心跳信号分类预测_tsfresh_XGB_20220327.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CatBoost训练10000次\n",
    "#2791.6345\n",
    "#CatBoost训练1000次\n",
    "#16044.4139\n",
    "#CatBoost训练30000次\n",
    "#2326.2931\n",
    "#tsfresh, XGBoost，n_estimators=2000\n",
    "#547.0449\n",
    "#tsfresh, XGBoost, n_estimators=2000，最大值转为1，其他为0\n",
    "#376.0000\n",
    "#tsfresh, XGBoost, n_estimators=2500，最大值转为1，其他为0\n",
    "#364.0000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('/home/aistudio/external-libraries/catboost103')\n",
    "sys.path.append('/home/aistudio/external-libraries/ipywidgets')\n",
    "sys.path.append('/home/aistudio/external-libraries/bayesian-optimization')\n",
    "sys.path.append('/home/aistudio/external-libraries/toad')\n",
    "sys.path.append('/home/aistudio/external-libraries/pandas-profiling')\n",
    "sys.path.append('/home/aistudio/external-libraries/tsfresh')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pickle\n",
    "%matplotlib inline\n",
    "import matplotlib # 注意这个也要import一次\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import roc_auc_score, accuracy_score, recall_score, mean_squared_error, roc_curve, auc\n",
    "from sklearn.model_selection import train_test_split\n",
    "import gc\n",
    "\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = pd.read_csv('./data/data133887/train.csv')\n",
    "test_data = pd.read_csv('./data/data133887/testA.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.concat((train_data, test_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'heartbeat_signals', 'label'], dtype='object')"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.DataFrame(data['heartbeat_signals'].str.split(',').tolist(), index=data['id'], dtype='float')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['label'] = train_data['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data.iloc[:, 0:205] = data.iloc[:, 0:205].astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f916ff4c190>]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data.iloc[10, 0:205])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "##保存数据data\n",
    "# with open('./work/features/data.pkl', 'wb') as file:\n",
    "#     pickle.dump(data, file) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 手动提取特征"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_des = data.drop(['isDefault'], axis=1).T.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_des.drop(['count'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "##保存数据data_des\n",
    "# with open('./work/features/data_des.pkl', 'wb') as file:\n",
    "#     pickle.dump(data_des, file) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 提取数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# with open('./work/features/data.pkl', 'rb') as file:\n",
    "#     data = pickle.load(file)\n",
    "\n",
    "# with open('./work/features/data_des.pkl', 'rb') as file:\n",
    "#     data_des = pickle.load(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data = pd.merge(data, data_des, left_index=True, right_index=True)\n",
    "# del data_des"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>196</th>\n",
       "      <th>197</th>\n",
       "      <th>198</th>\n",
       "      <th>199</th>\n",
       "      <th>200</th>\n",
       "      <th>201</th>\n",
       "      <th>202</th>\n",
       "      <th>203</th>\n",
       "      <th>204</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.991230</td>\n",
       "      <td>0.943533</td>\n",
       "      <td>0.764677</td>\n",
       "      <td>0.618571</td>\n",
       "      <td>0.379632</td>\n",
       "      <td>0.190822</td>\n",
       "      <td>0.040237</td>\n",
       "      <td>0.025995</td>\n",
       "      <td>0.031709</td>\n",
       "      <td>0.065524</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.971482</td>\n",
       "      <td>0.928969</td>\n",
       "      <td>0.572933</td>\n",
       "      <td>0.178457</td>\n",
       "      <td>0.122962</td>\n",
       "      <td>0.132360</td>\n",
       "      <td>0.094392</td>\n",
       "      <td>0.089575</td>\n",
       "      <td>0.030481</td>\n",
       "      <td>0.040499</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.959149</td>\n",
       "      <td>0.701378</td>\n",
       "      <td>0.231778</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.080698</td>\n",
       "      <td>0.128376</td>\n",
       "      <td>0.187448</td>\n",
       "      <td>0.280826</td>\n",
       "      <td>0.328261</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.975795</td>\n",
       "      <td>0.934088</td>\n",
       "      <td>0.659637</td>\n",
       "      <td>0.249921</td>\n",
       "      <td>0.237116</td>\n",
       "      <td>0.281445</td>\n",
       "      <td>0.249921</td>\n",
       "      <td>0.249921</td>\n",
       "      <td>0.241397</td>\n",
       "      <td>0.230670</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.055816</td>\n",
       "      <td>0.261294</td>\n",
       "      <td>0.359847</td>\n",
       "      <td>0.433143</td>\n",
       "      <td>0.453698</td>\n",
       "      <td>0.499004</td>\n",
       "      <td>0.542796</td>\n",
       "      <td>0.616904</td>\n",
       "      <td>0.676696</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</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",
       "      <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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119995</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.833028</td>\n",
       "      <td>0.634047</td>\n",
       "      <td>0.639118</td>\n",
       "      <td>0.623852</td>\n",
       "      <td>0.598042</td>\n",
       "      <td>0.613583</td>\n",
       "      <td>0.623852</td>\n",
       "      <td>0.628958</td>\n",
       "      <td>0.623852</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119996</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.825971</td>\n",
       "      <td>0.452105</td>\n",
       "      <td>0.082227</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.137119</td>\n",
       "      <td>0.201082</td>\n",
       "      <td>0.165688</td>\n",
       "      <td>0.158125</td>\n",
       "      <td>0.165688</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119997</th>\n",
       "      <td>0.951745</td>\n",
       "      <td>0.916261</td>\n",
       "      <td>0.667525</td>\n",
       "      <td>0.351985</td>\n",
       "      <td>0.255330</td>\n",
       "      <td>0.197383</td>\n",
       "      <td>0.173536</td>\n",
       "      <td>0.141934</td>\n",
       "      <td>0.134542</td>\n",
       "      <td>0.127112</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119998</th>\n",
       "      <td>0.927669</td>\n",
       "      <td>0.677190</td>\n",
       "      <td>0.242906</td>\n",
       "      <td>0.055362</td>\n",
       "      <td>0.102120</td>\n",
       "      <td>0.072236</td>\n",
       "      <td>0.021010</td>\n",
       "      <td>0.038288</td>\n",
       "      <td>0.048557</td>\n",
       "      <td>0.017530</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119999</th>\n",
       "      <td>0.665321</td>\n",
       "      <td>0.527064</td>\n",
       "      <td>0.516663</td>\n",
       "      <td>0.376442</td>\n",
       "      <td>0.489262</td>\n",
       "      <td>0.480725</td>\n",
       "      <td>0.459160</td>\n",
       "      <td>0.482864</td>\n",
       "      <td>0.469983</td>\n",
       "      <td>0.399270</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>120000 rows × 206 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               0         1         2         3         4         5         6  \\\n",
       "id                                                                             \n",
       "0       0.991230  0.943533  0.764677  0.618571  0.379632  0.190822  0.040237   \n",
       "1       0.971482  0.928969  0.572933  0.178457  0.122962  0.132360  0.094392   \n",
       "2       1.000000  0.959149  0.701378  0.231778  0.000000  0.080698  0.128376   \n",
       "3       0.975795  0.934088  0.659637  0.249921  0.237116  0.281445  0.249921   \n",
       "4       0.000000  0.055816  0.261294  0.359847  0.433143  0.453698  0.499004   \n",
       "...          ...       ...       ...       ...       ...       ...       ...   \n",
       "119995  1.000000  0.833028  0.634047  0.639118  0.623852  0.598042  0.613583   \n",
       "119996  1.000000  0.825971  0.452105  0.082227  0.000000  0.137119  0.201082   \n",
       "119997  0.951745  0.916261  0.667525  0.351985  0.255330  0.197383  0.173536   \n",
       "119998  0.927669  0.677190  0.242906  0.055362  0.102120  0.072236  0.021010   \n",
       "119999  0.665321  0.527064  0.516663  0.376442  0.489262  0.480725  0.459160   \n",
       "\n",
       "               7         8         9  ...  196  197  198  199  200  201  202  \\\n",
       "id                                    ...                                      \n",
       "0       0.025995  0.031709  0.065524  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "1       0.089575  0.030481  0.040499  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "2       0.187448  0.280826  0.328261  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "3       0.249921  0.241397  0.230670  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "4       0.542796  0.616904  0.676696  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "...          ...       ...       ...  ...  ...  ...  ...  ...  ...  ...  ...   \n",
       "119995  0.623852  0.628958  0.623852  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "119996  0.165688  0.158125  0.165688  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "119997  0.141934  0.134542  0.127112  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "119998  0.038288  0.048557  0.017530  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "119999  0.482864  0.469983  0.399270  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
       "\n",
       "        203  204  label  \n",
       "id                       \n",
       "0       0.0  0.0    0.0  \n",
       "1       0.0  0.0    0.0  \n",
       "2       0.0  0.0    2.0  \n",
       "3       0.0  0.0    0.0  \n",
       "4       0.0  0.0    2.0  \n",
       "...     ...  ...    ...  \n",
       "119995  0.0  0.0    NaN  \n",
       "119996  0.0  0.0    NaN  \n",
       "119997  0.0  0.0    NaN  \n",
       "119998  0.0  0.0    NaN  \n",
       "119999  0.0  0.0    NaN  \n",
       "\n",
       "[120000 rows x 206 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#切分数据\n",
    "X_train = data[data['label'].notnull()].drop(['label'], axis=1)\n",
    "y_train = data[data['label'].notnull()]['label']\n",
    "\n",
    "X_test = data[data['label'].isnull()].drop(['label'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_m = X_train.melt(id_vars=['id'], #value_vars=[range(0, 205)],\n",
    "                       var_name='time', value_name='heartbeat_signals', \n",
    "                       ignore_index=False).sort_values(by=['id', 'time'], ascending=True)\n",
    "\n",
    "X_test_m = X_test.melt(id_vars=['id'], #value_vars=[range(0, 205)],\n",
    "                     var_name='time', value_name='heartbeat_signals',\n",
    "                     ignore_index=False).sort_values(by=['id', 'time'], ascending=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib.pylab as plt\n",
    "\n",
    "from tsfresh import extract_features, extract_relevant_features, select_features\n",
    "from tsfresh.utilities.dataframe_functions import impute\n",
    "from tsfresh.feature_extraction import ComprehensiveFCParameters\n",
    "\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "extraction_settings = ComprehensiveFCParameters()\n",
    "batch_size = 2000\n",
    "sample_len = 205"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:0, end_ids:410000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:410000, end_ids:820000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.00it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:820000, end_ids:1230000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  1.50it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:1230000, end_ids:1640000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  2.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:1640000, end_ids:2050000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:2050000, end_ids:2460000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  1.00s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:2460000, end_ids:2870000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:00<00:00,  2.35it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:2870000, end_ids:3280000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:3280000, end_ids:3690000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  1.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:3690000, end_ids:4100000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  1.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:4100000, end_ids:4510000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:4510000, end_ids:4920000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  2.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:4920000, end_ids:5330000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  2.48it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:5330000, end_ids:5740000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:5740000, end_ids:6150000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  1.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:6150000, end_ids:6560000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  1.95it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:6560000, end_ids:6970000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  2.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:6970000, end_ids:7380000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:7380000, end_ids:7790000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:00<00:00,  2.76it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:7790000, end_ids:8200000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  1.01s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:8200000, end_ids:8610000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:01<00:00,  3.00it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:8610000, end_ids:9020000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:9020000, end_ids:9430000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:9430000, end_ids:9840000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:00<00:00,  2.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:9840000, end_ids:10250000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.74it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:10250000, end_ids:10660000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.36it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:10660000, end_ids:11070000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  1.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:11070000, end_ids:11480000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.97it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:11480000, end_ids:11890000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  1.01it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:11890000, end_ids:12300000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.37it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:12300000, end_ids:12710000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:12710000, end_ids:13120000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:13120000, end_ids:13530000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:13530000, end_ids:13940000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  1.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:13940000, end_ids:14350000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:14350000, end_ids:14760000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:14760000, end_ids:15170000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:15170000, end_ids:15580000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:56<00:00,  2.38it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:15580000, end_ids:15990000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:15990000, end_ids:16400000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:16400000, end_ids:16810000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  3.17it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:16810000, end_ids:17220000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  1.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:17220000, end_ids:17630000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.21it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:17630000, end_ids:18040000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:55<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:18040000, end_ids:18450000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:55<00:00,  1.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:18450000, end_ids:18860000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.31it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:18860000, end_ids:19270000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:05<00:00,  1.71it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:19270000, end_ids:19680000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:19680000, end_ids:20090000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  1.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start_ids:20090000, end_ids:20500000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  3.38it/s]\n"
     ]
    }
   ],
   "source": [
    "features_train = pd.DataFrame()\n",
    "for i in range(0, X_train_m.shape[0], batch_size * sample_len):\n",
    "    start_ids = i\n",
    "    end_ids = start_ids + batch_size * sample_len\n",
    "    print('start_ids:{}, end_ids:{}'.format(start_ids, end_ids))\n",
    "\n",
    "    temp_df = X_train_m.iloc[start_ids:end_ids]\n",
    "\n",
    "    temp_df = extract_features(temp_df, column_id='id', column_sort='time',\n",
    "                     default_fc_parameters=extraction_settings,\n",
    "                     # we impute = remove all NaN features automatically\n",
    "                     impute_function=impute)\n",
    "    features_train = features_train.append(temp_df)\n",
    "    del temp_df\n",
    "    gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('./work/features/features_train.pkl', 'wb') as file:\n",
    "    pickle.dump(features_train, file) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 59/59 [01:01<00:00,  1.04s/it]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.31it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.89it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:57<00:00,  2.97it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [01:00<00:00,  2.01it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  1.97it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:59<00:00,  3.08it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.69it/s]\n",
      "Feature Extraction: 100%|██████████| 59/59 [01:02<00:00,  1.05s/it]\n",
      "Feature Extraction: 100%|██████████| 59/59 [00:58<00:00,  2.15it/s]\n"
     ]
    }
   ],
   "source": [
    "features_test = pd.DataFrame()\n",
    "for i in range(0, X_test_m.shape[0], batch_size * sample_len):\n",
    "    start_ids = i\n",
    "    end_ids = start_ids + batch_size * sample_len\n",
    "\n",
    "    temp_df = X_test_m.iloc[start_ids:end_ids]\n",
    "\n",
    "    temp_df = extract_features(temp_df, column_id='id', column_sort='time',\n",
    "                     default_fc_parameters=extraction_settings,\n",
    "                     # we impute = remove all NaN features automatically\n",
    "                     impute_function=impute)\n",
    "    features_test = features_test.append(temp_df)\n",
    "    del temp_df\n",
    "    gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('./work/features/features_test.pkl', 'wb') as file:\n",
    "    pickle.dump(features_test, file) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# with open('./work/features/features_train.pkl', 'rb') as file:\n",
    "#     features_train = pickle.load(file) \n",
    "\n",
    "# with open('./work/features/features_test.pkl', 'rb') as file:\n",
    "#     features_test = pickle.load(file) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_filtered = select_features(features_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_filtered = features_test[X_train_filtered.columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train.set_index('id', inplace=True)\n",
    "X_test.set_index('id', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = pd.merge(X_train, X_train_filtered, left_index=True, right_index=True)\n",
    "X_test = pd.merge(X_test, X_test_filtered, left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16219"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del X_train_filtered, X_test_filtered, features_train, features_test, data\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 使用XGBoost训练"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_xgb = xgb.XGBClassifier(\n",
    "            max_depth=6, learning_rate=0.01, n_estimators=2500, \n",
    "            objective='multi:softprob', tree_method='gpu_hist', \n",
    "            num_class=4,\n",
    "            subsample=0.8, colsample_bytree=0.8, \n",
    "            min_child_samples=3, eval_metric=['auc', 'mae', 'merror', 'mlogloss'], reg_lambda=0.5,\n",
    "            seed=2022\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[00:12:15] WARNING: ../src/learner.cc:541: \n",
      "Parameters: { min_child_samples } might not be used.\n",
      "\n",
      "  This may not be accurate due to some parameters are only used in language bindings but\n",
      "  passed down to XGBoost core.  Or some parameters are not used but slip through this\n",
      "  verification. Please open an issue if you find above cases.\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model_xgb = model_xgb.fit(X_train, y_train, #eval_set=[(X_train, y_train)],\n",
    "                          verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = model_xgb.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = pd.get_dummies(y_pred.argmax(axis=1)).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame({'id' : X_test.index,\n",
    "                           'label_0' : y_pred[:, 0],\n",
    "                           'label_1' : y_pred[:, 1],\n",
    "                           'label_2' : y_pred[:, 2],\n",
    "                           'label_3' : y_pred[:, 3],\n",
    "                           })\n",
    "submission.to_csv('./work/sumission/submission_XGB.csv', index = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}