Negative_transfer_experiment.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Repositories uploaded!!\n"
     ]
    }
   ],
   "source": [
    "# from two_TrAdaBoostR2 import TwoStageTrAdaBoostR2 ##STrAdaBoost.R2\n",
    "# from TwoStageTrAdaBoostR2 import TwoStageTrAdaBoostR2 ##two-stage TrAdaBoost.R2\n",
    "\n",
    "import pandas as pd\n",
    "import sys\n",
    "import numpy as np\n",
    "from pandas import DataFrame\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import AdaBoostRegressor\n",
    "\n",
    "from keras.models import Sequential, load_model, Model\n",
    "from keras.layers import Input, Dense, Activation, Conv2D, Dropout, Flatten\n",
    "from keras import optimizers, utils, initializers, regularizers\n",
    "import keras.backend as K\n",
    "\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.preprocessing import StandardScaler #Importing the StandardScaler\n",
    "\n",
    "from itertools import combinations\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from scipy.stats.stats import pearsonr\n",
    "from math import sqrt\n",
    "\n",
    "#Geo plotting libraries\n",
    "import geopandas as gdp\n",
    "from matplotlib.colors import ListedColormap\n",
    "# import geoplot as glpt\n",
    "\n",
    "import xgboost as xgb\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn import linear_model\n",
    "from keras.models import Sequential\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "\n",
    "from sklearn.model_selection import KFold\n",
    "import matplotlib.lines as mlines\n",
    "\n",
    "import statistics \n",
    "\n",
    "from scipy.stats import rv_continuous\n",
    "from scipy.stats import *\n",
    "\n",
    "from statistics import mean\n",
    "from sklearn.cluster import KMeans\n",
    "from scipy.spatial import distance\n",
    "\n",
    "\n",
    "######### Instance Transfer repositories ####################\n",
    "from adapt.instance_based import TwoStageTrAdaBoostR2\n",
    "\n",
    "print(\"Repositories uploaded!!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Second Upload Completed!!\n"
     ]
    }
   ],
   "source": [
    "from adapt.instance_based import TrAdaBoost, TrAdaBoostR2, TwoStageTrAdaBoostR2\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from adapt.instance_based import KMM\n",
    "\n",
    "print(\"Second Upload Completed!!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Concrete Data\n",
      "-------------------------------------------\n",
      "(1030, 9)\n",
      "Target Set:  (653, 8) 63.398058252427184\n",
      "Source Set:  (121, 8) 11.747572815533982\n",
      "Test Set:  (256, 8) 24.854368932038835\n",
      "-------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "################################### Concrete ###########################################################################################################\n",
    "ConcreteData_df = pd.read_excel('UCI_regression/Concrete/Concrete_Data.xls') ## 'Cement' found to be correlated at 0.4 :: 100\n",
    "print(\"Concrete Data\")\n",
    "print(\"-------------------------------------------\")\n",
    "print(ConcreteData_df.shape)\n",
    "\n",
    "# concrete_cols = ConcreteData_df.columns\n",
    "# ss = StandardScaler()\n",
    "# ConcreteData_df[concrete_cols] = ss.fit_transform(ConcreteData_df[concrete_cols])\n",
    "# print(ConcreteData_df)\n",
    "\n",
    "\n",
    "drop_col_concrete = ['Cement']\n",
    "\n",
    "concrete_tgt_df = ConcreteData_df.loc[(ConcreteData_df['Cement'] <= 310)]\n",
    "concrete_tgt_df = concrete_tgt_df.drop(drop_col_concrete, axis = 1)\n",
    "concrete_tgt_df = concrete_tgt_df.reset_index(drop=True)\n",
    "print(\"Target Set: \",concrete_tgt_df.shape, ((concrete_tgt_df.shape[0]/ConcreteData_df.shape[0])*100))\n",
    "\n",
    "concrete_source_df = ConcreteData_df.loc[(ConcreteData_df['Cement'] > 310) & (ConcreteData_df['Cement'] <= 350)]\n",
    "concrete_source_df = concrete_source_df.drop(drop_col_concrete, axis = 1)\n",
    "concrete_source_df = concrete_source_df.reset_index(drop=True)\n",
    "print(\"Source Set: \",concrete_source_df.shape, ((concrete_source_df.shape[0]/ConcreteData_df.shape[0])*100))\n",
    "\n",
    "concrete_test_df = ConcreteData_df.loc[(ConcreteData_df['Cement'] > 350)]\n",
    "concrete_test_df = concrete_test_df.drop(drop_col_concrete, axis = 1)\n",
    "concrete_test_df = concrete_test_df.reset_index(drop=True)\n",
    "print(\"Test Set: \",concrete_test_df.shape, ((concrete_test_df.shape[0]/ConcreteData_df.shape[0])*100))\n",
    "\n",
    "#################### Splitting into features and target ####################\n",
    "target_column_concrete = ['ConcreteCompressiveStrength']\n",
    "\n",
    "concrete_tgt_df_y = concrete_tgt_df[target_column_concrete]\n",
    "concrete_tgt_df_X = concrete_tgt_df.drop(target_column_concrete, axis = 1)\n",
    "\n",
    "concrete_source_df_y = concrete_source_df[target_column_concrete]\n",
    "concrete_source_df_X = concrete_source_df.drop(target_column_concrete, axis = 1)\n",
    "\n",
    "concrete_test_df_y = concrete_test_df[target_column_concrete]\n",
    "concrete_test_df_X = concrete_test_df.drop(target_column_concrete, axis = 1)\n",
    "\n",
    "############## Merging the datasets ##########################################\n",
    "concrete_X_df = pd.concat([concrete_tgt_df_X, concrete_source_df_X], ignore_index=True)\n",
    "concrete_y_df = pd.concat([concrete_tgt_df_y, concrete_source_df_y], ignore_index=True)\n",
    "\n",
    "concrete_np_train_X = concrete_X_df.to_numpy()\n",
    "concrete_np_train_y = concrete_y_df.to_numpy()\n",
    "\n",
    "concrete_np_test_X = concrete_test_df_X.to_numpy()\n",
    "concrete_np_test_y = concrete_test_df_y.to_numpy()\n",
    "\n",
    "concrete_np_train_y_list = concrete_np_train_y.ravel()\n",
    "concrete_np_test_y_list = concrete_np_test_y.ravel()\n",
    "\n",
    "src_size_concrete = len(concrete_source_df_y)\n",
    "tgt_size_concrete = len(concrete_tgt_df_y)\n",
    "\n",
    "print(\"-------------------------------------------\")\n",
    "\n",
    "################################################################################################################################################################"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Specification requirement complete!\n"
     ]
    }
   ],
   "source": [
    "############# Transfer Learning specifications Concrete #########################\n",
    "\n",
    "from adapt.instance_based import TrAdaBoost, TrAdaBoostR2, TwoStageTrAdaBoostR2\n",
    "\n",
    "\n",
    "src_idx = np.arange(start=0, stop = (src_size_concrete - 1), step=1)\n",
    "tgt_idx = np.arange(start=src_size_concrete, stop = ((src_size_concrete + tgt_size_concrete)-1), step=1)\n",
    "\n",
    "\n",
    "def get_estimator(**kwargs):\n",
    "    return DecisionTreeRegressor(max_depth = 6)\n",
    "\n",
    "kwargs_TwoTrAda = {'steps': 30,\n",
    "                    'fold': 10,\n",
    "                  'learning_rate': 0.1}\n",
    "\n",
    "print(\"Specification requirement complete!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Two-Stage TrAdaboost.R2 (M + H, L)\n",
      "-------------------------------------------\n",
      "RMSE of Two-Stage TrAdaboost.R2: 14.582486877132064\n",
      "R^2 of TrAdaboostR2: 0.6125019055957691\n",
      "-------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "########################### TwoStagetrAdaBoostR2 Concrete #######################################\n",
    "\n",
    "from adapt.instance_based import TrAdaBoost, TrAdaBoostR2, TwoStageTrAdaBoostR2\n",
    "\n",
    "\n",
    "print(\"Two-Stage TrAdaboost.R2 (M + H, L)\")\n",
    "print(\"-------------------------------------------\")\n",
    "\n",
    "r2scorelist_TwoTrAda_concrete = []\n",
    "rmselist_TwoTrAda_concrete = []\n",
    "\n",
    "for x in range(0, 4):\n",
    "\n",
    "    model_TwoTrAda_concrete = TwoStageTrAdaBoostR2(get_estimator = get_estimator, n_estimators = 1000, cv=10) #, kwargs_TwoTrAda)\n",
    "    model_TwoTrAda_concrete.fit(concrete_np_train_X, concrete_np_train_y_list, src_idx, tgt_idx)\n",
    "\n",
    "    y_pred_TwoTrAda_concrete = model_TwoTrAda_concrete.predict(concrete_np_test_X)\n",
    "    \n",
    "    mse_TwoTrAda_concrete = sqrt(mean_squared_error(concrete_np_test_y, y_pred_TwoTrAda_concrete))\n",
    "    rmselist_TwoTrAda_concrete.append(mse_TwoTrAda_concrete)\n",
    "        \n",
    "    r2_score_TwoTrAda_concrete = pearsonr(concrete_np_test_y_list, y_pred_TwoTrAda_concrete)\n",
    "    r2_score_TwoTrAda_concrete = (r2_score_TwoTrAda_concrete[0])**2\n",
    "    r2scorelist_TwoTrAda_concrete.append(r2_score_TwoTrAda_concrete)\n",
    "\n",
    "\n",
    "print(\"RMSE of Two-Stage TrAdaboost.R2:\", statistics.mean(rmselist_TwoTrAda_concrete))\n",
    "print(\"R^2 of TrAdaboostR2:\", statistics.mean(r2scorelist_TwoTrAda_concrete))\n",
    "\n",
    "print(\"-------------------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adaboost.R2 (L + M, H)\n",
      "-------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/shrey/anaconda3/lib/python3.8/site-packages/sklearn/utils/validation.py:73: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(**kwargs)\n",
      "/Users/shrey/anaconda3/lib/python3.8/site-packages/sklearn/utils/validation.py:73: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(**kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE of Adaboost.R2: 16.415724044304717\n",
      "R^2 of AdaboostR2: 0.6126660748483308\n",
      "\n",
      "\n",
      "RMSE of Adaboost.R2: [16.686927605488663, 16.14452048312077]\n",
      "R^2 of AdaboostR2: [0.6058907027228013, 0.6194414469738604]\n",
      "-------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "########################### AdaBoostR2 Concrete #####################################################\n",
    "\n",
    "from sklearn.ensemble import AdaBoostRegressor\n",
    "\n",
    "\n",
    "print(\"Adaboost.R2 (L + M, H)\")\n",
    "print(\"-------------------------------------------\")\n",
    "\n",
    "r2scorelist_Ada_concrete = []\n",
    "rmselist_Ada_concrete = []\n",
    "\n",
    "for x in range(0, 2):\n",
    "\n",
    "    model_Ada_concrete = AdaBoostRegressor(DecisionTreeRegressor(max_depth = 6), learning_rate = 0.1, n_estimators = 100) \n",
    "    model_Ada_concrete.fit(concrete_tgt_df_X, concrete_tgt_df_y)\n",
    "    \n",
    "    y_pred_ada_concrete = model_Ada_concrete.predict(concrete_np_test_X)\n",
    "    \n",
    "    mse_Ada_concrete = sqrt(mean_squared_error(concrete_np_test_y, y_pred_ada_concrete))\n",
    "    rmselist_Ada_concrete.append(mse_Ada_concrete)\n",
    "        \n",
    "    r2_score_Ada_concrete = pearsonr(concrete_np_test_y_list, y_pred_ada_concrete)\n",
    "    r2_score_Ada_concrete = (r2_score_Ada_concrete[0])**2\n",
    "    r2scorelist_Ada_concrete.append(r2_score_Ada_concrete)\n",
    "\n",
    "\n",
    "print(\"RMSE of Adaboost.R2:\", statistics.mean(rmselist_Ada_concrete))\n",
    "print(\"R^2 of AdaboostR2:\", statistics.mean(r2scorelist_Ada_concrete))\n",
    "print(\"\\n\")\n",
    "print(\"RMSE of Adaboost.R2:\", rmselist_Ada_concrete)\n",
    "print(\"R^2 of AdaboostR2:\", r2scorelist_Ada_concrete)\n",
    "\n",
    "print(\"-------------------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient Boosting Regression Transfer Learning (L + M, H)\n",
      "-------------------------------------------\n",
      "RMSE of GBR: 11.297287802466293\n",
      "R^2 of GBR: 0.6852325217936286\n",
      "\n",
      "\n",
      "RMSE of GBR: [11.203363954598267, 11.391211650334318]\n",
      "R^2 of GBR: [0.6750797313177301, 0.6953853122695272]\n",
      "-------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "######################### Gradient Boosting Regression Transfer Learning Concrete #######################################\n",
    "\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "\n",
    "\n",
    "print(\"Gradient Boosting Regression Transfer Learning (L + M, H)\")\n",
    "print(\"-------------------------------------------\")\n",
    "\n",
    "r2scorelist_GBRTL_concrete = []\n",
    "rmselist_GBRTL_concrete = []\n",
    "\n",
    "for x in range(0, 2):\n",
    "\n",
    "    model_GBRTL_concrete = GradientBoostingRegressor(learning_rate = 0.1, max_depth = 6, n_estimators = 100, subsample = 0.5)\n",
    "    model_GBRTL_concrete.fit(concrete_np_train_X, concrete_np_train_y_list)\n",
    "\n",
    "    y_pred_GBRTL_concrete = model_GBRTL_concrete.predict(concrete_test_df_X) ##Using dataframe instead of the numpy matrix\n",
    "    \n",
    "    mse_GBRTL_concrete = sqrt(mean_squared_error(concrete_np_test_y, y_pred_GBRTL_concrete))\n",
    "    rmselist_GBRTL_concrete.append(mse_GBRTL_concrete)\n",
    "        \n",
    "    r2_score_GBRTL_concrete = pearsonr(concrete_np_test_y_list, y_pred_GBRTL_concrete)\n",
    "    r2_score_GBRTL_concrete = (r2_score_GBRTL_concrete[0])**2\n",
    "    r2scorelist_GBRTL_concrete.append(r2_score_GBRTL_concrete)\n",
    "\n",
    "\n",
    "print(\"RMSE of GBR:\", statistics.mean(rmselist_GBRTL_concrete))\n",
    "print(\"R^2 of GBR:\", statistics.mean(r2scorelist_GBRTL_concrete))\n",
    "print(\"\\n\")\n",
    "print(\"RMSE of GBR:\", rmselist_GBRTL_concrete)\n",
    "print(\"R^2 of GBR:\", r2scorelist_GBRTL_concrete)\n",
    "\n",
    "print(\"-------------------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "######################## Concrete #########################################################################\n",
    "TrAdaboost_rmse = [14.5734, 14.8614, 16.9801, 14.5824]\n",
    "Adaboost_rmse = [24.4403, 22.2673, 19.6041, 16.4157]\n",
    "GBR_rmse = [11.4610, 11.0932, 11.7888, 11.2972]\n",
    "\n",
    "TrAdaboost_r2 = [0.6389, 0.5835, 0.5196, 0.6125]\n",
    "Adaboost_r2 = [0.4624, 0.5643, 0.5579, 0.6126]\n",
    "GBR_r2 = [0.7007, 0.7009, 0.6808, 0.6852]\n",
    "\n",
    "\n",
    "#############################################################################################################\n",
    "\n",
    "xlim_list = [2, 4, 6, 8]\n",
    "box = mlines.Line2D([], [], color = 'tomato', marker = 's', linestyle = 'None',markersize = 20, label = 'TrAdaBoost.R2')\n",
    "circle = mlines.Line2D([], [], color = 'navy', marker = 'o', linestyle = 'None', markersize = 20, label='AdaBoost.R2')\n",
    "# triangle = mlines.Line2D([], [], color = 'olive', marker = 'v', linestyle = 'None', markersize = 20, label='GBR.TL')\n",
    "\n",
    "\n",
    "\n",
    "# line1 = mlines.Line2D([], [], color = 'black', marker='None',linestyle='-',\n",
    "#                           markersize = 30, label='Case 1', linewidth = 6.0)\n",
    "# line2 = mlines.Line2D([], [], color = 'black', marker='None', linestyle='--',\n",
    "#                           markersize = 30, label='Case 2', linewidth = 6.0)\n",
    "# line3 = mlines.Line2D([], [], color = 'black', marker='None', linestyle=':',\n",
    "#                           markersize = 30, label='Case 3', linewidth = 6.0)\n",
    "\n",
    "plt.figure(figsize = (15,15))\n",
    "\n",
    "plt.plot(xlim_list, TrAdaboost_r2, '-s', color = 'tomato', linewidth = 4.0, markersize = 20)\n",
    "plt.plot(xlim_list, Adaboost_r2, '-o', color = 'navy', linewidth = 4.0, markersize = 20)\n",
    "# plt.plot(xlim_list, GBR_r2, '-v', color = 'olive', linewidth = 4.0, markersize = 20)\n",
    "\n",
    "# plt.plot(xlim_list, tradaboostr2_vals_c1, '-ok', color = 'navy', linewidth = 4.0, markersize = 20)\n",
    "# plt.plot(xlim_list, tradaboostr2_vals_c2, '--ok', color = 'navy', linewidth = 4.0, markersize = 20)\n",
    "# plt.plot(xlim_list, tradaboostr2_vals_c3, ':ok', color = 'navy', linewidth = 4.0, markersize = 20)\n",
    "\n",
    "plt.xticks([2, 4, 6, 8], [35, 43, 53, 63], fontsize = 30)\n",
    "plt.yticks(fontsize = 30)\n",
    "\n",
    "plt.ylabel('R2 score', fontsize = 35)\n",
    "plt.xlabel('% Data as Target', fontsize = 40)\n",
    "\n",
    "plt.legend(handles= [box, circle], fontsize = 30, loc='lower right')\n",
    "\n",
    "# plt.title('UCI Auto MPG dataset', fontsize = 28)\n",
    "plt.savefig(\"paper_negative_transfer\")\n",
    "plt.show()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "def augmented_euclidean_dist(a, b):\n",
    "    # Write your code here.\n",
    "    # Remember to return the right object.\n",
    "    \n",
    "    for idx in range(0,len(a)):\n",
    "            if a == -999:\n",
    "\n",
    "        while len(a) != 0:\n",
    "            a.remove(-999)\n",
    "    \n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "# a = np.array([-999.0, 0.1, 2.3, 0.4])\n",
    "# b = np.array([9.0, 1.2, -999, 0.5])\n",
    "\n",
    "a = np.array([  1.  ,   2.  ,   3.  ,   5.5 ,  23.23, -11.11])\n",
    "b = np.array([  1.  ,   2.  ,   3.  ,   5.5 ,  23.23, -11.11])\n",
    "\n",
    "# a, b = list(zip(*[(a, b) for a, b in zip(a, b) if a == -999.0 and b == -999]))\n",
    "\n",
    "i = 0\n",
    "# for i in range(0, (a.size - 1)):\n",
    "while i < (a.size - 1):\n",
    "    if a[i] == -999.0:\n",
    "        a = np.delete(a,i)\n",
    "        b = np.delete(b,i)\n",
    "        i = i + 1\n",
    "        print(a)\n",
    "        print(b)\n",
    "    elif b[i] == -999.0:\n",
    "        a = np.delete(a,i)\n",
    "        b = np.delete(b,i)\n",
    "        i = i + 1\n",
    "        print(a)\n",
    "        print(b)\n",
    "    else:\n",
    "        i = i + 1\n",
    "    \n",
    "if a.size < 2 and b.size < 2:\n",
    "    print(\"No\")\n",
    "else:\n",
    "    temp = a - b\n",
    "    sum_sq = np.dot(temp.T, temp)\n",
    "    dist = np.sqrt(sum_sq)\n",
    "    print(dist)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([   1.2,    2.2,    3.1,    4.4, -999. ,    9.9, -999. ,   14. ,\n",
       "         16.2])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = np.array([-2, -1.4, -1.1, 0, 1.2, 2.2, 3.1, 4.4, -999.0, 9.9, -999.0, 14, 16.2])\n",
    "\n",
    "# for num in a:\n",
    "#     if num != -999.0:\n",
    "#         if num > 0.0:\n",
    "            \n",
    "a = np.array( [num for num in a if (num == -999.0) or (num > 0.0)] )\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}