms\n",
      "809:\tlearn: 0.2700649\ttotal: 605ms\tremaining: 142ms\n",
      "810:\tlearn: 0.2700191\ttotal: 606ms\tremaining: 141ms\n",
      "811:\tlearn: 0.2699431\ttotal: 606ms\tremaining: 140ms\n",
      "812:\tlearn: 0.2698867\ttotal: 607ms\tremaining: 140ms\n",
      "813:\tlearn: 0.2698309\ttotal: 608ms\tremaining: 139ms\n",
      "814:\tlearn: 0.2698068\ttotal: 609ms\tremaining: 138ms\n",
      "815:\tlearn: 0.2696642\ttotal: 609ms\tremaining: 137ms\n",
      "816:\tlearn: 0.2696008\ttotal: 610ms\tremaining: 137ms\n",
      "817:\tlearn: 0.2695302\ttotal: 611ms\tremaining: 136ms\n",
      "818:\tlearn: 0.2693911\ttotal: 611ms\tremaining: 135ms\n",
      "819:\tlearn: 0.2693259\ttotal: 612ms\tremaining: 134ms\n",
      "820:\tlearn: 0.2692495\ttotal: 612ms\tremaining: 134ms\n",
      "821:\tlearn: 0.2691633\ttotal: 613ms\tremaining: 133ms\n",
      "822:\tlearn: 0.2690476\ttotal: 614ms\tremaining: 132ms\n",
      "823:\tlearn: 0.2690202\ttotal: 614ms\tremaining: 131ms\n",
      "824:\tlearn: 0.2689352\ttotal: 615ms\tremaining: 130ms\n",
      "825:\tlearn: 0.2688748\ttotal: 616ms\tremaining: 130ms\n",
      "826:\tlearn: 0.2687797\ttotal: 616ms\tremaining: 129ms\n",
      "827:\tlearn: 0.2687126\ttotal: 617ms\tremaining: 128ms\n",
      "828:\tlearn: 0.2686538\ttotal: 617ms\tremaining: 127ms\n",
      "829:\tlearn: 0.2685354\ttotal: 618ms\tremaining: 127ms\n",
      "830:\tlearn: 0.2684915\ttotal: 619ms\tremaining: 126ms\n",
      "831:\tlearn: 0.2683619\ttotal: 619ms\tremaining: 125ms\n",
      "832:\tlearn: 0.2683025\ttotal: 620ms\tremaining: 124ms\n",
      "833:\tlearn: 0.2681660\ttotal: 621ms\tremaining: 124ms\n",
      "834:\tlearn: 0.2680908\ttotal: 621ms\tremaining: 123ms\n",
      "835:\tlearn: 0.2680447\ttotal: 622ms\tremaining: 122ms\n",
      "836:\tlearn: 0.2679273\ttotal: 622ms\tremaining: 121ms\n",
      "837:\tlearn: 0.2678113\ttotal: 623ms\tremaining: 120ms\n",
      "838:\tlearn: 0.2677689\ttotal: 624ms\tremaining: 120ms\n",
      "839:\tlearn: 0.2677161\ttotal: 625ms\tremaining: 119ms\n",
      "840:\tlearn: 0.2676199\ttotal: 625ms\tremaining: 118ms\n",
      "841:\tlearn: 0.2675347\ttotal: 626ms\tremaining: 117ms\n",
      "842:\tlearn: 0.2674904\ttotal: 626ms\tremaining: 117ms\n",
      "843:\tlearn: 0.2673733\ttotal: 627ms\tremaining: 116ms\n",
      "844:\tlearn: 0.2672932\ttotal: 628ms\tremaining: 115ms\n",
      "845:\tlearn: 0.2672545\ttotal: 628ms\tremaining: 114ms\n",
      "846:\tlearn: 0.2671769\ttotal: 629ms\tremaining: 114ms\n",
      "847:\tlearn: 0.2671362\ttotal: 630ms\tremaining: 113ms\n",
      "848:\tlearn: 0.2670937\ttotal: 630ms\tremaining: 112ms\n",
      "849:\tlearn: 0.2670268\ttotal: 632ms\tremaining: 111ms\n",
      "850:\tlearn: 0.2670042\ttotal: 633ms\tremaining: 111ms\n",
      "851:\tlearn: 0.2668447\ttotal: 633ms\tremaining: 110ms\n",
      "852:\tlearn: 0.2667932\ttotal: 634ms\tremaining: 109ms\n",
      "853:\tlearn: 0.2667089\ttotal: 634ms\tremaining: 108ms\n",
      "854:\tlearn: 0.2666538\ttotal: 635ms\tremaining: 108ms\n",
      "855:\tlearn: 0.2665748\ttotal: 636ms\tremaining: 107ms\n",
      "856:\tlearn: 0.2664636\ttotal: 636ms\tremaining: 106ms\n",
      "857:\tlearn: 0.2663827\ttotal: 637ms\tremaining: 105ms\n",
      "858:\tlearn: 0.2663339\ttotal: 638ms\tremaining: 105ms\n",
      "859:\tlearn: 0.2662571\ttotal: 638ms\tremaining: 104ms\n",
      "860:\tlearn: 0.2661633\ttotal: 639ms\tremaining: 103ms\n",
      "861:\tlearn: 0.2661253\ttotal: 639ms\tremaining: 102ms\n",
      "862:\tlearn: 0.2660902\ttotal: 640ms\tremaining: 102ms\n",
      "863:\tlearn: 0.2660328\ttotal: 641ms\tremaining: 101ms\n",
      "864:\tlearn: 0.2659809\ttotal: 641ms\tremaining: 100ms\n",
      "865:\tlearn: 0.2659040\ttotal: 642ms\tremaining: 99.3ms\n",
      "866:\tlearn: 0.2658256\ttotal: 643ms\tremaining: 98.6ms\n",
      "867:\tlearn: 0.2657332\ttotal: 643ms\tremaining: 97.8ms\n",
      "868:\tlearn: 0.2656294\ttotal: 644ms\tremaining: 97.1ms\n",
      "869:\tlearn: 0.2654970\ttotal: 645ms\tremaining: 96.3ms\n",
      "870:\tlearn: 0.2653849\ttotal: 645ms\tremaining: 95.6ms\n",
      "871:\tlearn: 0.2653595\ttotal: 646ms\tremaining: 94.8ms\n",
      "872:\tlearn: 0.2651930\ttotal: 647ms\tremaining: 94.1ms\n",
      "873:\tlearn: 0.2651246\ttotal: 647ms\tremaining: 93.3ms\n",
      "874:\tlearn: 0.2650509\ttotal: 648ms\tremaining: 92.5ms\n",
      "875:\tlearn: 0.2649912\ttotal: 649ms\tremaining: 91.8ms\n",
      "876:\tlearn: 0.2648833\ttotal: 649ms\tremaining: 91.1ms\n",
      "877:\tlearn: 0.2648104\ttotal: 650ms\tremaining: 90.3ms\n",
      "878:\tlearn: 0.2647431\ttotal: 650ms\tremaining: 89.5ms\n",
      "879:\tlearn: 0.2646933\ttotal: 651ms\tremaining: 88.8ms\n",
      "880:\tlearn: 0.2645831\ttotal: 652ms\tremaining: 88ms\n",
      "881:\tlearn: 0.2645709\ttotal: 652ms\tremaining: 87.3ms\n",
      "882:\tlearn: 0.2645449\ttotal: 653ms\tremaining: 86.5ms\n",
      "883:\tlearn: 0.2644652\ttotal: 654ms\tremaining: 85.8ms\n",
      "884:\tlearn: 0.2644049\ttotal: 654ms\tremaining: 85ms\n",
      "885:\tlearn: 0.2643803\ttotal: 655ms\tremaining: 84.3ms\n",
      "886:\tlearn: 0.2643338\ttotal: 656ms\tremaining: 83.5ms\n",
      "887:\tlearn: 0.2642734\ttotal: 656ms\tremaining: 82.8ms\n",
      "888:\tlearn: 0.2642304\ttotal: 657ms\tremaining: 82ms\n",
      "889:\tlearn: 0.2641918\ttotal: 658ms\tremaining: 81.3ms\n",
      "890:\tlearn: 0.2640418\ttotal: 658ms\tremaining: 80.5ms\n",
      "891:\tlearn: 0.2639362\ttotal: 659ms\tremaining: 79.8ms\n",
      "892:\tlearn: 0.2638536\ttotal: 659ms\tremaining: 79ms\n",
      "893:\tlearn: 0.2637420\ttotal: 660ms\tremaining: 78.3ms\n",
      "894:\tlearn: 0.2636940\ttotal: 661ms\tremaining: 77.5ms\n",
      "895:\tlearn: 0.2636152\ttotal: 662ms\tremaining: 76.8ms\n",
      "896:\tlearn: 0.2635656\ttotal: 662ms\tremaining: 76ms\n",
      "897:\tlearn: 0.2635247\ttotal: 663ms\tremaining: 75.3ms\n",
      "898:\tlearn: 0.2634601\ttotal: 664ms\tremaining: 74.5ms\n",
      "899:\tlearn: 0.2633579\ttotal: 664ms\tremaining: 73.8ms\n",
      "900:\tlearn: 0.2632939\ttotal: 665ms\tremaining: 73.1ms\n",
      "901:\tlearn: 0.2632288\ttotal: 666ms\tremaining: 72.3ms\n",
      "902:\tlearn: 0.2631159\ttotal: 666ms\tremaining: 71.6ms\n",
      "903:\tlearn: 0.2630408\ttotal: 667ms\tremaining: 70.8ms\n",
      "904:\tlearn: 0.2629883\ttotal: 667ms\tremaining: 70.1ms\n",
      "905:\tlearn: 0.2628759\ttotal: 668ms\tremaining: 69.3ms\n",
      "906:\tlearn: 0.2627762\ttotal: 669ms\tremaining: 68.6ms\n",
      "907:\tlearn: 0.2627216\ttotal: 669ms\tremaining: 67.8ms\n",
      "908:\tlearn: 0.2626669\ttotal: 670ms\tremaining: 67.1ms\n",
      "909:\tlearn: 0.2625715\ttotal: 671ms\tremaining: 66.3ms\n",
      "910:\tlearn: 0.2625116\ttotal: 672ms\tremaining: 65.6ms\n",
      "911:\tlearn: 0.2624625\ttotal: 672ms\tremaining: 64.8ms\n",
      "912:\tlearn: 0.2624057\ttotal: 673ms\tremaining: 64.1ms\n",
      "913:\tlearn: 0.2623001\ttotal: 673ms\tremaining: 63.4ms\n",
      "914:\tlearn: 0.2622136\ttotal: 674ms\tremaining: 62.6ms\n",
      "915:\tlearn: 0.2621741\ttotal: 675ms\tremaining: 61.9ms\n",
      "916:\tlearn: 0.2621582\ttotal: 675ms\tremaining: 61.1ms\n",
      "917:\tlearn: 0.2621209\ttotal: 676ms\tremaining: 60.4ms\n",
      "918:\tlearn: 0.2620493\ttotal: 677ms\tremaining: 59.7ms\n",
      "919:\tlearn: 0.2619846\ttotal: 678ms\tremaining: 58.9ms\n",
      "920:\tlearn: 0.2619244\ttotal: 678ms\tremaining: 58.2ms\n",
      "921:\tlearn: 0.2618836\ttotal: 679ms\tremaining: 57.5ms\n",
      "922:\tlearn: 0.2618391\ttotal: 680ms\tremaining: 56.7ms\n",
      "923:\tlearn: 0.2617922\ttotal: 680ms\tremaining: 56ms\n",
      "924:\tlearn: 0.2616677\ttotal: 681ms\tremaining: 55.2ms\n",
      "925:\tlearn: 0.2615721\ttotal: 682ms\tremaining: 54.5ms\n",
      "926:\tlearn: 0.2613791\ttotal: 682ms\tremaining: 53.7ms\n",
      "927:\tlearn: 0.2613187\ttotal: 683ms\tremaining: 53ms\n",
      "928:\tlearn: 0.2612451\ttotal: 684ms\tremaining: 52.2ms\n",
      "929:\tlearn: 0.2611513\ttotal: 684ms\tremaining: 51.5ms\n",
      "930:\tlearn: 0.2611305\ttotal: 685ms\tremaining: 50.8ms\n",
      "931:\tlearn: 0.2611147\ttotal: 686ms\tremaining: 50ms\n",
      "932:\tlearn: 0.2610829\ttotal: 687ms\tremaining: 49.3ms\n",
      "933:\tlearn: 0.2610238\ttotal: 687ms\tremaining: 48.6ms\n",
      "934:\tlearn: 0.2609851\ttotal: 688ms\tremaining: 47.8ms\n",
      "935:\tlearn: 0.2609699\ttotal: 689ms\tremaining: 47.1ms\n",
      "936:\tlearn: 0.2609145\ttotal: 689ms\tremaining: 46.3ms\n",
      "937:\tlearn: 0.2608562\ttotal: 690ms\tremaining: 45.6ms\n",
      "938:\tlearn: 0.2606699\ttotal: 691ms\tremaining: 44.9ms\n",
      "939:\tlearn: 0.2605617\ttotal: 691ms\tremaining: 44.1ms\n",
      "940:\tlearn: 0.2604650\ttotal: 692ms\tremaining: 43.4ms\n",
      "941:\tlearn: 0.2604206\ttotal: 693ms\tremaining: 42.6ms\n",
      "942:\tlearn: 0.2602751\ttotal: 693ms\tremaining: 41.9ms\n",
      "943:\tlearn: 0.2601291\ttotal: 694ms\tremaining: 41.2ms\n",
      "944:\tlearn: 0.2600287\ttotal: 695ms\tremaining: 40.4ms\n",
      "945:\tlearn: 0.2599803\ttotal: 695ms\tremaining: 39.7ms\n",
      "946:\tlearn: 0.2599213\ttotal: 696ms\tremaining: 38.9ms\n",
      "947:\tlearn: 0.2598235\ttotal: 697ms\tremaining: 38.2ms\n",
      "948:\tlearn: 0.2597407\ttotal: 697ms\tremaining: 37.5ms\n",
      "949:\tlearn: 0.2596421\ttotal: 698ms\tremaining: 36.7ms\n",
      "950:\tlearn: 0.2595577\ttotal: 699ms\tremaining: 36ms\n",
      "951:\tlearn: 0.2595366\ttotal: 699ms\tremaining: 35.3ms\n",
      "952:\tlearn: 0.2594915\ttotal: 700ms\tremaining: 34.5ms\n",
      "953:\tlearn: 0.2593586\ttotal: 701ms\tremaining: 33.8ms\n",
      "954:\tlearn: 0.2593088\ttotal: 701ms\tremaining: 33ms\n",
      "955:\tlearn: 0.2592481\ttotal: 702ms\tremaining: 32.3ms\n",
      "956:\tlearn: 0.2591562\ttotal: 703ms\tremaining: 31.6ms\n",
      "957:\tlearn: 0.2591095\ttotal: 704ms\tremaining: 30.8ms\n",
      "958:\tlearn: 0.2590656\ttotal: 704ms\tremaining: 30.1ms\n",
      "959:\tlearn: 0.2589966\ttotal: 705ms\tremaining: 29.4ms\n",
      "960:\tlearn: 0.2589570\ttotal: 706ms\tremaining: 28.6ms\n",
      "961:\tlearn: 0.2587948\ttotal: 706ms\tremaining: 27.9ms\n",
      "962:\tlearn: 0.2587453\ttotal: 707ms\tremaining: 27.2ms\n",
      "963:\tlearn: 0.2587204\ttotal: 708ms\tremaining: 26.4ms\n",
      "964:\tlearn: 0.2585535\ttotal: 709ms\tremaining: 25.7ms\n",
      "965:\tlearn: 0.2585187\ttotal: 709ms\tremaining: 25ms\n",
      "966:\tlearn: 0.2584789\ttotal: 710ms\tremaining: 24.2ms\n",
      "967:\tlearn: 0.2583782\ttotal: 710ms\tremaining: 23.5ms\n",
      "968:\tlearn: 0.2583313\ttotal: 711ms\tremaining: 22.7ms\n",
      "969:\tlearn: 0.2582963\ttotal: 712ms\tremaining: 22ms\n",
      "970:\tlearn: 0.2582476\ttotal: 713ms\tremaining: 21.3ms\n",
      "971:\tlearn: 0.2581764\ttotal: 713ms\tremaining: 20.5ms\n",
      "972:\tlearn: 0.2580777\ttotal: 714ms\tremaining: 19.8ms\n",
      "973:\tlearn: 0.2580014\ttotal: 715ms\tremaining: 19.1ms\n",
      "974:\tlearn: 0.2578564\ttotal: 715ms\tremaining: 18.3ms\n",
      "975:\tlearn: 0.2577739\ttotal: 716ms\tremaining: 17.6ms\n",
      "976:\tlearn: 0.2577058\ttotal: 717ms\tremaining: 16.9ms\n",
      "977:\tlearn: 0.2576693\ttotal: 717ms\tremaining: 16.1ms\n",
      "978:\tlearn: 0.2576251\ttotal: 718ms\tremaining: 15.4ms\n",
      "979:\tlearn: 0.2575577\ttotal: 719ms\tremaining: 14.7ms\n",
      "980:\tlearn: 0.2574599\ttotal: 719ms\tremaining: 13.9ms\n",
      "981:\tlearn: 0.2574419\ttotal: 720ms\tremaining: 13.2ms\n",
      "982:\tlearn: 0.2572823\ttotal: 721ms\tremaining: 12.5ms\n",
      "983:\tlearn: 0.2571710\ttotal: 722ms\tremaining: 11.7ms\n",
      "984:\tlearn: 0.2571457\ttotal: 722ms\tremaining: 11ms\n",
      "985:\tlearn: 0.2570506\ttotal: 723ms\tremaining: 10.3ms\n",
      "986:\tlearn: 0.2569420\ttotal: 724ms\tremaining: 9.53ms\n",
      "987:\tlearn: 0.2568287\ttotal: 724ms\tremaining: 8.8ms\n",
      "988:\tlearn: 0.2567428\ttotal: 725ms\tremaining: 8.06ms\n",
      "989:\tlearn: 0.2566604\ttotal: 726ms\tremaining: 7.33ms\n",
      "990:\tlearn: 0.2565893\ttotal: 726ms\tremaining: 6.6ms\n",
      "991:\tlearn: 0.2565196\ttotal: 727ms\tremaining: 5.86ms\n",
      "992:\tlearn: 0.2564509\ttotal: 728ms\tremaining: 5.13ms\n",
      "993:\tlearn: 0.2564196\ttotal: 729ms\tremaining: 4.4ms\n",
      "994:\tlearn: 0.2564096\ttotal: 729ms\tremaining: 3.66ms\n",
      "995:\tlearn: 0.2563272\ttotal: 730ms\tremaining: 2.93ms\n",
      "996:\tlearn: 0.2562156\ttotal: 731ms\tremaining: 2.2ms\n",
      "997:\tlearn: 0.2560615\ttotal: 731ms\tremaining: 1.47ms\n",
      "998:\tlearn: 0.2559430\ttotal: 732ms\tremaining: 732us\n",
      "999:\tlearn: 0.2559182\ttotal: 733ms\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x7f97ebd252b0>"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from catboost import CatBoostClassifier\n",
    "CBC = CatBoostClassifier()\n",
    "CBC.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "bf0bc406",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train_CBC = CBC.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "75b63c07",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_test_CBC = CBC.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "17ecce5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "CBC_train_accuarcy = accuracy_score(Y_train,y_pred_train_CBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "c7bc17fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "CBC_test_accuarcy = accuracy_score(Y_test,y_pred_test_CBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "5e62b769",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_CBC = cohen_kappa_score(Y_test,y_pred_test_CBC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "f7828650",
   "metadata": {
    "scrolled": false
   },
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logit_regression</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision_tree</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GradientBoostingClassifier</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CatBoostClassifier</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0.77</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0            Logit_regression         0.84        0.89               0.75\n",
       "1                         KNN         0.81        0.70               0.34\n",
       "2               Decision_tree         0.98        0.81               0.59\n",
       "3               Random_Forest         0.98        0.83               0.64\n",
       "4          AdaBoostClassifier         0.86        0.87               0.71\n",
       "5  GradientBoostingClassifier         0.90        0.88               0.74\n",
       "6          CatBoostClassifier         0.90        0.90               0.77"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card = Report_Card.append({'Model' : 'CatBoostClassifier', \n",
    "                    'Train_Score' : round(CBC_train_accuarcy,2),\n",
    "                    'Test_Score': round(CBC_test_accuarcy,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_CBC,2)\n",
    "                   }, ignore_index=True)\n",
    "Report_Card"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "eadc9d01",
   "metadata": {},
   "outputs": [],
   "source": [
    "# hypertuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "3aa07c99",
   "metadata": {},
   "outputs": [],
   "source": [
    "Report_Card2 = pd.DataFrame(columns= ['Model','Train_Score','Test_Score','Cohen_kappa_score'], index = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "73a62c5e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#AdaBoostClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18e3b6ae",
   "metadata": {},
   "source": [
    "from sklearn.model_selection import GridSearchCV </br>\n",
    "from sklearn.ensemble import AdaBoostClassifier </br>\n",
    "ABC = AdaBoostClassifier() </br>\n",
    "params = {'learning_rate' : [0.1,0.2,0.3,0.4,0.5,1,2,3,4,5,6,7], </br>\n",
    "          'n_estimators' : [100,200,300,400,500]} </br>\n",
    "ABC_CV = GridSearchCV(estimator=ABC, param_grid=params, scoring='f1', n_jobs=-1) </br>\n",
    "ABC_CV.fit(X_train,Y_train) </br>\n",
    "ABC_CV.best_params_ </br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "dc81cfa5",
   "metadata": {},
   "outputs": [],
   "source": [
    "ABC_200 = AdaBoostClassifier(learning_rate = 0.5, n_estimators = 200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "1d07e7d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AdaBoostClassifier(learning_rate=0.5, n_estimators=200)"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ABC_200.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "f3ade26e",
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_training_ABC_200 = accuracy_score(Y_train,ABC_200.predict(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "84443572",
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_testing_ABC_200 = accuracy_score(Y_test,ABC_200.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "aad2d197",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_ABC_200 = cohen_kappa_score(Y_test,ABC_200.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "16c159ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7018675530266754"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cohen_kappa_score_ABC_200"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "3a2b790d",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AdaBoost_hyper_tuned</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0  AdaBoost_hyper_tuned         0.87        0.86                0.7"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card2 = Report_Card2.append({'Model' : 'AdaBoost_hyper_tuned', \n",
    "                    'Train_Score' : round(accuracy_score_training_ABC_200,2),\n",
    "                    'Test_Score': round(accuracy_score_testing_ABC_200,2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_ABC_200,2)\n",
    "                    }, ignore_index=True)\n",
    "Report_Card2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "52378e80",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Gradient_decent"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92235ef8",
   "metadata": {},
   "source": [
    "from sklearn.model_selection import GridSearchCV</br>\n",
    "from sklearn.ensemble import GradientBoostingClassifier</br>\n",
    "GBC = GradientBoostingClassifier()</br>\n",
    "params = {'learning_rate' : [0.1,0.2,0.3,0.4,0.5,1,2,3], </br>\n",
    "          'n_estimators' : [100,150,200,250,300,400,500], </br>\n",
    "          'min_samples_split' : [2,3,4,5,6,7,8,9,10],</br>\n",
    "          'max_depth' : [1,2,3,4,5,6,7,8,9,10],</br>\n",
    "         'min_samples_leaf' : [1,2,3,4,5,6,7,8,9,10]}</br>\n",
    "GBC_CV = GridSearchCV(estimator=GBC, param_grid=params, scoring='accuracy', n_jobs=-1)</br>\n",
    "GBC_CV.fit(X_train,Y_train)</br>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7740337e",
   "metadata": {},
   "source": [
    "GBC_CV.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "89cbd556",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(learning_rate=0.2, max_depth=2, min_samples_leaf=2,\n",
       "                           min_samples_split=5)"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "GBC_CV_hyper_tune = GradientBoostingClassifier(learning_rate = 0.2,\n",
    "                                               max_depth = 2,\n",
    "                                               min_samples_leaf = 2,\n",
    "                                               min_samples_split = 5,\n",
    "                                               n_estimators = 100)\n",
    "GBC_CV_hyper_tune.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "09df2b86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8927789934354485"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(Y_train,GBC_CV_hyper_tune.predict(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "38c8d831",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8801020408163265"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(Y_test,GBC_CV_hyper_tune.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "2fc577e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7377739823512668"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cohen_kappa_score_GBC = cohen_kappa_score(Y_test,GBC_CV_hyper_tune.predict(X_test))\n",
    "cohen_kappa_score_GBC "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "28deee5b",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AdaBoost_hyper_tuned</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Graident_Boosting_hypertune</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0         AdaBoost_hyper_tuned         0.87        0.86               0.70\n",
       "1  Graident_Boosting_hypertune         0.89        0.88               0.74"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card2 = Report_Card2.append({'Model' : 'Graident_Boosting_hypertune', \n",
    "                    'Train_Score' : round(accuracy_score(Y_train,GBC_CV_hyper_tune.predict(X_train)),2),\n",
    "                    'Test_Score': round(accuracy_score(Y_test,GBC_CV_hyper_tune.predict(X_test)),2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_GBC,2)\n",
    "                    }, ignore_index=True)\n",
    "Report_Card2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "7c95d717",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Catboost"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c527882e",
   "metadata": {},
   "source": [
    "from catboost import CatBoostClassifier </br>\n",
    "CBC = CatBoostClassifier() </br>\n",
    "params = {'iterations' :  [500,600,700], 'depth' : [6,7,8,9,10]} </br>\n",
    "CBC_CV = GridSearchCV(estimator = CBC, param_grid= params, scoring='accuracy', n_jobs=-1) </br>\n",
    "CBC_CV.fit(X_train,Y_train) </br>\n",
    "CBC_CV.best_params_ </br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "d29215ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "CBC_500 = CatBoostClassifier(depth=6, iterations = 500, learning_rate = 0.02)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "cf2f74ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.6780217\ttotal: 704us\tremaining: 352ms\n",
      "1:\tlearn: 0.6626512\ttotal: 1.39ms\tremaining: 346ms\n",
      "2:\tlearn: 0.6486049\ttotal: 2.11ms\tremaining: 350ms\n",
      "3:\tlearn: 0.6348175\ttotal: 2.77ms\tremaining: 344ms\n",
      "4:\tlearn: 0.6218342\ttotal: 3.4ms\tremaining: 337ms\n",
      "5:\tlearn: 0.6130515\ttotal: 3.86ms\tremaining: 318ms\n",
      "6:\tlearn: 0.6021877\ttotal: 4.54ms\tremaining: 319ms\n",
      "7:\tlearn: 0.5906502\ttotal: 5.2ms\tremaining: 320ms\n",
      "8:\tlearn: 0.5806675\ttotal: 5.92ms\tremaining: 323ms\n",
      "9:\tlearn: 0.5708019\ttotal: 6.72ms\tremaining: 329ms\n",
      "10:\tlearn: 0.5595046\ttotal: 7.27ms\tremaining: 323ms\n",
      "11:\tlearn: 0.5494882\ttotal: 7.86ms\tremaining: 320ms\n",
      "12:\tlearn: 0.5406779\ttotal: 8.42ms\tremaining: 315ms\n",
      "13:\tlearn: 0.5302820\ttotal: 8.95ms\tremaining: 311ms\n",
      "14:\tlearn: 0.5221786\ttotal: 9.54ms\tremaining: 308ms\n",
      "15:\tlearn: 0.5149810\ttotal: 10.7ms\tremaining: 324ms\n",
      "16:\tlearn: 0.5070847\ttotal: 11.4ms\tremaining: 325ms\n",
      "17:\tlearn: 0.5024575\ttotal: 12ms\tremaining: 322ms\n",
      "18:\tlearn: 0.4952256\ttotal: 12.7ms\tremaining: 322ms\n",
      "19:\tlearn: 0.4885143\ttotal: 13.4ms\tremaining: 322ms\n",
      "20:\tlearn: 0.4823637\ttotal: 14ms\tremaining: 320ms\n",
      "21:\tlearn: 0.4764177\ttotal: 14.6ms\tremaining: 317ms\n",
      "22:\tlearn: 0.4716291\ttotal: 15.2ms\tremaining: 314ms\n",
      "23:\tlearn: 0.4662957\ttotal: 16ms\tremaining: 317ms\n",
      "24:\tlearn: 0.4617510\ttotal: 16.7ms\tremaining: 318ms\n",
      "25:\tlearn: 0.4575835\ttotal: 17.4ms\tremaining: 318ms\n",
      "26:\tlearn: 0.4521298\ttotal: 19ms\tremaining: 333ms\n",
      "27:\tlearn: 0.4493678\ttotal: 19.3ms\tremaining: 326ms\n",
      "28:\tlearn: 0.4446701\ttotal: 20.2ms\tremaining: 328ms\n",
      "29:\tlearn: 0.4418411\ttotal: 20.6ms\tremaining: 323ms\n",
      "30:\tlearn: 0.4382309\ttotal: 21.3ms\tremaining: 322ms\n",
      "31:\tlearn: 0.4336026\ttotal: 21.9ms\tremaining: 321ms\n",
      "32:\tlearn: 0.4298057\ttotal: 22.6ms\tremaining: 320ms\n",
      "33:\tlearn: 0.4269444\ttotal: 23ms\tremaining: 315ms\n",
      "34:\tlearn: 0.4236120\ttotal: 23.6ms\tremaining: 314ms\n",
      "35:\tlearn: 0.4206845\ttotal: 24.3ms\tremaining: 313ms\n",
      "36:\tlearn: 0.4176598\ttotal: 25ms\tremaining: 313ms\n",
      "37:\tlearn: 0.4140740\ttotal: 25.8ms\tremaining: 314ms\n",
      "38:\tlearn: 0.4113247\ttotal: 26.5ms\tremaining: 313ms\n",
      "39:\tlearn: 0.4086020\ttotal: 27.6ms\tremaining: 318ms\n",
      "40:\tlearn: 0.4060087\ttotal: 28.4ms\tremaining: 318ms\n",
      "41:\tlearn: 0.4038823\ttotal: 29.2ms\tremaining: 318ms\n",
      "42:\tlearn: 0.4025944\ttotal: 29.4ms\tremaining: 313ms\n",
      "43:\tlearn: 0.4000537\ttotal: 30.1ms\tremaining: 311ms\n",
      "44:\tlearn: 0.3975304\ttotal: 30.8ms\tremaining: 312ms\n",
      "45:\tlearn: 0.3953789\ttotal: 31.6ms\tremaining: 312ms\n",
      "46:\tlearn: 0.3930707\ttotal: 32.3ms\tremaining: 311ms\n",
      "47:\tlearn: 0.3905387\ttotal: 32.9ms\tremaining: 310ms\n",
      "48:\tlearn: 0.3886319\ttotal: 33.6ms\tremaining: 309ms\n",
      "49:\tlearn: 0.3866335\ttotal: 34.3ms\tremaining: 309ms\n",
      "50:\tlearn: 0.3851309\ttotal: 34.9ms\tremaining: 307ms\n",
      "51:\tlearn: 0.3829486\ttotal: 35.6ms\tremaining: 306ms\n",
      "52:\tlearn: 0.3808798\ttotal: 36.3ms\tremaining: 306ms\n",
      "53:\tlearn: 0.3794542\ttotal: 37ms\tremaining: 306ms\n",
      "54:\tlearn: 0.3777553\ttotal: 37.7ms\tremaining: 305ms\n",
      "55:\tlearn: 0.3762850\ttotal: 38.2ms\tremaining: 303ms\n",
      "56:\tlearn: 0.3752928\ttotal: 38.8ms\tremaining: 302ms\n",
      "57:\tlearn: 0.3738933\ttotal: 39.5ms\tremaining: 301ms\n",
      "58:\tlearn: 0.3726782\ttotal: 40.1ms\tremaining: 300ms\n",
      "59:\tlearn: 0.3710386\ttotal: 40.8ms\tremaining: 300ms\n",
      "60:\tlearn: 0.3692466\ttotal: 41.6ms\tremaining: 300ms\n",
      "61:\tlearn: 0.3671189\ttotal: 42.3ms\tremaining: 299ms\n",
      "62:\tlearn: 0.3663132\ttotal: 43.1ms\tremaining: 299ms\n",
      "63:\tlearn: 0.3651257\ttotal: 43.8ms\tremaining: 299ms\n",
      "64:\tlearn: 0.3638974\ttotal: 44.5ms\tremaining: 298ms\n",
      "65:\tlearn: 0.3628899\ttotal: 45.2ms\tremaining: 297ms\n",
      "66:\tlearn: 0.3621818\ttotal: 45.6ms\tremaining: 295ms\n",
      "67:\tlearn: 0.3607171\ttotal: 46.4ms\tremaining: 295ms\n",
      "68:\tlearn: 0.3596355\ttotal: 47ms\tremaining: 294ms\n",
      "69:\tlearn: 0.3582675\ttotal: 47.9ms\tremaining: 294ms\n",
      "70:\tlearn: 0.3571674\ttotal: 48.6ms\tremaining: 294ms\n",
      "71:\tlearn: 0.3558839\ttotal: 49.3ms\tremaining: 293ms\n",
      "72:\tlearn: 0.3544413\ttotal: 50.1ms\tremaining: 293ms\n",
      "73:\tlearn: 0.3535193\ttotal: 50.8ms\tremaining: 293ms\n",
      "74:\tlearn: 0.3525217\ttotal: 51.5ms\tremaining: 292ms\n",
      "75:\tlearn: 0.3514472\ttotal: 52ms\tremaining: 290ms\n",
      "76:\tlearn: 0.3507025\ttotal: 52.6ms\tremaining: 289ms\n",
      "77:\tlearn: 0.3503557\ttotal: 53.2ms\tremaining: 288ms\n",
      "78:\tlearn: 0.3497637\ttotal: 53.9ms\tremaining: 287ms\n",
      "79:\tlearn: 0.3490214\ttotal: 54.5ms\tremaining: 286ms\n",
      "80:\tlearn: 0.3482195\ttotal: 55.2ms\tremaining: 285ms\n",
      "81:\tlearn: 0.3474710\ttotal: 55.8ms\tremaining: 285ms\n",
      "82:\tlearn: 0.3469702\ttotal: 56.6ms\tremaining: 284ms\n",
      "83:\tlearn: 0.3466670\ttotal: 57ms\tremaining: 283ms\n",
      "84:\tlearn: 0.3457288\ttotal: 57.8ms\tremaining: 282ms\n",
      "85:\tlearn: 0.3451455\ttotal: 58.6ms\tremaining: 282ms\n",
      "86:\tlearn: 0.3441529\ttotal: 59.4ms\tremaining: 282ms\n",
      "87:\tlearn: 0.3437151\ttotal: 60.1ms\tremaining: 281ms\n",
      "88:\tlearn: 0.3426732\ttotal: 60.8ms\tremaining: 281ms\n",
      "89:\tlearn: 0.3419023\ttotal: 61.4ms\tremaining: 280ms\n",
      "90:\tlearn: 0.3410220\ttotal: 62.1ms\tremaining: 279ms\n",
      "91:\tlearn: 0.3407118\ttotal: 62.6ms\tremaining: 278ms\n",
      "92:\tlearn: 0.3406309\ttotal: 63.2ms\tremaining: 276ms\n",
      "93:\tlearn: 0.3401525\ttotal: 63.8ms\tremaining: 275ms\n",
      "94:\tlearn: 0.3397137\ttotal: 64.2ms\tremaining: 274ms\n",
      "95:\tlearn: 0.3392672\ttotal: 64.9ms\tremaining: 273ms\n",
      "96:\tlearn: 0.3387997\ttotal: 65.4ms\tremaining: 272ms\n",
      "97:\tlearn: 0.3384334\ttotal: 66.1ms\tremaining: 271ms\n",
      "98:\tlearn: 0.3378721\ttotal: 66.6ms\tremaining: 270ms\n",
      "99:\tlearn: 0.3375209\ttotal: 67.2ms\tremaining: 269ms\n",
      "100:\tlearn: 0.3370511\ttotal: 67.8ms\tremaining: 268ms\n",
      "101:\tlearn: 0.3367505\ttotal: 68.5ms\tremaining: 267ms\n",
      "102:\tlearn: 0.3361491\ttotal: 69.1ms\tremaining: 266ms\n",
      "103:\tlearn: 0.3356464\ttotal: 69.7ms\tremaining: 265ms\n",
      "104:\tlearn: 0.3355091\ttotal: 70.3ms\tremaining: 265ms\n",
      "105:\tlearn: 0.3352114\ttotal: 71.1ms\tremaining: 264ms\n",
      "106:\tlearn: 0.3349649\ttotal: 71.6ms\tremaining: 263ms\n",
      "107:\tlearn: 0.3344710\ttotal: 72.3ms\tremaining: 262ms\n",
      "108:\tlearn: 0.3341077\ttotal: 72.9ms\tremaining: 262ms\n",
      "109:\tlearn: 0.3337878\ttotal: 73.9ms\tremaining: 262ms\n",
      "110:\tlearn: 0.3335701\ttotal: 74.4ms\tremaining: 261ms\n",
      "111:\tlearn: 0.3327369\ttotal: 75.1ms\tremaining: 260ms\n",
      "112:\tlearn: 0.3322775\ttotal: 75.9ms\tremaining: 260ms\n",
      "113:\tlearn: 0.3321076\ttotal: 76.5ms\tremaining: 259ms\n",
      "114:\tlearn: 0.3320104\ttotal: 76.9ms\tremaining: 258ms\n",
      "115:\tlearn: 0.3314145\ttotal: 77.8ms\tremaining: 257ms\n",
      "116:\tlearn: 0.3308674\ttotal: 78.5ms\tremaining: 257ms\n",
      "117:\tlearn: 0.3302893\ttotal: 79.2ms\tremaining: 256ms\n",
      "118:\tlearn: 0.3300085\ttotal: 79.9ms\tremaining: 256ms\n",
      "119:\tlearn: 0.3293644\ttotal: 80.6ms\tremaining: 255ms\n",
      "120:\tlearn: 0.3289917\ttotal: 81.2ms\tremaining: 254ms\n",
      "121:\tlearn: 0.3285834\ttotal: 81.8ms\tremaining: 253ms\n",
      "122:\tlearn: 0.3281035\ttotal: 82.6ms\tremaining: 253ms\n",
      "123:\tlearn: 0.3277623\ttotal: 83.3ms\tremaining: 253ms\n",
      "124:\tlearn: 0.3269761\ttotal: 84ms\tremaining: 252ms\n",
      "125:\tlearn: 0.3266234\ttotal: 84.5ms\tremaining: 251ms\n",
      "126:\tlearn: 0.3263203\ttotal: 85ms\tremaining: 250ms\n",
      "127:\tlearn: 0.3260792\ttotal: 85.9ms\tremaining: 250ms\n",
      "128:\tlearn: 0.3255958\ttotal: 86.5ms\tremaining: 249ms\n",
      "129:\tlearn: 0.3252415\ttotal: 87.1ms\tremaining: 248ms\n",
      "130:\tlearn: 0.3249424\ttotal: 87.8ms\tremaining: 247ms\n",
      "131:\tlearn: 0.3243922\ttotal: 88.5ms\tremaining: 247ms\n",
      "132:\tlearn: 0.3240092\ttotal: 89.3ms\tremaining: 246ms\n",
      "133:\tlearn: 0.3235333\ttotal: 89.9ms\tremaining: 246ms\n",
      "134:\tlearn: 0.3231176\ttotal: 90.5ms\tremaining: 245ms\n",
      "135:\tlearn: 0.3230159\ttotal: 91.2ms\tremaining: 244ms\n",
      "136:\tlearn: 0.3226909\ttotal: 91.9ms\tremaining: 243ms\n",
      "137:\tlearn: 0.3223148\ttotal: 92.8ms\tremaining: 243ms\n",
      "138:\tlearn: 0.3218236\ttotal: 93.5ms\tremaining: 243ms\n",
      "139:\tlearn: 0.3214792\ttotal: 94.1ms\tremaining: 242ms\n",
      "140:\tlearn: 0.3212187\ttotal: 94.9ms\tremaining: 242ms\n",
      "141:\tlearn: 0.3208079\ttotal: 95.5ms\tremaining: 241ms\n",
      "142:\tlearn: 0.3206904\ttotal: 96.2ms\tremaining: 240ms\n",
      "143:\tlearn: 0.3206508\ttotal: 96.5ms\tremaining: 239ms\n",
      "144:\tlearn: 0.3204170\ttotal: 97.2ms\tremaining: 238ms\n",
      "145:\tlearn: 0.3201722\ttotal: 97.8ms\tremaining: 237ms\n",
      "146:\tlearn: 0.3197336\ttotal: 98.4ms\tremaining: 236ms\n",
      "147:\tlearn: 0.3194170\ttotal: 99ms\tremaining: 235ms\n",
      "148:\tlearn: 0.3191067\ttotal: 99.6ms\tremaining: 235ms\n",
      "149:\tlearn: 0.3186848\ttotal: 100ms\tremaining: 234ms\n",
      "150:\tlearn: 0.3185051\ttotal: 101ms\tremaining: 233ms\n",
      "151:\tlearn: 0.3182813\ttotal: 102ms\tremaining: 233ms\n",
      "152:\tlearn: 0.3182491\ttotal: 102ms\tremaining: 231ms\n",
      "153:\tlearn: 0.3180672\ttotal: 103ms\tremaining: 231ms\n",
      "154:\tlearn: 0.3180087\ttotal: 103ms\tremaining: 230ms\n",
      "155:\tlearn: 0.3177097\ttotal: 104ms\tremaining: 229ms\n",
      "156:\tlearn: 0.3176056\ttotal: 104ms\tremaining: 228ms\n",
      "157:\tlearn: 0.3173745\ttotal: 105ms\tremaining: 228ms\n",
      "158:\tlearn: 0.3169623\ttotal: 106ms\tremaining: 227ms\n",
      "159:\tlearn: 0.3163541\ttotal: 107ms\tremaining: 227ms\n",
      "160:\tlearn: 0.3159418\ttotal: 107ms\tremaining: 226ms\n",
      "161:\tlearn: 0.3156892\ttotal: 108ms\tremaining: 225ms\n",
      "162:\tlearn: 0.3153701\ttotal: 108ms\tremaining: 224ms\n",
      "163:\tlearn: 0.3149940\ttotal: 109ms\tremaining: 224ms\n",
      "164:\tlearn: 0.3148494\ttotal: 110ms\tremaining: 223ms\n",
      "165:\tlearn: 0.3146528\ttotal: 110ms\tremaining: 222ms\n",
      "166:\tlearn: 0.3143604\ttotal: 111ms\tremaining: 222ms\n",
      "167:\tlearn: 0.3140795\ttotal: 112ms\tremaining: 221ms\n",
      "168:\tlearn: 0.3139324\ttotal: 112ms\tremaining: 220ms\n",
      "169:\tlearn: 0.3135749\ttotal: 113ms\tremaining: 220ms\n",
      "170:\tlearn: 0.3133811\ttotal: 114ms\tremaining: 219ms\n",
      "171:\tlearn: 0.3133198\ttotal: 115ms\tremaining: 218ms\n",
      "172:\tlearn: 0.3129278\ttotal: 115ms\tremaining: 218ms\n",
      "173:\tlearn: 0.3128381\ttotal: 116ms\tremaining: 217ms\n",
      "174:\tlearn: 0.3126017\ttotal: 116ms\tremaining: 216ms\n",
      "175:\tlearn: 0.3123650\ttotal: 117ms\tremaining: 215ms\n",
      "176:\tlearn: 0.3120825\ttotal: 118ms\tremaining: 215ms\n",
      "177:\tlearn: 0.3116047\ttotal: 118ms\tremaining: 214ms\n",
      "178:\tlearn: 0.3113092\ttotal: 119ms\tremaining: 213ms\n",
      "179:\tlearn: 0.3111511\ttotal: 120ms\tremaining: 213ms\n",
      "180:\tlearn: 0.3106952\ttotal: 120ms\tremaining: 212ms\n",
      "181:\tlearn: 0.3104823\ttotal: 121ms\tremaining: 212ms\n",
      "182:\tlearn: 0.3101818\ttotal: 122ms\tremaining: 211ms\n",
      "183:\tlearn: 0.3099848\ttotal: 122ms\tremaining: 210ms\n",
      "184:\tlearn: 0.3096503\ttotal: 123ms\tremaining: 209ms\n",
      "185:\tlearn: 0.3095030\ttotal: 124ms\tremaining: 209ms\n",
      "186:\tlearn: 0.3093022\ttotal: 124ms\tremaining: 208ms\n",
      "187:\tlearn: 0.3090152\ttotal: 125ms\tremaining: 207ms\n",
      "188:\tlearn: 0.3088391\ttotal: 126ms\tremaining: 207ms\n",
      "189:\tlearn: 0.3087640\ttotal: 126ms\tremaining: 206ms\n",
      "190:\tlearn: 0.3084865\ttotal: 127ms\tremaining: 205ms\n",
      "191:\tlearn: 0.3081560\ttotal: 127ms\tremaining: 204ms\n",
      "192:\tlearn: 0.3078451\ttotal: 128ms\tremaining: 204ms\n",
      "193:\tlearn: 0.3074931\ttotal: 129ms\tremaining: 203ms\n",
      "194:\tlearn: 0.3068737\ttotal: 129ms\tremaining: 202ms\n",
      "195:\tlearn: 0.3066557\ttotal: 130ms\tremaining: 202ms\n",
      "196:\tlearn: 0.3066157\ttotal: 130ms\tremaining: 200ms\n",
      "197:\tlearn: 0.3064797\ttotal: 131ms\tremaining: 199ms\n",
      "198:\tlearn: 0.3062289\ttotal: 132ms\tremaining: 199ms\n",
      "199:\tlearn: 0.3060957\ttotal: 132ms\tremaining: 198ms\n",
      "200:\tlearn: 0.3057717\ttotal: 133ms\tremaining: 197ms\n",
      "201:\tlearn: 0.3056512\ttotal: 133ms\tremaining: 197ms\n",
      "202:\tlearn: 0.3055394\ttotal: 134ms\tremaining: 196ms\n",
      "203:\tlearn: 0.3053639\ttotal: 135ms\tremaining: 195ms\n",
      "204:\tlearn: 0.3051282\ttotal: 135ms\tremaining: 195ms\n",
      "205:\tlearn: 0.3046341\ttotal: 136ms\tremaining: 194ms\n",
      "206:\tlearn: 0.3044638\ttotal: 137ms\tremaining: 193ms\n",
      "207:\tlearn: 0.3043150\ttotal: 137ms\tremaining: 193ms\n",
      "208:\tlearn: 0.3041440\ttotal: 138ms\tremaining: 192ms\n",
      "209:\tlearn: 0.3039063\ttotal: 138ms\tremaining: 191ms\n",
      "210:\tlearn: 0.3036335\ttotal: 139ms\tremaining: 191ms\n",
      "211:\tlearn: 0.3035308\ttotal: 140ms\tremaining: 190ms\n",
      "212:\tlearn: 0.3033867\ttotal: 141ms\tremaining: 190ms\n",
      "213:\tlearn: 0.3030912\ttotal: 141ms\tremaining: 189ms\n",
      "214:\tlearn: 0.3026557\ttotal: 142ms\tremaining: 188ms\n",
      "215:\tlearn: 0.3023316\ttotal: 143ms\tremaining: 188ms\n",
      "216:\tlearn: 0.3019771\ttotal: 143ms\tremaining: 187ms\n",
      "217:\tlearn: 0.3015183\ttotal: 144ms\tremaining: 186ms\n",
      "218:\tlearn: 0.3012545\ttotal: 145ms\tremaining: 186ms\n",
      "219:\tlearn: 0.3009470\ttotal: 146ms\tremaining: 185ms\n",
      "220:\tlearn: 0.3005900\ttotal: 146ms\tremaining: 184ms\n",
      "221:\tlearn: 0.3005617\ttotal: 147ms\tremaining: 183ms\n",
      "222:\tlearn: 0.3004479\ttotal: 147ms\tremaining: 183ms\n",
      "223:\tlearn: 0.3003757\ttotal: 148ms\tremaining: 182ms\n",
      "224:\tlearn: 0.3000904\ttotal: 148ms\tremaining: 181ms\n",
      "225:\tlearn: 0.2999829\ttotal: 149ms\tremaining: 181ms\n",
      "226:\tlearn: 0.2998724\ttotal: 150ms\tremaining: 180ms\n",
      "227:\tlearn: 0.2997238\ttotal: 150ms\tremaining: 179ms\n",
      "228:\tlearn: 0.2995134\ttotal: 151ms\tremaining: 179ms\n",
      "229:\tlearn: 0.2993131\ttotal: 152ms\tremaining: 178ms\n",
      "230:\tlearn: 0.2990143\ttotal: 152ms\tremaining: 177ms\n",
      "231:\tlearn: 0.2988248\ttotal: 153ms\tremaining: 177ms\n",
      "232:\tlearn: 0.2987920\ttotal: 153ms\tremaining: 176ms\n",
      "233:\tlearn: 0.2984714\ttotal: 154ms\tremaining: 175ms\n",
      "234:\tlearn: 0.2981296\ttotal: 155ms\tremaining: 174ms\n",
      "235:\tlearn: 0.2979819\ttotal: 155ms\tremaining: 174ms\n",
      "236:\tlearn: 0.2979802\ttotal: 156ms\tremaining: 173ms\n",
      "237:\tlearn: 0.2977127\ttotal: 156ms\tremaining: 172ms\n",
      "238:\tlearn: 0.2976244\ttotal: 157ms\tremaining: 171ms\n",
      "239:\tlearn: 0.2974888\ttotal: 157ms\tremaining: 171ms\n",
      "240:\tlearn: 0.2972298\ttotal: 158ms\tremaining: 170ms\n",
      "241:\tlearn: 0.2969534\ttotal: 159ms\tremaining: 169ms\n",
      "242:\tlearn: 0.2969207\ttotal: 159ms\tremaining: 168ms\n",
      "243:\tlearn: 0.2968214\ttotal: 160ms\tremaining: 168ms\n",
      "244:\tlearn: 0.2966215\ttotal: 160ms\tremaining: 167ms\n",
      "245:\tlearn: 0.2964191\ttotal: 161ms\tremaining: 166ms\n",
      "246:\tlearn: 0.2963642\ttotal: 161ms\tremaining: 165ms\n",
      "247:\tlearn: 0.2962057\ttotal: 162ms\tremaining: 165ms\n",
      "248:\tlearn: 0.2960640\ttotal: 163ms\tremaining: 164ms\n",
      "249:\tlearn: 0.2959562\ttotal: 163ms\tremaining: 163ms\n",
      "250:\tlearn: 0.2957673\ttotal: 164ms\tremaining: 163ms\n",
      "251:\tlearn: 0.2956080\ttotal: 165ms\tremaining: 162ms\n",
      "252:\tlearn: 0.2954178\ttotal: 166ms\tremaining: 162ms\n",
      "253:\tlearn: 0.2950691\ttotal: 166ms\tremaining: 161ms\n",
      "254:\tlearn: 0.2949707\ttotal: 167ms\tremaining: 160ms\n",
      "255:\tlearn: 0.2947476\ttotal: 168ms\tremaining: 160ms\n",
      "256:\tlearn: 0.2945682\ttotal: 168ms\tremaining: 159ms\n",
      "257:\tlearn: 0.2944772\ttotal: 169ms\tremaining: 159ms\n",
      "258:\tlearn: 0.2941961\ttotal: 170ms\tremaining: 158ms\n",
      "259:\tlearn: 0.2939593\ttotal: 170ms\tremaining: 157ms\n",
      "260:\tlearn: 0.2938386\ttotal: 171ms\tremaining: 156ms\n",
      "261:\tlearn: 0.2937656\ttotal: 171ms\tremaining: 156ms\n",
      "262:\tlearn: 0.2937252\ttotal: 172ms\tremaining: 155ms\n",
      "263:\tlearn: 0.2936641\ttotal: 173ms\tremaining: 155ms\n",
      "264:\tlearn: 0.2934684\ttotal: 173ms\tremaining: 154ms\n",
      "265:\tlearn: 0.2933116\ttotal: 174ms\tremaining: 153ms\n",
      "266:\tlearn: 0.2930051\ttotal: 175ms\tremaining: 153ms\n",
      "267:\tlearn: 0.2928714\ttotal: 175ms\tremaining: 152ms\n",
      "268:\tlearn: 0.2925892\ttotal: 176ms\tremaining: 151ms\n",
      "269:\tlearn: 0.2925587\ttotal: 176ms\tremaining: 150ms\n",
      "270:\tlearn: 0.2923396\ttotal: 177ms\tremaining: 150ms\n",
      "271:\tlearn: 0.2921157\ttotal: 178ms\tremaining: 149ms\n",
      "272:\tlearn: 0.2917801\ttotal: 178ms\tremaining: 148ms\n",
      "273:\tlearn: 0.2916242\ttotal: 179ms\tremaining: 147ms\n",
      "274:\tlearn: 0.2915423\ttotal: 180ms\tremaining: 147ms\n",
      "275:\tlearn: 0.2914579\ttotal: 180ms\tremaining: 146ms\n",
      "276:\tlearn: 0.2913473\ttotal: 181ms\tremaining: 146ms\n",
      "277:\tlearn: 0.2912041\ttotal: 182ms\tremaining: 145ms\n",
      "278:\tlearn: 0.2910728\ttotal: 182ms\tremaining: 144ms\n",
      "279:\tlearn: 0.2908103\ttotal: 183ms\tremaining: 144ms\n",
      "280:\tlearn: 0.2907059\ttotal: 184ms\tremaining: 143ms\n",
      "281:\tlearn: 0.2906260\ttotal: 184ms\tremaining: 143ms\n",
      "282:\tlearn: 0.2904438\ttotal: 185ms\tremaining: 142ms\n",
      "283:\tlearn: 0.2901482\ttotal: 186ms\tremaining: 141ms\n",
      "284:\tlearn: 0.2900151\ttotal: 186ms\tremaining: 141ms\n",
      "285:\tlearn: 0.2898926\ttotal: 187ms\tremaining: 140ms\n",
      "286:\tlearn: 0.2898604\ttotal: 187ms\tremaining: 139ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "287:\tlearn: 0.2896901\ttotal: 188ms\tremaining: 139ms\n",
      "288:\tlearn: 0.2895967\ttotal: 189ms\tremaining: 138ms\n",
      "289:\tlearn: 0.2895286\ttotal: 190ms\tremaining: 137ms\n",
      "290:\tlearn: 0.2890514\ttotal: 190ms\tremaining: 137ms\n",
      "291:\tlearn: 0.2889790\ttotal: 191ms\tremaining: 136ms\n",
      "292:\tlearn: 0.2888725\ttotal: 192ms\tremaining: 135ms\n",
      "293:\tlearn: 0.2886043\ttotal: 192ms\tremaining: 135ms\n",
      "294:\tlearn: 0.2882934\ttotal: 193ms\tremaining: 134ms\n",
      "295:\tlearn: 0.2881435\ttotal: 193ms\tremaining: 133ms\n",
      "296:\tlearn: 0.2879678\ttotal: 194ms\tremaining: 133ms\n",
      "297:\tlearn: 0.2878881\ttotal: 195ms\tremaining: 132ms\n",
      "298:\tlearn: 0.2877009\ttotal: 195ms\tremaining: 131ms\n",
      "299:\tlearn: 0.2876895\ttotal: 196ms\tremaining: 130ms\n",
      "300:\tlearn: 0.2873960\ttotal: 196ms\tremaining: 130ms\n",
      "301:\tlearn: 0.2872253\ttotal: 197ms\tremaining: 129ms\n",
      "302:\tlearn: 0.2869601\ttotal: 198ms\tremaining: 129ms\n",
      "303:\tlearn: 0.2867638\ttotal: 199ms\tremaining: 128ms\n",
      "304:\tlearn: 0.2866266\ttotal: 199ms\tremaining: 127ms\n",
      "305:\tlearn: 0.2864389\ttotal: 200ms\tremaining: 127ms\n",
      "306:\tlearn: 0.2862793\ttotal: 201ms\tremaining: 126ms\n",
      "307:\tlearn: 0.2859026\ttotal: 201ms\tremaining: 126ms\n",
      "308:\tlearn: 0.2855366\ttotal: 202ms\tremaining: 125ms\n",
      "309:\tlearn: 0.2853026\ttotal: 203ms\tremaining: 124ms\n",
      "310:\tlearn: 0.2850987\ttotal: 204ms\tremaining: 124ms\n",
      "311:\tlearn: 0.2848136\ttotal: 204ms\tremaining: 123ms\n",
      "312:\tlearn: 0.2845751\ttotal: 205ms\tremaining: 122ms\n",
      "313:\tlearn: 0.2845173\ttotal: 206ms\tremaining: 122ms\n",
      "314:\tlearn: 0.2843984\ttotal: 206ms\tremaining: 121ms\n",
      "315:\tlearn: 0.2843622\ttotal: 207ms\tremaining: 121ms\n",
      "316:\tlearn: 0.2842249\ttotal: 208ms\tremaining: 120ms\n",
      "317:\tlearn: 0.2841515\ttotal: 209ms\tremaining: 119ms\n",
      "318:\tlearn: 0.2838266\ttotal: 209ms\tremaining: 119ms\n",
      "319:\tlearn: 0.2837659\ttotal: 210ms\tremaining: 118ms\n",
      "320:\tlearn: 0.2835485\ttotal: 211ms\tremaining: 117ms\n",
      "321:\tlearn: 0.2835231\ttotal: 211ms\tremaining: 117ms\n",
      "322:\tlearn: 0.2832955\ttotal: 212ms\tremaining: 116ms\n",
      "323:\tlearn: 0.2829555\ttotal: 213ms\tremaining: 116ms\n",
      "324:\tlearn: 0.2826467\ttotal: 213ms\tremaining: 115ms\n",
      "325:\tlearn: 0.2824853\ttotal: 214ms\tremaining: 114ms\n",
      "326:\tlearn: 0.2824439\ttotal: 215ms\tremaining: 114ms\n",
      "327:\tlearn: 0.2822381\ttotal: 215ms\tremaining: 113ms\n",
      "328:\tlearn: 0.2820211\ttotal: 216ms\tremaining: 112ms\n",
      "329:\tlearn: 0.2816249\ttotal: 217ms\tremaining: 112ms\n",
      "330:\tlearn: 0.2814703\ttotal: 218ms\tremaining: 111ms\n",
      "331:\tlearn: 0.2814498\ttotal: 218ms\tremaining: 110ms\n",
      "332:\tlearn: 0.2814121\ttotal: 219ms\tremaining: 110ms\n",
      "333:\tlearn: 0.2811507\ttotal: 219ms\tremaining: 109ms\n",
      "334:\tlearn: 0.2810744\ttotal: 220ms\tremaining: 108ms\n",
      "335:\tlearn: 0.2807456\ttotal: 221ms\tremaining: 108ms\n",
      "336:\tlearn: 0.2805499\ttotal: 221ms\tremaining: 107ms\n",
      "337:\tlearn: 0.2803793\ttotal: 222ms\tremaining: 106ms\n",
      "338:\tlearn: 0.2802364\ttotal: 223ms\tremaining: 106ms\n",
      "339:\tlearn: 0.2800871\ttotal: 224ms\tremaining: 105ms\n",
      "340:\tlearn: 0.2800668\ttotal: 224ms\tremaining: 105ms\n",
      "341:\tlearn: 0.2797370\ttotal: 225ms\tremaining: 104ms\n",
      "342:\tlearn: 0.2795218\ttotal: 225ms\tremaining: 103ms\n",
      "343:\tlearn: 0.2793370\ttotal: 226ms\tremaining: 103ms\n",
      "344:\tlearn: 0.2792685\ttotal: 227ms\tremaining: 102ms\n",
      "345:\tlearn: 0.2791769\ttotal: 228ms\tremaining: 101ms\n",
      "346:\tlearn: 0.2789486\ttotal: 228ms\tremaining: 101ms\n",
      "347:\tlearn: 0.2787908\ttotal: 229ms\tremaining: 100ms\n",
      "348:\tlearn: 0.2787111\ttotal: 230ms\tremaining: 99.4ms\n",
      "349:\tlearn: 0.2786064\ttotal: 231ms\tremaining: 98.8ms\n",
      "350:\tlearn: 0.2783995\ttotal: 231ms\tremaining: 98.1ms\n",
      "351:\tlearn: 0.2782961\ttotal: 232ms\tremaining: 97.5ms\n",
      "352:\tlearn: 0.2779172\ttotal: 233ms\tremaining: 96.8ms\n",
      "353:\tlearn: 0.2778510\ttotal: 233ms\tremaining: 96.2ms\n",
      "354:\tlearn: 0.2774378\ttotal: 234ms\tremaining: 95.5ms\n",
      "355:\tlearn: 0.2772798\ttotal: 235ms\tremaining: 94.9ms\n",
      "356:\tlearn: 0.2769573\ttotal: 235ms\tremaining: 94.2ms\n",
      "357:\tlearn: 0.2767985\ttotal: 236ms\tremaining: 93.6ms\n",
      "358:\tlearn: 0.2767358\ttotal: 237ms\tremaining: 93ms\n",
      "359:\tlearn: 0.2767180\ttotal: 237ms\tremaining: 92.2ms\n",
      "360:\tlearn: 0.2766371\ttotal: 238ms\tremaining: 91.5ms\n",
      "361:\tlearn: 0.2763343\ttotal: 238ms\tremaining: 90.8ms\n",
      "362:\tlearn: 0.2762215\ttotal: 239ms\tremaining: 90.2ms\n",
      "363:\tlearn: 0.2761174\ttotal: 240ms\tremaining: 89.6ms\n",
      "364:\tlearn: 0.2760435\ttotal: 240ms\tremaining: 88.9ms\n",
      "365:\tlearn: 0.2758539\ttotal: 241ms\tremaining: 88.3ms\n",
      "366:\tlearn: 0.2757661\ttotal: 242ms\tremaining: 87.6ms\n",
      "367:\tlearn: 0.2755547\ttotal: 243ms\tremaining: 87ms\n",
      "368:\tlearn: 0.2752305\ttotal: 243ms\tremaining: 86.3ms\n",
      "369:\tlearn: 0.2750523\ttotal: 244ms\tremaining: 85.6ms\n",
      "370:\tlearn: 0.2749120\ttotal: 244ms\tremaining: 84.9ms\n",
      "371:\tlearn: 0.2746812\ttotal: 246ms\tremaining: 84.5ms\n",
      "372:\tlearn: 0.2745199\ttotal: 246ms\tremaining: 83.8ms\n",
      "373:\tlearn: 0.2743848\ttotal: 247ms\tremaining: 83.2ms\n",
      "374:\tlearn: 0.2739751\ttotal: 248ms\tremaining: 82.6ms\n",
      "375:\tlearn: 0.2736813\ttotal: 248ms\tremaining: 81.9ms\n",
      "376:\tlearn: 0.2735909\ttotal: 249ms\tremaining: 81.2ms\n",
      "377:\tlearn: 0.2734079\ttotal: 250ms\tremaining: 80.6ms\n",
      "378:\tlearn: 0.2732785\ttotal: 250ms\tremaining: 79.9ms\n",
      "379:\tlearn: 0.2731625\ttotal: 251ms\tremaining: 79.2ms\n",
      "380:\tlearn: 0.2729586\ttotal: 251ms\tremaining: 78.5ms\n",
      "381:\tlearn: 0.2728858\ttotal: 252ms\tremaining: 77.9ms\n",
      "382:\tlearn: 0.2726898\ttotal: 253ms\tremaining: 77.2ms\n",
      "383:\tlearn: 0.2725564\ttotal: 253ms\tremaining: 76.5ms\n",
      "384:\tlearn: 0.2723761\ttotal: 254ms\tremaining: 75.9ms\n",
      "385:\tlearn: 0.2722521\ttotal: 255ms\tremaining: 75.3ms\n",
      "386:\tlearn: 0.2721664\ttotal: 256ms\tremaining: 74.6ms\n",
      "387:\tlearn: 0.2719699\ttotal: 256ms\tremaining: 74ms\n",
      "388:\tlearn: 0.2716972\ttotal: 257ms\tremaining: 73.3ms\n",
      "389:\tlearn: 0.2715577\ttotal: 258ms\tremaining: 72.7ms\n",
      "390:\tlearn: 0.2711651\ttotal: 258ms\tremaining: 72ms\n",
      "391:\tlearn: 0.2710642\ttotal: 259ms\tremaining: 71.4ms\n",
      "392:\tlearn: 0.2708973\ttotal: 260ms\tremaining: 70.7ms\n",
      "393:\tlearn: 0.2706381\ttotal: 260ms\tremaining: 70.1ms\n",
      "394:\tlearn: 0.2705597\ttotal: 261ms\tremaining: 69.4ms\n",
      "395:\tlearn: 0.2703559\ttotal: 262ms\tremaining: 68.7ms\n",
      "396:\tlearn: 0.2702133\ttotal: 262ms\tremaining: 68.1ms\n",
      "397:\tlearn: 0.2701020\ttotal: 263ms\tremaining: 67.4ms\n",
      "398:\tlearn: 0.2700648\ttotal: 264ms\tremaining: 66.8ms\n",
      "399:\tlearn: 0.2699635\ttotal: 264ms\tremaining: 66.1ms\n",
      "400:\tlearn: 0.2696352\ttotal: 265ms\tremaining: 65.4ms\n",
      "401:\tlearn: 0.2695423\ttotal: 266ms\tremaining: 64.8ms\n",
      "402:\tlearn: 0.2694365\ttotal: 266ms\tremaining: 64.1ms\n",
      "403:\tlearn: 0.2693132\ttotal: 267ms\tremaining: 63.5ms\n",
      "404:\tlearn: 0.2692017\ttotal: 268ms\tremaining: 62.8ms\n",
      "405:\tlearn: 0.2690144\ttotal: 268ms\tremaining: 62.1ms\n",
      "406:\tlearn: 0.2686358\ttotal: 269ms\tremaining: 61.5ms\n",
      "407:\tlearn: 0.2685617\ttotal: 270ms\tremaining: 60.8ms\n",
      "408:\tlearn: 0.2684243\ttotal: 270ms\tremaining: 60.2ms\n",
      "409:\tlearn: 0.2683893\ttotal: 271ms\tremaining: 59.5ms\n",
      "410:\tlearn: 0.2683277\ttotal: 272ms\tremaining: 58.8ms\n",
      "411:\tlearn: 0.2681435\ttotal: 272ms\tremaining: 58.2ms\n",
      "412:\tlearn: 0.2680211\ttotal: 273ms\tremaining: 57.5ms\n",
      "413:\tlearn: 0.2677987\ttotal: 274ms\tremaining: 56.8ms\n",
      "414:\tlearn: 0.2676963\ttotal: 274ms\tremaining: 56.2ms\n",
      "415:\tlearn: 0.2675537\ttotal: 275ms\tremaining: 55.6ms\n",
      "416:\tlearn: 0.2674742\ttotal: 276ms\tremaining: 54.9ms\n",
      "417:\tlearn: 0.2673786\ttotal: 276ms\tremaining: 54.2ms\n",
      "418:\tlearn: 0.2673000\ttotal: 277ms\tremaining: 53.6ms\n",
      "419:\tlearn: 0.2671896\ttotal: 278ms\tremaining: 52.9ms\n",
      "420:\tlearn: 0.2670208\ttotal: 278ms\tremaining: 52.3ms\n",
      "421:\tlearn: 0.2669462\ttotal: 279ms\tremaining: 51.6ms\n",
      "422:\tlearn: 0.2666903\ttotal: 280ms\tremaining: 50.9ms\n",
      "423:\tlearn: 0.2666733\ttotal: 281ms\tremaining: 50.3ms\n",
      "424:\tlearn: 0.2664480\ttotal: 281ms\tremaining: 49.6ms\n",
      "425:\tlearn: 0.2663699\ttotal: 282ms\tremaining: 49ms\n",
      "426:\tlearn: 0.2662465\ttotal: 283ms\tremaining: 48.3ms\n",
      "427:\tlearn: 0.2662190\ttotal: 283ms\tremaining: 47.6ms\n",
      "428:\tlearn: 0.2660766\ttotal: 284ms\tremaining: 47ms\n",
      "429:\tlearn: 0.2659788\ttotal: 284ms\tremaining: 46.3ms\n",
      "430:\tlearn: 0.2658204\ttotal: 285ms\tremaining: 45.7ms\n",
      "431:\tlearn: 0.2654723\ttotal: 286ms\tremaining: 45ms\n",
      "432:\tlearn: 0.2652415\ttotal: 286ms\tremaining: 44.3ms\n",
      "433:\tlearn: 0.2650050\ttotal: 287ms\tremaining: 43.6ms\n",
      "434:\tlearn: 0.2649339\ttotal: 288ms\tremaining: 43ms\n",
      "435:\tlearn: 0.2647333\ttotal: 288ms\tremaining: 42.3ms\n",
      "436:\tlearn: 0.2645435\ttotal: 289ms\tremaining: 41.7ms\n",
      "437:\tlearn: 0.2641839\ttotal: 290ms\tremaining: 41ms\n",
      "438:\tlearn: 0.2641040\ttotal: 290ms\tremaining: 40.3ms\n",
      "439:\tlearn: 0.2640396\ttotal: 291ms\tremaining: 39.7ms\n",
      "440:\tlearn: 0.2638183\ttotal: 292ms\tremaining: 39ms\n",
      "441:\tlearn: 0.2635876\ttotal: 292ms\tremaining: 38.4ms\n",
      "442:\tlearn: 0.2634898\ttotal: 293ms\tremaining: 37.8ms\n",
      "443:\tlearn: 0.2633742\ttotal: 294ms\tremaining: 37.1ms\n",
      "444:\tlearn: 0.2632768\ttotal: 295ms\tremaining: 36.4ms\n",
      "445:\tlearn: 0.2630285\ttotal: 295ms\tremaining: 35.7ms\n",
      "446:\tlearn: 0.2629419\ttotal: 296ms\tremaining: 35.1ms\n",
      "447:\tlearn: 0.2629139\ttotal: 296ms\tremaining: 34.4ms\n",
      "448:\tlearn: 0.2628464\ttotal: 297ms\tremaining: 33.7ms\n",
      "449:\tlearn: 0.2628117\ttotal: 298ms\tremaining: 33.1ms\n",
      "450:\tlearn: 0.2627877\ttotal: 298ms\tremaining: 32.4ms\n",
      "451:\tlearn: 0.2623903\ttotal: 299ms\tremaining: 31.7ms\n",
      "452:\tlearn: 0.2621604\ttotal: 299ms\tremaining: 31.1ms\n",
      "453:\tlearn: 0.2619655\ttotal: 300ms\tremaining: 30.4ms\n",
      "454:\tlearn: 0.2617162\ttotal: 301ms\tremaining: 29.7ms\n",
      "455:\tlearn: 0.2615928\ttotal: 301ms\tremaining: 29.1ms\n",
      "456:\tlearn: 0.2614487\ttotal: 302ms\tremaining: 28.4ms\n",
      "457:\tlearn: 0.2613730\ttotal: 302ms\tremaining: 27.7ms\n",
      "458:\tlearn: 0.2612938\ttotal: 303ms\tremaining: 27.1ms\n",
      "459:\tlearn: 0.2611167\ttotal: 304ms\tremaining: 26.4ms\n",
      "460:\tlearn: 0.2609989\ttotal: 304ms\tremaining: 25.8ms\n",
      "461:\tlearn: 0.2608613\ttotal: 305ms\tremaining: 25.1ms\n",
      "462:\tlearn: 0.2607640\ttotal: 306ms\tremaining: 24.4ms\n",
      "463:\tlearn: 0.2607372\ttotal: 306ms\tremaining: 23.8ms\n",
      "464:\tlearn: 0.2606182\ttotal: 307ms\tremaining: 23.1ms\n",
      "465:\tlearn: 0.2604102\ttotal: 307ms\tremaining: 22.4ms\n",
      "466:\tlearn: 0.2602902\ttotal: 308ms\tremaining: 21.8ms\n",
      "467:\tlearn: 0.2602580\ttotal: 309ms\tremaining: 21.1ms\n",
      "468:\tlearn: 0.2601854\ttotal: 310ms\tremaining: 20.5ms\n",
      "469:\tlearn: 0.2600898\ttotal: 310ms\tremaining: 19.8ms\n",
      "470:\tlearn: 0.2600454\ttotal: 311ms\tremaining: 19.1ms\n",
      "471:\tlearn: 0.2600309\ttotal: 311ms\tremaining: 18.5ms\n",
      "472:\tlearn: 0.2598827\ttotal: 312ms\tremaining: 17.8ms\n",
      "473:\tlearn: 0.2597131\ttotal: 313ms\tremaining: 17.2ms\n",
      "474:\tlearn: 0.2596493\ttotal: 314ms\tremaining: 16.5ms\n",
      "475:\tlearn: 0.2593524\ttotal: 314ms\tremaining: 15.8ms\n",
      "476:\tlearn: 0.2591360\ttotal: 315ms\tremaining: 15.2ms\n",
      "477:\tlearn: 0.2588903\ttotal: 316ms\tremaining: 14.5ms\n",
      "478:\tlearn: 0.2586963\ttotal: 316ms\tremaining: 13.9ms\n",
      "479:\tlearn: 0.2585929\ttotal: 317ms\tremaining: 13.2ms\n",
      "480:\tlearn: 0.2584573\ttotal: 318ms\tremaining: 12.5ms\n",
      "481:\tlearn: 0.2582543\ttotal: 318ms\tremaining: 11.9ms\n",
      "482:\tlearn: 0.2579687\ttotal: 319ms\tremaining: 11.2ms\n",
      "483:\tlearn: 0.2578254\ttotal: 319ms\tremaining: 10.6ms\n",
      "484:\tlearn: 0.2574777\ttotal: 320ms\tremaining: 9.9ms\n",
      "485:\tlearn: 0.2574504\ttotal: 321ms\tremaining: 9.24ms\n",
      "486:\tlearn: 0.2574067\ttotal: 321ms\tremaining: 8.58ms\n",
      "487:\tlearn: 0.2573131\ttotal: 322ms\tremaining: 7.92ms\n",
      "488:\tlearn: 0.2570615\ttotal: 323ms\tremaining: 7.26ms\n",
      "489:\tlearn: 0.2570337\ttotal: 323ms\tremaining: 6.6ms\n",
      "490:\tlearn: 0.2569267\ttotal: 324ms\tremaining: 5.94ms\n",
      "491:\tlearn: 0.2567245\ttotal: 325ms\tremaining: 5.28ms\n",
      "492:\tlearn: 0.2565968\ttotal: 325ms\tremaining: 4.62ms\n",
      "493:\tlearn: 0.2562541\ttotal: 326ms\tremaining: 3.96ms\n",
      "494:\tlearn: 0.2561525\ttotal: 327ms\tremaining: 3.3ms\n",
      "495:\tlearn: 0.2560640\ttotal: 327ms\tremaining: 2.64ms\n",
      "496:\tlearn: 0.2560054\ttotal: 328ms\tremaining: 1.98ms\n",
      "497:\tlearn: 0.2558971\ttotal: 329ms\tremaining: 1.32ms\n",
      "498:\tlearn: 0.2558274\ttotal: 329ms\tremaining: 660us\n",
      "499:\tlearn: 0.2556538\ttotal: 330ms\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x7f97ebe278b0>"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CBC_500.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "aa14d140",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9015317286652079"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(Y_train, CBC_500.predict(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "7f847f3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8903061224489796"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(Y_test, CBC_500.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "23be3acf",
   "metadata": {},
   "outputs": [],
   "source": [
    "cohen_kappa_score_CB =  cohen_kappa_score(Y_test,CBC_500.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "b17af8b2",
   "metadata": {},
   "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>Model</th>\n",
       "      <th>Train_Score</th>\n",
       "      <th>Test_Score</th>\n",
       "      <th>Cohen_kappa_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AdaBoost_hyper_tuned</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Graident_Boosting_hypertune</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CatBoost_hypertuned</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.76</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Model  Train_Score  Test_Score  Cohen_kappa_score\n",
       "0         AdaBoost_hyper_tuned         0.87        0.86               0.70\n",
       "1  Graident_Boosting_hypertune         0.89        0.88               0.74\n",
       "2          CatBoost_hypertuned         0.90        0.89               0.76"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Report_Card2 = Report_Card2.append({'Model' : 'CatBoost_hypertuned', \n",
    "                    'Train_Score' : round(accuracy_score(Y_train, CBC_500.predict(X_train)),2),\n",
    "                    'Test_Score': round(accuracy_score(Y_test, CBC_500.predict(X_test)),2),\n",
    "                    'Cohen_kappa_score' : round(cohen_kappa_score_CB,2)\n",
    "                    }, ignore_index=True)\n",
    "Report_Card2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "ba497d0e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.94      0.92       565\n",
      "           1       0.90      0.83      0.87       349\n",
      "\n",
      "    accuracy                           0.90       914\n",
      "   macro avg       0.90      0.89      0.89       914\n",
      "weighted avg       0.90      0.90      0.90       914\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(Y_train, CBC_500.predict(X_train)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "db0f2b20",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.93      0.92       249\n",
      "           1       0.87      0.82      0.84       143\n",
      "\n",
      "    accuracy                           0.89       392\n",
      "   macro avg       0.89      0.87      0.88       392\n",
      "weighted avg       0.89      0.89      0.89       392\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(Y_test, CBC_500.predict(X_test)))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}