TF2.0 Linear Regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2020-04-30 10:26:27--  https://raw.githubusercontent.com/lazyprogrammer/machine_learning_examples/master/tf2.0/moore.csv\n",
      "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.228.133\n",
      "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.228.133|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 2302 (2.2K) [text/plain]\n",
      "Saving to: ‘moore.csv’\n",
      "\n",
      "moore.csv           100%[===================>]   2.25K  --.-KB/s    in 0s      \n",
      "\n",
      "2020-04-30 10:26:28 (18.7 MB/s) - ‘moore.csv’ saved [2302/2302]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!wget https://raw.githubusercontent.com/lazyprogrammer/machine_learning_examples/master/tf2.0/moore.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('moore.csv', header=None).values\n",
    "X = data[:,0].reshape(-1,1)\n",
    "Y = data[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f6eeb2d07f0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.scatter(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f6eeb18cc88>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "Y = np.log(Y)\n",
    "plt.scatter(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = X - X.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 162 samples\n",
      "Epoch 1/200\n",
      "162/162 [==============================] - 0s 735us/sample - loss: 382.2536\n",
      "Epoch 2/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 285.3438\n",
      "Epoch 3/200\n",
      "162/162 [==============================] - 0s 32us/sample - loss: 271.9848\n",
      "Epoch 4/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 196.5573\n",
      "Epoch 5/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 175.7914\n",
      "Epoch 6/200\n",
      "162/162 [==============================] - 0s 56us/sample - loss: 184.1604\n",
      "Epoch 7/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 164.2979\n",
      "Epoch 8/200\n",
      "162/162 [==============================] - 0s 63us/sample - loss: 124.2249\n",
      "Epoch 9/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 70.4681\n",
      "Epoch 10/200\n",
      "162/162 [==============================] - 0s 62us/sample - loss: 52.6681\n",
      "Epoch 11/200\n",
      "162/162 [==============================] - 0s 56us/sample - loss: 47.4714\n",
      "Epoch 12/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 37.9664\n",
      "Epoch 13/200\n",
      "162/162 [==============================] - 0s 58us/sample - loss: 38.3607\n",
      "Epoch 14/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 24.0965\n",
      "Epoch 15/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 9.0055\n",
      "Epoch 16/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 6.2119\n",
      "Epoch 17/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 4.5819\n",
      "Epoch 18/200\n",
      "162/162 [==============================] - 0s 59us/sample - loss: 5.1217\n",
      "Epoch 19/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 5.4940\n",
      "Epoch 20/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 3.2075\n",
      "Epoch 21/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 2.5253\n",
      "Epoch 22/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 1.7719\n",
      "Epoch 23/200\n",
      "162/162 [==============================] - 0s 74us/sample - loss: 2.4604\n",
      "Epoch 24/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 2.0039\n",
      "Epoch 25/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 1.7247\n",
      "Epoch 26/200\n",
      "162/162 [==============================] - 0s 62us/sample - loss: 1.1422\n",
      "Epoch 27/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 1.1201\n",
      "Epoch 28/200\n",
      "162/162 [==============================] - 0s 64us/sample - loss: 1.0662\n",
      "Epoch 29/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9598\n",
      "Epoch 30/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9616\n",
      "Epoch 31/200\n",
      "162/162 [==============================] - 0s 62us/sample - loss: 0.9200\n",
      "Epoch 32/200\n",
      "162/162 [==============================] - 0s 38us/sample - loss: 0.9287\n",
      "Epoch 33/200\n",
      "162/162 [==============================] - 0s 60us/sample - loss: 0.9281\n",
      "Epoch 34/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9194\n",
      "Epoch 35/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 0.9003\n",
      "Epoch 36/200\n",
      "162/162 [==============================] - 0s 63us/sample - loss: 0.9165\n",
      "Epoch 37/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.8911\n",
      "Epoch 38/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.9650\n",
      "Epoch 39/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.9088\n",
      "Epoch 40/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.8870\n",
      "Epoch 41/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.8893\n",
      "Epoch 42/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 0.8841\n",
      "Epoch 43/200\n",
      "162/162 [==============================] - 0s 69us/sample - loss: 1.0029\n",
      "Epoch 44/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9057\n",
      "Epoch 45/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 1.5161\n",
      "Epoch 46/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 1.3543\n",
      "Epoch 47/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 1.1282\n",
      "Epoch 48/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 1.2852\n",
      "Epoch 49/200\n",
      "162/162 [==============================] - 0s 56us/sample - loss: 1.1557\n",
      "Epoch 50/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 1.0212\n",
      "Epoch 51/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.8945\n",
      "Epoch 52/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.8896\n",
      "Epoch 53/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9859\n",
      "Epoch 54/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.8857\n",
      "Epoch 55/200\n",
      "162/162 [==============================] - 0s 66us/sample - loss: 0.9055\n",
      "Epoch 56/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9751\n",
      "Epoch 57/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9398\n",
      "Epoch 58/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 0.9595\n",
      "Epoch 59/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 0.8977\n",
      "Epoch 60/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 1.1008\n",
      "Epoch 61/200\n",
      "162/162 [==============================] - 0s 66us/sample - loss: 1.1379\n",
      "Epoch 62/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9301\n",
      "Epoch 63/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 1.4791\n",
      "Epoch 64/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 1.2809\n",
      "Epoch 65/200\n",
      "162/162 [==============================] - 0s 56us/sample - loss: 0.9639\n",
      "Epoch 66/200\n",
      "162/162 [==============================] - 0s 58us/sample - loss: 0.9344\n",
      "Epoch 67/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9911\n",
      "Epoch 68/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.9345\n",
      "Epoch 69/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 0.9614\n",
      "Epoch 70/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.9447\n",
      "Epoch 71/200\n",
      "162/162 [==============================] - 0s 67us/sample - loss: 0.8933\n",
      "Epoch 72/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.8797\n",
      "Epoch 73/200\n",
      "162/162 [==============================] - 0s 56us/sample - loss: 0.9313\n",
      "Epoch 74/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.9015\n",
      "Epoch 75/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.9062\n",
      "Epoch 76/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.9497\n",
      "Epoch 77/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9172\n",
      "Epoch 78/200\n",
      "162/162 [==============================] - 0s 66us/sample - loss: 0.9477\n",
      "Epoch 79/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.8778\n",
      "Epoch 80/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9046\n",
      "Epoch 81/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9294\n",
      "Epoch 82/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.8855\n",
      "Epoch 83/200\n",
      "162/162 [==============================] - 0s 59us/sample - loss: 0.8943\n",
      "Epoch 84/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9248\n",
      "Epoch 85/200\n",
      "162/162 [==============================] - 0s 66us/sample - loss: 0.9146\n",
      "Epoch 86/200\n",
      "162/162 [==============================] - 0s 63us/sample - loss: 0.9082\n",
      "Epoch 87/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9078\n",
      "Epoch 88/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 1.0751\n",
      "Epoch 89/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 0.9557\n",
      "Epoch 90/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.9230\n",
      "Epoch 91/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9515\n",
      "Epoch 92/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 0.9890\n",
      "Epoch 93/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.9055\n",
      "Epoch 94/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.9412\n",
      "Epoch 95/200\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9155\n",
      "Epoch 96/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9693\n",
      "Epoch 97/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.8924\n",
      "Epoch 98/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 0.8905\n",
      "Epoch 99/200\n",
      "162/162 [==============================] - 0s 60us/sample - loss: 0.9067\n",
      "Epoch 100/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.8877\n",
      "Epoch 101/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.9061\n",
      "Epoch 102/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.8949\n",
      "Epoch 103/200\n",
      "162/162 [==============================] - 0s 40us/sample - loss: 0.8747\n",
      "Epoch 104/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9131\n",
      "Epoch 105/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9478\n",
      "Epoch 106/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9064\n",
      "Epoch 107/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 0.8747\n",
      "Epoch 108/200\n",
      "162/162 [==============================] - 0s 60us/sample - loss: 0.9901\n",
      "Epoch 109/200\n",
      "162/162 [==============================] - 0s 37us/sample - loss: 0.9424\n",
      "Epoch 110/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.8912\n",
      "Epoch 111/200\n",
      "162/162 [==============================] - 0s 38us/sample - loss: 0.8809\n",
      "Epoch 112/200\n",
      "162/162 [==============================] - 0s 59us/sample - loss: 0.9980\n",
      "Epoch 113/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 1.2149\n",
      "Epoch 114/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 1.1149\n",
      "Epoch 115/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 1.0116\n",
      "Epoch 116/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.8897\n",
      "Epoch 117/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 1.0080\n",
      "Epoch 118/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9089\n",
      "Epoch 119/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9101\n",
      "Epoch 120/200\n",
      "162/162 [==============================] - 0s 38us/sample - loss: 0.9087\n",
      "Epoch 121/200\n",
      "162/162 [==============================] - 0s 57us/sample - loss: 0.9471\n",
      "Epoch 122/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9738\n",
      "Epoch 123/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9623\n",
      "Epoch 124/200\n",
      "162/162 [==============================] - 0s 38us/sample - loss: 0.9215\n",
      "Epoch 125/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.8770\n",
      "Epoch 126/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 0.8922\n",
      "Epoch 127/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.8980\n",
      "Epoch 128/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 0.9012\n",
      "Epoch 129/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.8879\n",
      "Epoch 130/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.9999\n",
      "Epoch 131/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 1.0430\n",
      "Epoch 132/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.9503\n",
      "Epoch 133/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.9337\n",
      "Epoch 134/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9052\n",
      "Epoch 135/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 0.9067\n",
      "Epoch 136/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.9688\n",
      "Epoch 137/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9070\n",
      "Epoch 138/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9854\n",
      "Epoch 139/200\n",
      "162/162 [==============================] - 0s 40us/sample - loss: 0.9253\n",
      "Epoch 140/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.8825\n",
      "Epoch 141/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 0.8769\n",
      "Epoch 142/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.8764\n",
      "Epoch 143/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.8779\n",
      "Epoch 144/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9098\n",
      "Epoch 145/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.9003\n",
      "Epoch 146/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.8801\n",
      "Epoch 147/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.8792\n",
      "Epoch 148/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9450\n",
      "Epoch 149/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 1.0508\n",
      "Epoch 150/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 1.0013\n",
      "Epoch 151/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9857\n",
      "Epoch 152/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9423\n",
      "Epoch 153/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9690\n",
      "Epoch 154/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9228\n",
      "Epoch 155/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.8721\n",
      "Epoch 156/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.8890\n",
      "Epoch 157/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 1.0460\n",
      "Epoch 158/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 1.3519\n",
      "Epoch 159/200\n",
      "162/162 [==============================] - 0s 52us/sample - loss: 1.2064\n",
      "Epoch 160/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 1.1454\n",
      "Epoch 161/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 1.0501\n",
      "Epoch 162/200\n",
      "162/162 [==============================] - 0s 42us/sample - loss: 0.9531\n",
      "Epoch 163/200\n",
      "162/162 [==============================] - 0s 53us/sample - loss: 0.9112\n",
      "Epoch 164/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 1.0058\n",
      "Epoch 165/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.9253\n",
      "Epoch 166/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.9054\n",
      "Epoch 167/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.8846\n",
      "Epoch 168/200\n",
      "162/162 [==============================] - 0s 50us/sample - loss: 0.9597\n",
      "Epoch 169/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 1.1225\n",
      "Epoch 170/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 1.2042\n",
      "Epoch 171/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 1.2411\n",
      "Epoch 172/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 1.0760\n",
      "Epoch 173/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 1.0733\n",
      "Epoch 174/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 1.1827\n",
      "Epoch 175/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9520\n",
      "Epoch 176/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9871\n",
      "Epoch 177/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 1.4557\n",
      "Epoch 178/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 1.1890\n",
      "Epoch 179/200\n",
      "162/162 [==============================] - 0s 55us/sample - loss: 0.9111\n",
      "Epoch 180/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 1.7189\n",
      "Epoch 181/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 1.3600\n",
      "Epoch 182/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 1.2683\n",
      "Epoch 183/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 1.6606\n",
      "Epoch 184/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 1.3488\n",
      "Epoch 185/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 1.0641\n",
      "Epoch 186/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 1.0181\n",
      "Epoch 187/200\n",
      "162/162 [==============================] - 0s 44us/sample - loss: 0.8955\n",
      "Epoch 188/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.8732\n",
      "Epoch 189/200\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "162/162 [==============================] - 0s 43us/sample - loss: 0.9207\n",
      "Epoch 190/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 0.8884\n",
      "Epoch 191/200\n",
      "162/162 [==============================] - 0s 54us/sample - loss: 0.9499\n",
      "Epoch 192/200\n",
      "162/162 [==============================] - 0s 46us/sample - loss: 0.9150\n",
      "Epoch 193/200\n",
      "162/162 [==============================] - 0s 43us/sample - loss: 0.9227\n",
      "Epoch 194/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.9252\n",
      "Epoch 195/200\n",
      "162/162 [==============================] - 0s 48us/sample - loss: 0.8824\n",
      "Epoch 196/200\n",
      "162/162 [==============================] - 0s 45us/sample - loss: 0.9004\n",
      "Epoch 197/200\n",
      "162/162 [==============================] - 0s 47us/sample - loss: 0.9109\n",
      "Epoch 198/200\n",
      "162/162 [==============================] - 0s 49us/sample - loss: 0.9040\n",
      "Epoch 199/200\n",
      "162/162 [==============================] - 0s 51us/sample - loss: 0.9445\n",
      "Epoch 200/200\n",
      "162/162 [==============================] - 0s 41us/sample - loss: 0.9336\n"
     ]
    }
   ],
   "source": [
    "model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Input(shape=(1,)),\n",
    "    tf.keras.layers.Dense(1)\n",
    "])\n",
    "\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(0.001, 0.9), loss='mse')\n",
    "\n",
    "def schedule(epoch, lr):\n",
    "    if epoch >= 50:\n",
    "        lr = 0.00001\n",
    "    return 0.001\n",
    "\n",
    "\n",
    "scheduler = tf.keras.callbacks.LearningRateScheduler(schedule)\n",
    "\n",
    "r = model.fit(X, Y, epochs=200, callbacks=[scheduler])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f6e984200b8>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.plot(r.history['loss'],label='loss')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f6dfeec7dd8>]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deZhU1bX38e/qAWgQaBBEaEAcEJBJpOMQNBEnEI0ixoGbqBm8xNx4zWCIIN6o0UQSosY3emO80RsTDXEChDggDtcpgmkEBAQcUIQGAYUGlAZ6WO8fXQ09nFPdVV3dVV31+zwPD1X7nFO1u+BZtXuftdc2d0dERNJXVrI7ICIizUuBXkQkzSnQi4ikOQV6EZE0p0AvIpLmcpLdgSDdunXzfv36JbsbIiKtxuLFiz919+5Bx1Iy0Pfr14+ioqJkd0NEpNUws3VhxzR1IyKS5hToRUTSnAK9iEiaU6AXEUlzCvQiImkuJbNuREQyyZwlxcyYv4aNJaX0ys9j8pgBjB9RkLDXV6AXEUmiOUuKmTprOaVlFQAUl5QyddZygIQFe03diIgk0Yz5a/YH+WqlZRXMmL8mYe+hEb2ISAsJmqLZWFIaeG5YezwU6EVEWsCcJcVMfnwZZRVVmz0Vl5Qy+fFl5LfPZfvusnrn98rPS9h7a+pGRKQF3Dxv5f4gX62swtlbVkFebnat9rzcbCaPGZCw91agFxFpAUGjdoDdZZXcNmEoBfl5GFCQn8dtE4Yq60ZEJJ2MH1HA+cf24vO95XRsl5vw19eIXkSkBeTnBQfw/LxciktKufLBIi5/4E0qKz3wvKZQoBcRaQE3nTeY3Cyr1ZZjMHrgIZx5x8v884PPGDekJ4kP82DuzfGyTVNYWOiqRy8iqSye1aw1r+l2UFva5Waxfnspg3p2ZNvn+9iya2/cK2PNbLG7FwYd04heRCRG1atZi0tKcQ6sZp2zpLjBayvdcWDr53vZvruMy086jPc+2cXmXXv3v9bkx5Y16rUaS4FeRCRG8axmnbOkmJ89/jabduzZ31ZeUcnjResprzOxUlbp3DR3ZcL6q6wbEZEYRVvNGjSlc3L/bkydtZx9FZW1zt9TXhn4OgAlpcHpmPFQoBcRiVGv/DyKA4J9fvvceqtfr310KW1zs+v9BtCSNHUjIhKj0QO7B7Z/sbe83urXCod9UUbuLaHBQG9mfczsJTNbZWYrzeyHkfYZZrbazN42s9lmlh9y/UdmttzMlpqZUmlEpNV7afXWwPZ9FcFZjOXNkBsfi8ZM3ZQD17r7W2bWEVhsZguABcBUdy83s18DU4HrQl5jtLt/mpgui4i0jLAUykRWlmwJDQZ6d98EbIo83mVmq4ACd3+uxmkLga83TxdFRJouWt570DEgdEOQsDn6MPl5uQm9uRqrmObozawfMAJYVOfQd4BnQi5z4DkzW2xmk6K89iQzKzKzoq1bg38tEhGJR7S897BjN89bGZpCOXnMgHoVJwEG9exITu3Fr+RmGTedN7j5frhGaHSgN7ODgCeAH7n7zhrt06ia3nk45NJR7n4ccDbwAzP7StBJ7n6fuxe6e2H37sE3OkRE4hEt7z3sWFi1yY0lpYwfUcDPxgygXU5VCM3NNq4982ie+eFX+O3Fx9aqRDnjouEJrUQZj0alV5pZLlVB/mF3n1Wj/QrgXOB0D6ml4O4bI39vMbPZwPHAK03tuIhIY0XLe4/1NmnnvBzuf+1Dfv3sasoi2TTdD2pLn67tm9jL2goSuPFIg4HezAy4H1jl7nfUaB9L1c3Xr7r77pBrOwBZkbn9DsBZwC8S0nMRkUYKm1PvlZ/HJzv2UBFDza9deyq45R/vkGXs/5LYuGMPU2ctp2jdNp5YXBw4rx9NXp08+2RsPDIKuAw4LZIiudTMxgF3Ax2BBZG2ewHMrJeZPR25tgfwmpktA94EnnL3ZxPWexGRRgiaU68OprEEeYAKd7q0z6VuxmRpWQUzF62Pa6PvpG884u6vARZw6OmAtuqpmnGRx2uB4U3poIhIU40fUcBjRR/z+gfb9rcd17cz40cUMGP+msDRfrZZ4JdAz07t+GTnnnrtQOiXRkMZOuNHFDTrPL5WxopI2rthzvJaQR7g9Q+2ccOc5aGj/eP6dg58rdOPOYT89rHvAvXNE/vG1J5IqnUjImklKCf+4UUfB5778KKPuXX8UID91/Ts3I6vHN2dR4rWB17z0uqtxLONR/X7zFy0ngp3ss2YeEKf/e3NSRuPiEirE7b4ac6S4lpFxaAq9bFu/ZmaPpp+zv7HH2z9nOtnLWfRh9tCz4equexYI2fN92kO0TYe0YheRFqV6gVOQZktN89bWS+oRwvy1faWV3Dv/63lnpfep11uFr++cChTn1hOUCmybDMO7dwuppWxebnJnSXXHL2IJNWcJcWMmv4ih095ilHTX2xwZ6Voi5/CFjmFaZNtvPnhNsbd9Sp3Pv8uY4YcyvPXfpVLvtQ3MMhD1Q3XyWMGkF1n/9fsLOObJ/atF1SzgNsmDIupX4mmQC8iSRPPlnxhI+lYRtgHGBf/8Q22fbGPgzu04R/LNnLBPf9kzpJiuoTccO3SPpeidduoqJNfWf38jktqr4y945Jjk74yVnP0IpIwsW6YPWr6i4EBuiA/j9ennBZ4zZFTnw5MY8w2278fayxGD+jOGx98Vmu3p7zcbAxnd1n9cX1+Xi4795TVy6MHyDJYe1vzzsWH0Ry9iDS7aHPnYcE+WmmCMGG56rEufIKq0gXvbv683pZ+0XaD2lFaFvplkuSy86E0dSMiCRHPhtm9Quq5hLVDeA2Ygvy80GOd2+VQZ0qdnCyYds6gmGvLR+tbqlKgF5GEiGd0Hq00QZiwbfxGD+we+Hptc7LIzs6qN9qufh4WuNuHZMqMHtg9NIsm2dk1YVKzVyLS6sQzOh8/oiDmOi9h2/i9tHor40cUcOHIArKtavhuVO3Xuv2LffXOr/SqdMywL5s2OfXrzVe/z20ThqVkdk0YzdGLSEJMHjOg1hw9NK4KY6x1XqJl3cxZUswTi4v3z9c7VWmPdTNkqm3fXbb/veveRP7xI0sDr6muRx90TbKza8Io0ItIQqRC8Jv+zOp69wnCgnxDopU2huYvRJZICvQikjCJDn5he7mGCasq2dB7BGULXTiyoFZteUh8nfiWojx6EWkRsebY1w3AUH+DjrraZGexr6J+7ntYbZr8vFw6tM0JzeWfPGZAq5meUR69iCRVtBz7onXbAis6hqVrRjP9wqFMm72i3pfDhSMLeOTN9ZTVmMap3rS7obn4VA3ssVCgF5FmFxa0p81ezhf7DrRXuPPQwqqSwrGWNCjIz2PCcb3JMgschRce1jWwPWzjkdaYLx9GgV5Eml1YLn3NIF/TzEXrQ3d4ClJz7jxsFB7WHm+2UGuiQC8igWKdU48mLIMlTEMBfvSA7qz+ZBef7NjT5L6lQrZQc1OgF5F64qlbE83kMQNi2hAky6Bn5+Avh+4HteV/v318zH2IJl3m4sM0uDLWzPqY2UtmtsrMVprZDyPtXc1sgZm9F/m7S8j1V0TOec/Mrkj0DyAiiRdP3ZoG1Y3pDmEVA9rmZPGTM48mp06BmnY5WUw7Z1D8fchQjRnRlwPXuvtbZtYRWGxmC4BvAS+4+3QzmwJMAa6reaGZdQVuBAqp+mdebGZz3X17In8IEUmseOrWRDNj/ppaGS9Avec1lZZV8r///JDySqdtThZ7yyv3pzvGO/JO5FRUa9NgoHf3TcCmyONdZrYKKADOB06NnPYg8H/UCfTAGGCBu28DiHxBjAVmJqDvIpIAQQGwoVWhsYpnU5DNO/dyz78dx7ihh2JmDV8QRaKnolqbmObozawfMAJYBPSIfAng7pvM7JCASwqAmlupb4i0Bb32JGASQN++fWPplojEKSwAHte3c2BwDqscWfP1gkbN0TbTDloE9eUjD+YP3xxJ57zgXZ5iFW0qKhMCfaOrV5rZQcATwI/cfWdjLwtoC/z3dvf73L3Q3Qu7d4/+n0lEEiMsAC5cGzy7GlY5EqJvCxgth2bq2QP3V4/MyTKuOe0o/vbvJyYsyEPip6Jam0aN6M0sl6og/7C7z4o0bzaznpHRfE9gS8ClGzgwvQPQm6opHhFJAWGBLiy9MVpgjPcG7m+fW0NFpfOTM4/me189grYh5YGbItFTUa1NY7JuDLgfWOXud9Q4NBeozqK5Angy4PL5wFlm1iWSlXNWpE1EUkBYoMsOmROPFhijjZrDNtoGGNSzE8/86BSuOb1/swR5iG+Dk3TSmKmbUcBlwGlmtjTyZxwwHTjTzN4Dzow8x8wKzexPAJGbsLcA/4r8+UX1jVkRSb6wADjxhD4xB8awqZbOebnc+LXB5GbX//K49Et9+PukEzmy+0Fx9L7x4tngJJ00JuvmNYLn2gFODzi/CLiyxvMHgAfi7aCINJ9oq0LDasNA8E3XsMQYs6r3+WDL5/zh5Q8or3TycrO5ftxALjupXwv9pKmxKCpZKZ4qUyySIRIVZOYsKY5plStERu7/Wk/vLnncOn4Ipw44kKSXKfntYWWXE/WbhcoUi2S4ROaR3zxvZb2gXlbhZBn1NuCGqnIGjy3ewPe+cgQ/PKM/7dscCDuZlN+ezBRPBXqRNBI2Ok5kkNm+uyywvdKDR/YF+Xnce9lIBvfqXO+aTMpvT2aKpwK9SJqINjpuqSDjdYb0WQY/PuNoBvfqHPgllEn57clM8Wz0gikRSW3RRsdhwSSeIBNWiAygvM7UTaXD7QveDV1MlR+SdpmO+e3JTPFUoBdJE9FGx4kMMnWDeWP6FfYl5E7G5LcnM8VTUzciaSLa1EAiN9eINVGvV35e6JfQjtIy7rzk2IzIuoHkpXgq0IukiYa2xEtUkAnb4s8M2uVkB75/tH1ZUyG/Pd0p0IukiXhH7WGZOkHt5w3vRWG/Liz6sP4C92+c0DfqIqt035c1lWnBlEgGC1vEc+HIAp5YXFyrvW1OFgX5eaz99At6dGrL1p17qaRqhD/xhD7cOn5og++VKVM0yRBtwZQCvUgGGzX9xcAplWjTM7++cBgXjezd5M1AJLG0MlZEAsVaptgdLi7s05xdkmag9EqRDBaWrx4WGArSML89EyjQi2SwoPz63GyjTcCqKN08bb0U6EUyWM1FPFB1w7WswhnQoyM/PevojK3fnm40Ry+S4c4d1pMtu/Zw54L3yDKY8rVjuPykfmRnGVef1j/Z3ZMEUKAXyRBB6Y1HdO/Afzz0FhsiN2V7dGpLl/ZtyM5SRk06UaAXyQBBlS2vfWwZlZVOzfyazTv3pm09+EymQC+SRmKpR19R6YF7hLZ0PXgtpGp+DQZ6M3sAOBfY4u5DIm2PANW33/OBEnc/NuDaj4BdQAVQHpbMLyKxCQqOQMz16MOWS7ZUPfhM2mEqmRozov8zcDfwl+oGd7+k+rGZ3Q7siHL9aHf/NN4OikhtYcGxXW5WYCng3zy7ms55uZSU1t8ZKmwFbEvVg8+kHaaSqcH0Snd/BahfwQiwqjXQFwMzE9wvEQkRFhzDtvjbuGMPJaVl1L2/mpebzcQT+iS1Hnwm7TCVTE3Noz8F2Ozu74Ucd+A5M1tsZpOa+F4iQuxB0AxmfH0Yt180vF5e/K3jhyZtMwwI/80hHXeYSqam3oydSPTR/Ch332hmhwALzGx15DeEeiJfBJMA+vbt28RuiaSvsA1G8vNy2VteWWu0n23GzecN5qJIfZoLjutd77pk1oNvqIa+JEbcI3ozywEmAI+EnePuGyN/bwFmA8dHOfc+dy9098Lu3bvH2y2RtDFnSTGjpr/I4VOeYtT0F5mzpBiIvvfosN6d97cd3KENt188nG+edFiL9jsWydxeL5M0ZUR/BrDa3TcEHTSzDkCWu++KPD4L+EUT3k8kYzQmG6U666Zn53acPqgHdz7/Ltt3l3HVV4/kh6f3J69NdujrJ0NYGqV2mGp+DY7ozWwm8AYwwMw2mNl3I4cupc60jZn1MrOnI097AK+Z2TLgTeApd382cV0XSV/RslGgKti/PuU0XvnZaI7q0ZG/LlxHQZf2zLv6ZKacPTAlg/zUWcspLinFOfDFVf1bijSvBkf07j4xpP1bAW0bgXGRx2uB4U3sn0hGCpqDr9l+/ey3mblo/f48+MJ+XXhk0kkpW7pAaZTJpeqVIikobPMmM7jqoSL+ViPIAxR9tJ0b565okb7FQ2mUyaVAL5KCwnb4dIdnV2wOPDZz0fpm7FHTKI0yuRToRdJE2PZ/qSBappA0PwV6kRSUn5cb2H5Qm2yyQ+Z1wtpTgdIok0vVK0VS0M+/dgzXPrqs1jy8AbdeMJSiddt4aOHH9a6ZeEJqb9qtNMrkUaAXSTHvbt7F7194v15lyZxIRs2t44cCVXPyFe5kmzHxhD7720XqMk/Beb3CwkIvKipKdjdEWtSesgrueel97n35A8orPfCGbEF+Hq9POa3lOycpz8wWh5WC1xy9SAp444PPOPuuV/n9i+/ztWG9QrNulI4o8dDUjUgLCFv+X7J7H796ehWPFm2gb9f2/PW7x3NK/+4s+nBb4KIppSNKPBToRZrZnCXFTH5sGWWVVcP04pJSfvroUhav287TyzdRUlrG9089kmtOO1CfZvTA7oE3XEcPVME/iZ0CvUgzu2nuyv1Bvlq5w18XrmN4n3wemjCUQT071Tr+0uqtga8V1i4SjQK9SDML2sKv2qzvf5l5yzZy5YNFtaZ1VDJAEkmBXiSJ5i3bGFiOOL99buDWgJqjl3go60akGX2+tzz0WJaFV3V0RyUDJGEU6EViFLbzU13Pv7OZM+94OfR1Kj18KmZHaZlKBkjCaOpGJAaN2flp88493DR3Jc+s+IQBPTpSVlHJp5/vq/daBZFpmLA0SpUMkETRiF4kBtE20KisdB5auI4zbn+ZF1ZvYfKYAfzjmpO54ZxjQqdhVNVRWoJG9CIxCJtqKS4p5aI/vsHiddtpm5PFvvJK/rboYwoiI3MgcMFUtWjHRJpKgV4kBr3y80K3+Vu9aSe52cbe8kqg/rROWPDWFI00N03diMQgaKoF4Ev9utCxXS5lFbUXRtXc0FskWRoM9Gb2gJltMbMVNdpuMrNiM1sa+TMu5NqxZrbGzN43symJ7LhIMowfUcAN5wyifaRUQXaW8f2vHsljV32ZzTv3BF6jRU6SbI2ZuvkzcDfwlzrtd7r7b8MuMrNs4B7gTGAD8C8zm+vu78TZV5GECys2FsTdeXLpRm5f8C77yiv5j1OP5JrT+9MuMsIPm9bRIidJtgZH9O7+CrAtjtc+Hnjf3de6+z7g78D5cbyOSLOYs6SYyY8vo7ikFKdqTn3y48sC8+I//mw3lz/wJj96ZCl9u7Zn3n+ezM/GDtwf5EH7okrqasrN2KvN7HKgCLjW3bfXOV4A1NyWfgNwQtiLmdkkYBJA3759m9Atkca5ed7KenPqZRXOzfNWMn5EAXOWFPObZ1ezccceDGiTk8Uvzh/MQW1y6tWmqXlDVRk0kmriDfR/AG4BPPL37cB36pwTtFNx6HZW7n4fcB9U7TAVZ79EGi2olkx1+5wlxVz3+NvsrajKoHGq/kO/u3kXTywuDl0wpQwaSUVxZd24+2Z3r3D3SuB/qJqmqWsDUHO34t7AxnjeT6SlTZu9fH+Qr7anvJKZi9aHLpgSSVVxBXoz61nj6QXAioDT/gX0N7PDzawNcCkwN573E2kO+Xm5ge1m8MW+isBjFSF7/CmzRlJZY9IrZwJvAAPMbIOZfRf4jZktN7O3gdHAjyPn9jKzpwHcvRy4GpgPrAIedfeVzfRziMTspvMGk5tVf4bx0E7t6H5Q28Brsi1oRlKZNZLaGpyjd/eJAc33h5y7ERhX4/nTwNNx906kGY0fUUClOzfPe4cdkc1Bzh3WkzsvOZan3t5Uq3gZVGXQXDiyoNYcfXW7MmsklakEgmSsNZ/s4uFFH7OjtIyTj+rGLy8YwmEHdwCImkFTeFhXZdZIq2IeMueYTIWFhV5UVJTsbkgr09jFT3vKKrj7xfe59+UP6Nguh/869xguGFGAhUzLiLQGZrbY3QuDjmlEL2mhMXXiAf75/qdcP3s5H322mwnHFXDDOcfQtUObpPRZpKUo0EtKijY6DzoWrU78+BEFbP9iH798ehWPL97AYQe35+ErT2DUUd2S8aOJtDgFekk50UbnQOCxukG+WnFJKbOXbOCWf6xiZ2kZPxh9JP95Wv9apQtE0p0CvaScaKPz6sd1j2WbBea4t83J4sePLGNE33xumzCUgYd2ar6Oi6QoBXpJOWEbexSXlAbW1YCqhUx5udmBI/tbzh9MhzY5fPfP9evTiGQCbTwiKSdsUVK2WejCpIL8PG6bMLTWQqdhvTvz8uTRdGyXy7Q5K2pVqZw6a3lglUqRdKRALyknrMxAhXtoKeD/PO0olq4v4dMv9nJop3b88bKRzL36ZA7t3K7BqSCRdKepG2l2sWzuAVWj86Dpm7CNtscOOZS7XniPT3bu4YqT+nHtWUfTsd2BOjZhdWhUn0YyhUb00qyqM2himTZpaAOP8SMKeH3Kabwx9XSGFnTm/tc+pHNeLrO+/2VuOm9wrSAP4XVoVJ9GMoUCvTSreKZNxo8o4MKRBfvn6rPNuHDkgTrvlZXOX9/4iDPueJmX1mzhurEDmfefJzOib5fA15s8ZgC52bXn/XOzrcH6NHOWFDNq+oscPuUpRk1/UXP60mpp6kaaVTzTJnOWFPPE4uL9c/UV7jyxuJjCw7oyqGcnps56m7c+LqlXnyaqutP+DVT+aOxKW5HWQIFemlXnvFxKSuvv5NQ5pBY8hP8WcMOcFewpq6BTXi53XjKc8cc2rj7NjPlrKKuss2Vgpe9fNRtLH6JdI5KqFOilWYXF4WjxOWy0//neci48rjfTzhkUU32aeH6r0A1cSSeao5dmVRKyL2tYO4TfJD24Qxtuv3h4zEXI4rkZqxu4kk4U6CVhgm5exhMwf3rW0fVunrbLyeK/zj0mrn41lMWTqGtEUpWmbiQhwm5eHte3c2BO/OiB3QNfZ91nX/DEW8WUVThtsrPYV1FJQRNLFkTbRCSR14ikKgV6SYiwm5cL124PPP+l1VtrPS+rqOR/Xl3LXc+/R5vsLL4+sjf/fP9TNu3Yk5D+jR9REHOQjucakVSkQC8JEXaTMqycQc3zl3y8namzlrP6k12MHXwoJx7RlV8/u0apjSIJ0mCgN7MHgHOBLe4+JNI2A/gasA/4APi2u5cEXPsRsAuoAMrDtrmS1iOsnEGvkLIFYeWDe+XnsWtPGb+dv4a/LFxHj47tuO+ykZw1+FBGTX9RqY0iCdSYm7F/BsbWaVsADHH3YcC7wNQo149292MV5Fu/aOUMwm5ennhE8GrVow7pwJl3vMJf3lhH+9xsPtm5h5vnvcOcJcVKbRRJsAYDvbu/Amyr0/acu5dHni4EejdD3yTFNLSI6LYJQynIz8M4UDb4o8+Cg/PL735KlkGb7Cy+2Fd7iia/ffBiKqU2isQnEXP03wEeCTnmwHNm5sAf3f2+sBcxs0nAJIC+ffsmoFuSaA2NtINuXv74kaVRX3NvRWWt56VlFbTNyaq3iYhSG0Xi16Q8ejObBpQDD4ecMsrdjwPOBn5gZl8Jey13v8/dC929sHv34NQ7aTmJyomPtlFIWEbNjtKywN8OND8vEp+4R/RmdgVVN2lPdw9OrXD3jZG/t5jZbOB44JV431NaRlhO/IUjC3hicXGjR9p7yioY1LNjvZu01dfMmL8m8AZur0jdeQV2kcSIa0RvZmOB64Dz3H13yDkdzKxj9WPgLGBFvB2VlhM2F//S6q2NHmm//v6njPndKzy/agvH9+vKoZ3a1bsmbNFUWLuIxKcx6ZUzgVOBbma2AbiRqiybtsCCSPXAhe5+lZn1Av7k7uOAHsDsyPEc4G/u/myz/BSSUNHm4hsaaW/7Yh+3PvUOs94qpt/B7fnblSfw5aO6BZ5bd9FUQ+0iEp8GA727Twxovj/k3I3AuMjjtcDwJvVOEiLWrfzCcuKjzcW7O7PeKubWp95h155yrh59FFefdhTt6qRc1qQ0SpGWoaJmaa45tvKr66NPv+Cb9y/i2seWcXi3Djx1zSn8dMyAqEEeVCFSpKUo0Ke5eLfya8xcfFlFJfe89D5jfvcKb6/fwUUje/PJjj2M/d0rjdp6TxUio9NWhpIoqnWT5uKdHmloLv6tj7dzfaQ+zdlDDuXEIw5m+jOrY6pPowqR4bSVoSSSAn2ai2e+PZpde8qYMX8Nf124jkM7teN/Li/kzGN6xF2fRmmUwbSVoSSSAn2amzxmQK2RIcQ/PfLsik+4ae5KNu/aw7e+3I9rzxrAQW2r/gvpxmpi6fOURFKgT3OJmB7ZtKOUG59cyXPvbGZQz07ce9lIju2TX+ucRP/mkOn0eUoiKdBngHinRyoqnYcWrmPG/DWUV1Yy9eyBfOfkw8nNrn8PP5G/OYg+T0ksBfoUFWvue6Kt2rSTqbOWs3R9Caf078Yvxw+l78HtQ8/XjdXE0ucpiWQhZWqSqrCw0IuKipLdjaSpm3EBVaO5lijstaesgrteeI8/vvwBAJUOvTq342djByrIiKQwM1sctu+HRvQpKNEZF4397eC19z5l2pzlrPtsd62doTbu2KPUPpFWTAumUlAiMy4aszL2s8/38pNHlvLN+xeRZUa3Dm3qbf/X0CIrEUldCvQpKJGlAaL9duDuPL54A2fc8TLz3t7INacdxTM/PIXPvtgX+FpK7RNpnRToU1AiSwOEBefiklK+8adF/PSxZRzR/SCeuuYUfnJWVX0a1aARSS8K9CmosbVmGiNacF6+YQe/vGAIj33vJI7u0XF/u2rQiKQX3YxNUYkqDRCUjw0wvHdn7ru8kB6d2gW+N7S+1L5kp6SKpCoF+jQ3fkQBpfsquPWpd/hiXwXZZnx7VD9uOPeYBq9rTUFSRcBEwinQp7lnV3zC7154l9KyCr49qnZ9mkRL5ohaRcBEwinQJ1G0wNjUoLlpRyk/f3IlCyL1ae67rJDhderTJFKyR9QqAiYSToE+SaIFRiDuoFlR6fz1jY/47XPvNlifJt5+B3nhNbgAAAv7SURBVH0BJXtErSJgIuEU6JOkoZ2f4gmaqzbtZMqs5SxbX8JXju7OL8cPoU/X8Po0sYr25ZTsEbWKgImEa1SgN7MHgHOBLe4+JNLWFXgE6Ad8BFzs7tsDrr0CuCHy9FZ3f7Dp3W794gmMYcdK91XVp/nTq2vpnJfLXZcey3nDe2FmDfYjlimiaF9OyR5Rt9ZMIZGW0NgR/Z+Bu4G/1GibArzg7tPNbErk+XU1L4p8GdwIFAIOLDazuUFfCJmmocDY2KD56ntbmTZ7BR9v283Fhb25ftwg8tu3aVQfYp1Xj/bldOclxyZ9RN3aMoVEWkqjJm7d/RVgW53m84Hq0fmDwPiAS8cAC9x9WyS4LwDGxtnXtBJtUdLkMQPIza49Gs/NtlpB87PP9/LjR5Zy2f1vkpNlzPz3E/nN14c3OshD7BuHR1sxm8hFXiKSWE2Zo+/h7psA3H2TmR0ScE4BsL7G8w2RtnrMbBIwCaBv375N6FbrEG2qYc6S4qrff2qKPHd3nniruCovfm8515x2FP8x+ija1fnSaIxYp48mjxnA5MeWUVZ5oHO5WQe+gDSiFklNzX0zNmiSOLAAvrvfB9wHVfXom7NTqSIsMM6Yv6ZWMAUoq3Rue3oVjxat558ffEbhYV24bcJQ+tcoXRCruObV6/6LNnwbQESSrCk5d5vNrCdA5O8tAedsAPrUeN4b2NiE98wIYSPqzbv2sry4qj7No987qUlBHmKvaTNj/hrKKup8AVW4yheLpLimBPq5wBWRx1cATwacMx84y8y6mFkX4KxIm0QRNqLOy83mhZ98lW+ccBhZWU0fSsc6r57sFEoRiU9j0ytnAqcC3cxsA1WZNNOBR83su8DHwEWRcwuBq9z9SnffZma3AP+KvNQv3L3uTV2pY/KYAUx54m32lFfub2uTncVtE4ZySEARsmrxrKaNZV492SmUIhKfRgV6d58Ycuj0gHOLgCtrPH8AeCCu3mUgd6ddbhZtcrL2B/qendtxXQN7trZECQItShJpnbQyNoVsLKmqT/P8qs0c07MT0y8cyrDejatP0xIlCLQoSaR1UqBPouqpluKSUjrn5bKnrIIsM64fN5DvjDqcnBjq07TU/LlSKEVaHwX6APHMdcd6Td2plh2lZWQZTBs3iO+eckTMfdb8uYiE0VaCdVQH4OKSUpwDc91zlhQn9JpfP7u63lRLpcMDr38UV7+1/Z+IhFGgryPWsgDxXPPKu1vZtGNP4LF4p1pUgkBEwmjqpo5EVpWs2/7p53u59R/vMGfpRnKyjPLK+guAmzLVovlzEQmiEX0d0Qp3hemclxu13d15tGg9Z9zxMk8t38Q1p/fntglDNdUiIi1CI/o64skVDyv7bgZrt37OtNkreGPtZ3ypXxd+dcGB+jS52VlKVRSRZqdAX0c8ueIlu8sC27fvLmPsXa/SNqdqVeslhX1qlS7QVIuItAQF+gTIb5/L9pBgf+YxPbjx3GOili4QEWlOCvR1zFlSzOTHl+2v0lhcUsrkx5cB4aUEPKSococ22dzzb8c1Sz9FRBpLN2PruHneysBSvDfPWxl6TUlp8Gh+976KwHYRkZaU9iP6WFeshk3BVLfXfb0rTzmcdjUKkNWkVakikgrSOtAnuqJj0OvdPO8dsrOsXl68UiVFJFWk9dRNPKtc80Ny4vPzcgNfD6Bbhzb89qLhWpUqIikprUf08axyvem8wYEbYN903mB+9MjSwGu27NqrVEkRSVlpHegbquh4w5zlzFy0ngp3ss2YeEIfbh0/FKifR9+lQxuys4yKBJctEBFpbmkT6INuukZb5XrDnOU8tPDj/e0V7vuf3zr+wLTLp5/v5ZZ/vMOTSzdySMe2lOwuY19FZb3XExFJVWkxRx9WJhgIreg4c9H6wNeqbq+uT3P67S/zzPJP+OHp/Xn1utH85uvDNBcvIq1K3CN6MxsAPFKj6Qjg5+7+uxrnnAo8CXwYaZrl7r+I9z3DRLvp+vqU0wIDcUXIKqcKd9Zu/ZzrZy9n4dptHN+vK7+aMISjDqmqT6O5eBFpbeIO9O6+BjgWwMyygWJgdsCpr7r7ufG+T2PEc9M12yww2Bsw9q5XaRdSn0ZEpLVJ1NTN6cAH7r4uQa8Xk3hKC088oU9guwNnHdOD56/9KhOP76sgLyKtXqIC/aXAzJBjJ5nZMjN7xswGh72AmU0ysyIzK9q6dWtMbx7PNnq3jh/KN0/sW+sD6NAmm//91pe4+9+O45COKkImIumhyYHezNoA5wGPBRx+CzjM3YcDvwfmhL2Ou9/n7oXuXti9e/eY+hDPNnruzpeP7MbBHduSZXDlyYfz5rQzGD3wkJjeW0Qk1SUivfJs4C1331z3gLvvrPH4aTP7bzPr5u6fJuB9a4nlJmlxSSk/n7OCF1ZvYUhBJx644ksM7d050V0SEUkJiQj0EwmZtjGzQ4HN7u5mdjxVv0F8loD3jEtFpfPgPz/it8+twR1uOGcQ3/pyP3Ky0yLLVEQkUJMCvZm1B84Evlej7SoAd78X+DrwfTMrB0qBS93Dqrc3rxXFO7h+9nLe3rCDUwd055bzh9Cna/tkdEVEpEU1KdC7+27g4Dpt99Z4fDdwd1Peo6l27yvnruff40+vfUiX9m34/cQRnDusJxa20auISJpJmxIIQV5+dyvTZi9nw/ZSJh7fhyljB9G5fXB1ShGRdJWWgb5mfZoju3fg0e+dxPGHd012t0REkiKtAr2781jRBn759CpK91XwozP68/1Tj6RtTnbDF4uIpKm0CfSl+yr49p/frKpPc3hXfnXBUI465KBkd0tEJOnSJtDntcmm38EduGBEAReNVH0aEZFqaRPoAaZfOCzZXRARSTlaKSQikuYU6EVE0pwCvYhImlOgFxFJcwr0IiJpToFeRCTNKdCLiKQ5BXoRkTRnSSoPH5WZbQWaa6PxbkDCd7hqRTL95wd9BqDPANLvMzjM3QP3YU3JQN+czKzI3QuT3Y9kyfSfH/QZgD4DyKzPQFM3IiJpToFeRCTNZWKgvy/ZHUiyTP/5QZ8B6DOADPoMMm6OXkQk02TiiF5EJKMo0IuIpLmMCPRmdouZvW1mS83sOTPrFWk3M/t/ZvZ+5Phxye5rczGzGWa2OvJzzjaz/BrHpkY+gzVmNiaZ/WxOZnaRma00s0ozK6xzLFM+g7GRn/F9M5uS7P60BDN7wMy2mNmKGm1dzWyBmb0X+btLMvvY3DIi0AMz3H2Yux8L/AP4eaT9bKB/5M8k4A9J6l9LWAAMcfdhwLvAVAAzOwa4FBgMjAX+28zSdTf1FcAE4JWajZnyGUR+pnuo+n9/DDAx8rOnuz9T9e9a0xTgBXfvD7wQeZ62MiLQu/vOGk87ANV3oM8H/uJVFgL5ZtazxTvYAtz9OXcvjzxdCPSOPD4f+Lu773X3D4H3geOT0cfm5u6r3H1NwKFM+QyOB95397Xuvg/4O1U/e1pz91eAbXWazwcejDx+EBjfop1qYRkR6AHM7Jdmth74BgdG9AXA+hqnbYi0pbvvAM9EHmfqZ1BTpnwGmfJzNkYPd98EEPn7kCT3p1mlzebgZvY8cGjAoWnu/qS7TwOmmdlU4GrgRsACzm+1+aYNfQaRc6YB5cDD1ZcFnJ/Wn0HQZQFtrfYziCJTfk6pI20Cvbuf0chT/wY8RVWg3wD0qXGsN7AxwV1rMQ19BmZ2BXAucLofWECRUZ9BiLT6DKLIlJ+zMTabWU933xSZrt2S7A41p4yYujGz/jWengesjjyeC1weyb45EdhR/etcujGzscB1wHnuvrvGobnApWbW1swOp+rG9JvJ6GMSZcpn8C+gv5kdbmZtqLoBPTfJfUqWucAVkcdXAGG/7aWFtBnRN2C6mQ0AKqkqf3xVpP1pYBxVN992A99OTvdaxN1AW2CBmQEsdPer3H2lmT0KvEPVlM4P3L0iif1sNmZ2AfB7oDvwlJktdfcxmfIZuHu5mV0NzAeygQfcfWWSu9XszGwmcCrQzcw2UPXb/HTgUTP7LvAxcFHyetj8VAJBRCTNZcTUjYhIJlOgFxFJcwr0IiJpToFeRCTNKdCLiKQ5BXoRkTSnQC8ikub+P/uXrVc69E3yAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Yhat = model.predict(X).flatten()\n",
    "plt.scatter(X,Y)\n",
    "plt.plot(X, Yhat)"
   ]
  },
  {
   "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
}