learn_pytorch_28_pytorch实战项目之多元线性回归房价预测2完整版.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.34727322  0.00728832  0.53615372 ...  1.07761115 -0.48852299\n",
      " -0.42084081]\n",
      "354\n",
      "迭代次数：100,  loss值：0.65879297\n",
      "迭代次数：200,  loss值：0.45392793\n",
      "迭代次数：300,  loss值：0.33683515\n",
      "迭代次数：400,  loss值：0.27156311\n",
      "迭代次数：500,  loss值：0.23424567\n",
      "迭代次数：600,  loss值：0.21107636\n",
      "迭代次数：700,  loss值：0.19508490\n",
      "迭代次数：800,  loss值：0.18303314\n",
      "迭代次数：900,  loss值：0.17341208\n",
      "迭代次数：1000,  loss值：0.16545159\n",
      "迭代次数：1100,  loss值：0.15870337\n",
      "迭代次数：1200,  loss值：0.15287676\n",
      "迭代次数：1300,  loss值：0.14777122\n",
      "迭代次数：1400,  loss值：0.14324392\n",
      "迭代次数：1500,  loss值：0.13919130\n",
      "迭代次数：1600,  loss值：0.13553673\n",
      "迭代次数：1700,  loss值：0.13222224\n",
      "迭代次数：1800,  loss值：0.12920287\n",
      "迭代次数：1900,  loss值：0.12644278\n",
      "迭代次数：2000,  loss值：0.12391267\n",
      "迭代次数：2100,  loss值：0.12158803\n",
      "迭代次数：2200,  loss值：0.11944795\n",
      "迭代次数：2300,  loss值：0.11747428\n",
      "迭代次数：2400,  loss值：0.11565106\n",
      "迭代次数：2500,  loss值：0.11396409\n",
      "迭代次数：2600,  loss值：0.11240061\n",
      "迭代次数：2700,  loss值：0.11094917\n",
      "迭代次数：2800,  loss值：0.10959933\n",
      "迭代次数：2900,  loss值：0.10834166\n",
      "迭代次数：3000,  loss值：0.10716758\n",
      "迭代次数：3100,  loss值：0.10606938\n",
      "迭代次数：3200,  loss值：0.10504004\n",
      "迭代次数：3300,  loss值：0.10407327\n",
      "迭代次数：3400,  loss值：0.10316339\n",
      "迭代次数：3500,  loss值：0.10230533\n",
      "迭代次数：3600,  loss值：0.10149455\n",
      "迭代次数：3700,  loss值：0.10072701\n",
      "迭代次数：3800,  loss值：0.09999906\n",
      "迭代次数：3900,  loss值：0.09930751\n",
      "迭代次数：4000,  loss值：0.09864950\n",
      "迭代次数：4100,  loss值：0.09802243\n",
      "迭代次数：4200,  loss值：0.09742402\n",
      "迭代次数：4300,  loss值：0.09685224\n",
      "迭代次数：4400,  loss值：0.09630521\n",
      "迭代次数：4500,  loss值：0.09578127\n",
      "迭代次数：4600,  loss值：0.09527891\n",
      "迭代次数：4700,  loss值：0.09479676\n",
      "迭代次数：4800,  loss值：0.09433357\n",
      "迭代次数：4900,  loss值：0.09388817\n",
      "迭代次数：5000,  loss值：0.09345952\n",
      "迭代次数：5100,  loss值：0.09304662\n",
      "迭代次数：5200,  loss值：0.09264859\n",
      "迭代次数：5300,  loss值：0.09226458\n",
      "迭代次数：5400,  loss值：0.09189382\n",
      "迭代次数：5500,  loss值：0.09153559\n",
      "迭代次数：5600,  loss值：0.09118922\n",
      "迭代次数：5700,  loss值：0.09085409\n",
      "迭代次数：5800,  loss值：0.09052962\n",
      "迭代次数：5900,  loss值：0.09021527\n",
      "迭代次数：6000,  loss值：0.08991054\n",
      "迭代次数：6100,  loss值：0.08961495\n",
      "迭代次数：6200,  loss值：0.08932807\n",
      "迭代次数：6300,  loss值：0.08904944\n",
      "迭代次数：6400,  loss值：0.08877869\n",
      "迭代次数：6500,  loss值：0.08851542\n",
      "迭代次数：6600,  loss值：0.08825926\n",
      "迭代次数：6700,  loss值：0.08800984\n",
      "迭代次数：6800,  loss值：0.08776681\n",
      "迭代次数：6900,  loss值：0.08752984\n",
      "迭代次数：7000,  loss值：0.08729856\n",
      "迭代次数：7100,  loss值：0.08707266\n",
      "迭代次数：7200,  loss值：0.08685184\n",
      "迭代次数：7300,  loss值：0.08663577\n",
      "迭代次数：7400,  loss值：0.08642415\n",
      "迭代次数：7500,  loss值：0.08621670\n",
      "迭代次数：7600,  loss值：0.08601315\n",
      "迭代次数：7700,  loss值：0.08581325\n",
      "迭代次数：7800,  loss值：0.08561675\n",
      "迭代次数：7900,  loss值：0.08542343\n",
      "迭代次数：8000,  loss值：0.08523309\n",
      "迭代次数：8100,  loss值：0.08504555\n",
      "迭代次数：8200,  loss值：0.08486059\n",
      "迭代次数：8300,  loss值：0.08467810\n",
      "迭代次数：8400,  loss值：0.08449794\n",
      "迭代次数：8500,  loss值：0.08431996\n",
      "迭代次数：8600,  loss值：0.08414409\n",
      "迭代次数：8700,  loss值：0.08397018\n",
      "迭代次数：8800,  loss值：0.08379817\n",
      "迭代次数：8900,  loss值：0.08362798\n",
      "迭代次数：9000,  loss值：0.08345956\n",
      "迭代次数：9100,  loss值：0.08329282\n",
      "迭代次数：9200,  loss值：0.08312773\n",
      "迭代次数：9300,  loss值：0.08296424\n",
      "迭代次数：9400,  loss值：0.08280233\n",
      "迭代次数：9500,  loss值：0.08264194\n",
      "迭代次数：9600,  loss值：0.08248308\n",
      "迭代次数：9700,  loss值：0.08232572\n",
      "迭代次数：9800,  loss值：0.08216984\n",
      "迭代次数：9900,  loss值：0.08201544\n",
      "迭代次数：10000,  loss值：0.08186252\n",
      "迭代次数：10100,  loss值：0.08171108\n",
      "迭代次数：10200,  loss值：0.08156110\n",
      "迭代次数：10300,  loss值：0.08141260\n",
      "迭代次数：10400,  loss值：0.08126558\n",
      "迭代次数：10500,  loss值：0.08112004\n",
      "迭代次数：10600,  loss值：0.08097599\n",
      "迭代次数：10700,  loss值：0.08083346\n",
      "迭代次数：10800,  loss值：0.08069242\n",
      "迭代次数：10900,  loss值：0.08055288\n",
      "迭代次数：11000,  loss值：0.08041486\n",
      "迭代次数：11100,  loss值：0.08027834\n",
      "迭代次数：11200,  loss值：0.08014333\n",
      "迭代次数：11300,  loss值：0.08000983\n",
      "迭代次数：11400,  loss值：0.07987782\n",
      "迭代次数：11500,  loss值：0.07974738\n",
      "迭代次数：11600,  loss值：0.07961851\n",
      "迭代次数：11700,  loss值：0.07949118\n",
      "迭代次数：11800,  loss值：0.07936536\n",
      "迭代次数：11900,  loss值：0.07924105\n",
      "迭代次数：12000,  loss值：0.07911827\n",
      "迭代次数：12100,  loss值：0.07899697\n",
      "迭代次数：12200,  loss值：0.07887714\n",
      "迭代次数：12300,  loss值：0.07875881\n",
      "迭代次数：12400,  loss值：0.07864191\n",
      "迭代次数：12500,  loss值：0.07852647\n",
      "迭代次数：12600,  loss值：0.07841246\n",
      "迭代次数：12700,  loss值：0.07829992\n",
      "迭代次数：12800,  loss值：0.07818875\n",
      "迭代次数：12900,  loss值：0.07807896\n",
      "迭代次数：13000,  loss值：0.07797053\n",
      "迭代次数：13100,  loss值：0.07786346\n",
      "迭代次数：13200,  loss值：0.07775770\n",
      "迭代次数：13300,  loss值：0.07765326\n",
      "迭代次数：13400,  loss值：0.07755017\n",
      "迭代次数：13500,  loss值：0.07744835\n",
      "迭代次数：13600,  loss值：0.07734783\n",
      "迭代次数：13700,  loss值：0.07724860\n",
      "迭代次数：13800,  loss值：0.07715048\n",
      "迭代次数：13900,  loss值：0.07705362\n",
      "迭代次数：14000,  loss值：0.07695797\n",
      "迭代次数：14100,  loss值：0.07686351\n",
      "迭代次数：14200,  loss值：0.07677020\n",
      "迭代次数：14300,  loss值：0.07667803\n",
      "迭代次数：14400,  loss值：0.07658693\n",
      "迭代次数：14500,  loss值：0.07649699\n",
      "迭代次数：14600,  loss值：0.07640808\n",
      "迭代次数：14700,  loss值：0.07632022\n",
      "迭代次数：14800,  loss值：0.07623339\n",
      "迭代次数：14900,  loss值：0.07614752\n",
      "迭代次数：15000,  loss值：0.07606261\n",
      "迭代次数：15100,  loss值：0.07597866\n",
      "迭代次数：15200,  loss值：0.07589561\n",
      "迭代次数：15300,  loss值：0.07581338\n",
      "迭代次数：15400,  loss值：0.07573199\n",
      "迭代次数：15500,  loss值：0.07565141\n",
      "迭代次数：15600,  loss值：0.07557163\n",
      "迭代次数：15700,  loss值：0.07549260\n",
      "迭代次数：15800,  loss值：0.07541419\n",
      "迭代次数：15900,  loss值：0.07533648\n",
      "迭代次数：16000,  loss值：0.07525941\n",
      "迭代次数：16100,  loss值：0.07518294\n",
      "迭代次数：16200,  loss值：0.07510702\n",
      "迭代次数：16300,  loss值：0.07503171\n",
      "迭代次数：16400,  loss值：0.07495691\n",
      "迭代次数：16500,  loss值：0.07488260\n",
      "迭代次数：16600,  loss值：0.07480886\n",
      "迭代次数：16700,  loss值：0.07473549\n",
      "迭代次数：16800,  loss值：0.07466265\n",
      "迭代次数：16900,  loss值：0.07459046\n",
      "迭代次数：17000,  loss值：0.07451830\n",
      "迭代次数：17100,  loss值：0.07444680\n",
      "迭代次数：17200,  loss值：0.07437575\n",
      "迭代次数：17300,  loss值：0.07430513\n",
      "迭代次数：17400,  loss值：0.07423496\n",
      "迭代次数：17500,  loss值：0.07416523\n",
      "迭代次数：17600,  loss值：0.07409594\n",
      "迭代次数：17700,  loss值：0.07402703\n",
      "迭代次数：17800,  loss值：0.07395857\n",
      "迭代次数：17900,  loss值：0.07389057\n",
      "迭代次数：18000,  loss值：0.07382287\n",
      "迭代次数：18100,  loss值：0.07375564\n",
      "迭代次数：18200,  loss值：0.07368882\n",
      "迭代次数：18300,  loss值：0.07362242\n",
      "迭代次数：18400,  loss值：0.07355647\n",
      "迭代次数：18500,  loss值：0.07349088\n",
      "迭代次数：18600,  loss值：0.07342573\n",
      "迭代次数：18700,  loss值：0.07336096\n",
      "迭代次数：18800,  loss值：0.07329663\n",
      "迭代次数：18900,  loss值：0.07323270\n",
      "迭代次数：19000,  loss值：0.07316910\n",
      "迭代次数：19100,  loss值：0.07310595\n",
      "迭代次数：19200,  loss值：0.07304330\n",
      "迭代次数：19300,  loss值：0.07298087\n",
      "迭代次数：19400,  loss值：0.07291891\n",
      "迭代次数：19500,  loss值：0.07285737\n",
      "迭代次数：19600,  loss值：0.07279618\n",
      "迭代次数：19700,  loss值：0.07273539\n",
      "迭代次数：19800,  loss值：0.07267500\n",
      "迭代次数：19900,  loss值：0.07261498\n",
      "迭代次数：20000,  loss值：0.07255533\n",
      "迭代次数：20100,  loss值：0.07249609\n",
      "迭代次数：20200,  loss值：0.07243723\n",
      "迭代次数：20300,  loss值：0.07237867\n",
      "迭代次数：20400,  loss值：0.07232048\n",
      "迭代次数：20500,  loss值：0.07226270\n",
      "迭代次数：20600,  loss值：0.07220535\n",
      "迭代次数：20700,  loss值：0.07214828\n",
      "迭代次数：20800,  loss值：0.07209156\n",
      "迭代次数：20900,  loss值：0.07203525\n",
      "迭代次数：21000,  loss值：0.07197933\n",
      "迭代次数：21100,  loss值：0.07192365\n",
      "迭代次数：21200,  loss值：0.07186836\n",
      "迭代次数：21300,  loss值：0.07181339\n",
      "迭代次数：21400,  loss值：0.07175878\n",
      "迭代次数：21500,  loss值：0.07170465\n",
      "迭代次数：21600,  loss值：0.07165059\n",
      "迭代次数：21700,  loss值：0.07159702\n",
      "迭代次数：21800,  loss值：0.07154380\n",
      "迭代次数：21900,  loss值：0.07149089\n",
      "迭代次数：22000,  loss值：0.07143832\n",
      "迭代次数：22100,  loss值：0.07138607\n",
      "迭代次数：22200,  loss值：0.07133414\n",
      "迭代次数：22300,  loss值：0.07128254\n",
      "迭代次数：22400,  loss值：0.07123131\n",
      "迭代次数：22500,  loss值：0.07118042\n",
      "迭代次数：22600,  loss值：0.07112984\n",
      "迭代次数：22700,  loss值：0.07107966\n",
      "迭代次数：22800,  loss值：0.07102967\n",
      "迭代次数：22900,  loss值：0.07098010\n",
      "迭代次数：23000,  loss值：0.07093077\n",
      "迭代次数：23100,  loss值：0.07088184\n",
      "迭代次数：23200,  loss值：0.07083322\n",
      "迭代次数：23300,  loss值：0.07078494\n",
      "迭代次数：23400,  loss值：0.07073700\n",
      "迭代次数：23500,  loss值：0.07068934\n",
      "迭代次数：23600,  loss值：0.07064205\n",
      "迭代次数：23700,  loss值：0.07059503\n",
      "迭代次数：23800,  loss值：0.07054835\n",
      "迭代次数：23900,  loss值：0.07050198\n",
      "迭代次数：24000,  loss值：0.07045599\n",
      "迭代次数：24100,  loss值：0.07041029\n",
      "迭代次数：24200,  loss值：0.07036493\n",
      "迭代次数：24300,  loss值：0.07031988\n",
      "迭代次数：24400,  loss值：0.07027514\n",
      "迭代次数：24500,  loss值：0.07023074\n",
      "迭代次数：24600,  loss值：0.07018664\n",
      "迭代次数：24700,  loss值：0.07014289\n",
      "迭代次数：24800,  loss值：0.07009943\n",
      "迭代次数：24900,  loss值：0.07005630\n",
      "迭代次数：25000,  loss值：0.07001355\n",
      "迭代次数：25100,  loss值：0.06997097\n",
      "迭代次数：25200,  loss值：0.06992879\n",
      "迭代次数：25300,  loss值：0.06988698\n",
      "迭代次数：25400,  loss值：0.06984548\n",
      "迭代次数：25500,  loss值：0.06980427\n",
      "迭代次数：25600,  loss值：0.06976338\n",
      "迭代次数：25700,  loss值：0.06972283\n",
      "迭代次数：25800,  loss值：0.06968265\n",
      "迭代次数：25900,  loss值：0.06964278\n",
      "迭代次数：26000,  loss值：0.06960312\n",
      "迭代次数：26100,  loss值：0.06956381\n",
      "迭代次数：26200,  loss值：0.06952489\n",
      "迭代次数：26300,  loss值：0.06948613\n",
      "迭代次数：26400,  loss值：0.06944779\n",
      "迭代次数：26500,  loss值：0.06940976\n",
      "迭代次数：26600,  loss值：0.06937204\n",
      "迭代次数：26700,  loss值：0.06933461\n",
      "迭代次数：26800,  loss值：0.06929751\n",
      "迭代次数：26900,  loss值：0.06926075\n",
      "迭代次数：27000,  loss值：0.06922428\n",
      "迭代次数：27100,  loss值：0.06918811\n",
      "迭代次数：27200,  loss值：0.06915229\n",
      "迭代次数：27300,  loss值：0.06911677\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "迭代次数：27400,  loss值：0.06908152\n",
      "迭代次数：27500,  loss值：0.06904660\n",
      "迭代次数：27600,  loss值：0.06901200\n",
      "迭代次数：27700,  loss值：0.06897772\n",
      "迭代次数：27800,  loss值：0.06894370\n",
      "迭代次数：27900,  loss值：0.06891000\n",
      "迭代次数：28000,  loss值：0.06887662\n",
      "迭代次数：28100,  loss值：0.06884357\n",
      "迭代次数：28200,  loss值：0.06881081\n",
      "迭代次数：28300,  loss值：0.06877832\n",
      "迭代次数：28400,  loss值：0.06874618\n",
      "迭代次数：28500,  loss值：0.06871432\n",
      "迭代次数：28600,  loss值：0.06868274\n",
      "迭代次数：28700,  loss值：0.06865146\n",
      "迭代次数：28800,  loss值：0.06862050\n",
      "迭代次数：28900,  loss值：0.06858998\n",
      "迭代次数：29000,  loss值：0.06855948\n",
      "迭代次数：29100,  loss值：0.06852943\n",
      "迭代次数：29200,  loss值：0.06849965\n",
      "迭代次数：29300,  loss值：0.06847019\n",
      "迭代次数：29400,  loss值：0.06844106\n",
      "迭代次数：29500,  loss值：0.06841215\n",
      "迭代次数：29600,  loss值：0.06838357\n",
      "迭代次数：29700,  loss值：0.06835531\n",
      "迭代次数：29800,  loss值：0.06832732\n",
      "迭代次数：29900,  loss值：0.06829963\n",
      "迭代次数：30000,  loss值：0.06827224\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1 导入需要的模块、包等\n",
    "import torch\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import torch.nn as nn\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "\n",
    "# 2 数据处理\n",
    "# 2.1  加载数据\n",
    "train_data = pd.read_csv(\"./kg_train.csv\")\n",
    "test_data = pd.read_csv(\"./kg_test.csv\")\n",
    "\n",
    "# print(train_data.shape)\n",
    "# print(test_data.shape)\n",
    "#\n",
    "# print(train_data.head())\n",
    "\n",
    "# 2.2 数据拼接\n",
    "all_features = pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))\n",
    "\n",
    "# 2.3 数据预处理:数值特征标准化处理，缺失值用0填充\n",
    "numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index\n",
    "all_features[numeric_features] = all_features[numeric_features].apply(lambda x: (x - x.mean()) / (x.std()))\n",
    "all_features = all_features.fillna(0)\n",
    "\n",
    "train_labels_mean = np.mean(train_data.SalePrice.values)\n",
    "train_labels_std = np.std(train_data.SalePrice.values)\n",
    "\n",
    "train_labels = (train_data.SalePrice.values - train_labels_mean) / train_labels_std\n",
    "\n",
    "print(train_labels)\n",
    "# 2.4 数据预处理:非数值特征one-hot编码\n",
    "all_features = pd.get_dummies(all_features, dummy_na=True)\n",
    "\n",
    "# 2.5 转为numpy的数据类型，再转为tensor\n",
    "n_train = train_data.shape[0]\n",
    "train_features = torch.tensor(all_features[:n_train].values, dtype=torch.float)\n",
    "train_labels = torch.tensor(train_labels, dtype=torch.float).unsqueeze(1)\n",
    "\n",
    "test_features = torch.tensor(all_features[n_train:].values, dtype=torch.float)\n",
    "print(train_features.shape[1])\n",
    "\n",
    "# 3 创建模型\n",
    "class PolyLinearRegression(nn.Module):\n",
    "    def __init__(self, input_feature_numbers, out_labels_numbers):\n",
    "        super(PolyLinearRegression, self).__init__()\n",
    "        self.ploy = nn.Linear(input_feature_numbers, out_labels_numbers)\n",
    "\n",
    "    def forward(self, x):\n",
    "        out = self.ploy(x)\n",
    "        return out\n",
    "\n",
    "# 4 模型训练与测试\n",
    "def train_and_test():\n",
    "    if torch.cuda.is_available():\n",
    "        device = torch.device(\"cuda:0\")\n",
    "        model = PolyLinearRegression(train_features.shape[1], train_labels.shape[1])\n",
    "        model.to(device)\n",
    "    else:\n",
    "        device = torch.device(\"cpu\")\n",
    "        model = PolyLinearRegression(train_features.shape[1], train_labels.shape[1])\n",
    "        model.to(device)\n",
    "\n",
    "    criterion = nn.MSELoss()\n",
    "    optimizer = torch.optim.Adam(model.parameters(), lr=0.0001)\n",
    "\n",
    "    loss_list = []\n",
    "    for i in range(30000):\n",
    "        i += 1\n",
    "\n",
    "        output_y = model(train_features)\n",
    "        # 计算损失\n",
    "        loss = criterion(output_y, train_labels)\n",
    "\n",
    "        # 2 反向传播\n",
    "        # 梯度要清0\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        # 更新权重参数\n",
    "        optimizer.step()\n",
    "\n",
    "        # 打印\n",
    "        if i % 100 == 0:\n",
    "            print(\"迭代次数：%d,  loss值：%.8f\" % (i, loss.item()))\n",
    "            loss_list.append(loss.item())\n",
    "\n",
    "    # 测试模式\n",
    "    model.eval()\n",
    "    predict = model(train_features).data.numpy()\n",
    "\n",
    "    plt.plot(predict, \"ro\")\n",
    "    plt.plot(train_labels, \"k\")\n",
    "    plt.show()\n",
    "\n",
    "    plt.plot(loss_list)\n",
    "    plt.show()\n",
    "    plt.show()\n",
    "\n",
    "    predict_price = predict * train_labels_std + train_labels_mean\n",
    "    plt.plot(predict_price[0:100], \"r\")\n",
    "    plt.plot(train_data.SalePrice.values[0:100], \"k\")\n",
    "    plt.show()\n",
    "\n",
    "    # 假设训练集和测试集同分布\n",
    "    test_result = model(test_features).detach().numpy() * train_labels_std + train_labels_mean\n",
    "\n",
    "    plt.plot(test_result, \"k\")\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "train_and_test()"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}