HousePricing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.layers import Input, Dense, Conv2D, Flatten, Activation, BatchNormalization, MaxPooling2D, Dropout\n",
    "from keras.models import Model, Sequential\n",
    "from keras.utils.np_utils import to_categorical\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.wrappers.scikit_learn import KerasRegressor\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder, StandardScaler\n",
    "from sklearn.model_selection import cross_val_score, KFold\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import accuracy_score\n",
    "get_ipython().run_line_magic('matplotlib', 'inline')\n",
    "from collections import defaultdict\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_columns', 500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>LotConfig</th>\n",
       "      <th>LandSlope</th>\n",
       "      <th>Neighborhood</th>\n",
       "      <th>Condition1</th>\n",
       "      <th>Condition2</th>\n",
       "      <th>BldgType</th>\n",
       "      <th>HouseStyle</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>RoofStyle</th>\n",
       "      <th>RoofMatl</th>\n",
       "      <th>Exterior1st</th>\n",
       "      <th>Exterior2nd</th>\n",
       "      <th>MasVnrType</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>ExterCond</th>\n",
       "      <th>Foundation</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>BsmtCond</th>\n",
       "      <th>BsmtExposure</th>\n",
       "      <th>BsmtFinType1</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinType2</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <th>Heating</th>\n",
       "      <th>HeatingQC</th>\n",
       "      <th>CentralAir</th>\n",
       "      <th>Electrical</th>\n",
       "      <th>1stFlrSF</th>\n",
       "      <th>2ndFlrSF</th>\n",
       "      <th>LowQualFinSF</th>\n",
       "      <th>GrLivArea</th>\n",
       "      <th>BsmtFullBath</th>\n",
       "      <th>BsmtHalfBath</th>\n",
       "      <th>FullBath</th>\n",
       "      <th>HalfBath</th>\n",
       "      <th>BedroomAbvGr</th>\n",
       "      <th>KitchenAbvGr</th>\n",
       "      <th>KitchenQual</th>\n",
       "      <th>TotRmsAbvGrd</th>\n",
       "      <th>Functional</th>\n",
       "      <th>Fireplaces</th>\n",
       "      <th>FireplaceQu</th>\n",
       "      <th>GarageType</th>\n",
       "      <th>GarageYrBlt</th>\n",
       "      <th>GarageFinish</th>\n",
       "      <th>GarageCars</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>GarageQual</th>\n",
       "      <th>GarageCond</th>\n",
       "      <th>PavedDrive</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8450</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>CollgCr</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2003</td>\n",
       "      <td>2003</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>BrkFace</td>\n",
       "      <td>196.0</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>GLQ</td>\n",
       "      <td>706</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>856</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>856</td>\n",
       "      <td>854</td>\n",
       "      <td>0</td>\n",
       "      <td>1710</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>8</td>\n",
       "      <td>Typ</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>2003.0</td>\n",
       "      <td>RFn</td>\n",
       "      <td>2</td>\n",
       "      <td>548</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>61</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>208500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>80.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>FR2</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>Veenker</td>\n",
       "      <td>Feedr</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>1Story</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>1976</td>\n",
       "      <td>1976</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>MetalSd</td>\n",
       "      <td>MetalSd</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>CBlock</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>Gd</td>\n",
       "      <td>ALQ</td>\n",
       "      <td>978</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0</td>\n",
       "      <td>284</td>\n",
       "      <td>1262</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>1262</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1262</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>6</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1976.0</td>\n",
       "      <td>RFn</td>\n",
       "      <td>2</td>\n",
       "      <td>460</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>298</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>181500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>68.0</td>\n",
       "      <td>11250</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>CollgCr</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2001</td>\n",
       "      <td>2002</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>BrkFace</td>\n",
       "      <td>162.0</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>Mn</td>\n",
       "      <td>GLQ</td>\n",
       "      <td>486</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0</td>\n",
       "      <td>434</td>\n",
       "      <td>920</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>920</td>\n",
       "      <td>866</td>\n",
       "      <td>0</td>\n",
       "      <td>1786</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>6</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>2001.0</td>\n",
       "      <td>RFn</td>\n",
       "      <td>2</td>\n",
       "      <td>608</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>223500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>70</td>\n",
       "      <td>RL</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Corner</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>Crawfor</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1915</td>\n",
       "      <td>1970</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>Wd Sdng</td>\n",
       "      <td>Wd Shng</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>BrkTil</td>\n",
       "      <td>TA</td>\n",
       "      <td>Gd</td>\n",
       "      <td>No</td>\n",
       "      <td>ALQ</td>\n",
       "      <td>216</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0</td>\n",
       "      <td>540</td>\n",
       "      <td>756</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Gd</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>961</td>\n",
       "      <td>756</td>\n",
       "      <td>0</td>\n",
       "      <td>1717</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>7</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>Detchd</td>\n",
       "      <td>1998.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>3</td>\n",
       "      <td>642</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>272</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2006</td>\n",
       "      <td>WD</td>\n",
       "      <td>Abnorml</td>\n",
       "      <td>140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>84.0</td>\n",
       "      <td>14260</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>FR2</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>NoRidge</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>2000</td>\n",
       "      <td>2000</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>BrkFace</td>\n",
       "      <td>350.0</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>Av</td>\n",
       "      <td>GLQ</td>\n",
       "      <td>655</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0</td>\n",
       "      <td>490</td>\n",
       "      <td>1145</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>1145</td>\n",
       "      <td>1053</td>\n",
       "      <td>0</td>\n",
       "      <td>2198</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>9</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>2000.0</td>\n",
       "      <td>RFn</td>\n",
       "      <td>3</td>\n",
       "      <td>836</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>192</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0   1          60       RL         65.0     8450   Pave   NaN      Reg   \n",
       "1   2          20       RL         80.0     9600   Pave   NaN      Reg   \n",
       "2   3          60       RL         68.0    11250   Pave   NaN      IR1   \n",
       "3   4          70       RL         60.0     9550   Pave   NaN      IR1   \n",
       "4   5          60       RL         84.0    14260   Pave   NaN      IR1   \n",
       "\n",
       "  LandContour Utilities LotConfig LandSlope Neighborhood Condition1  \\\n",
       "0         Lvl    AllPub    Inside       Gtl      CollgCr       Norm   \n",
       "1         Lvl    AllPub       FR2       Gtl      Veenker      Feedr   \n",
       "2         Lvl    AllPub    Inside       Gtl      CollgCr       Norm   \n",
       "3         Lvl    AllPub    Corner       Gtl      Crawfor       Norm   \n",
       "4         Lvl    AllPub       FR2       Gtl      NoRidge       Norm   \n",
       "\n",
       "  Condition2 BldgType HouseStyle  OverallQual  OverallCond  YearBuilt  \\\n",
       "0       Norm     1Fam     2Story            7            5       2003   \n",
       "1       Norm     1Fam     1Story            6            8       1976   \n",
       "2       Norm     1Fam     2Story            7            5       2001   \n",
       "3       Norm     1Fam     2Story            7            5       1915   \n",
       "4       Norm     1Fam     2Story            8            5       2000   \n",
       "\n",
       "   YearRemodAdd RoofStyle RoofMatl Exterior1st Exterior2nd MasVnrType  \\\n",
       "0          2003     Gable  CompShg     VinylSd     VinylSd    BrkFace   \n",
       "1          1976     Gable  CompShg     MetalSd     MetalSd       None   \n",
       "2          2002     Gable  CompShg     VinylSd     VinylSd    BrkFace   \n",
       "3          1970     Gable  CompShg     Wd Sdng     Wd Shng       None   \n",
       "4          2000     Gable  CompShg     VinylSd     VinylSd    BrkFace   \n",
       "\n",
       "   MasVnrArea ExterQual ExterCond Foundation BsmtQual BsmtCond BsmtExposure  \\\n",
       "0       196.0        Gd        TA      PConc       Gd       TA           No   \n",
       "1         0.0        TA        TA     CBlock       Gd       TA           Gd   \n",
       "2       162.0        Gd        TA      PConc       Gd       TA           Mn   \n",
       "3         0.0        TA        TA     BrkTil       TA       Gd           No   \n",
       "4       350.0        Gd        TA      PConc       Gd       TA           Av   \n",
       "\n",
       "  BsmtFinType1  BsmtFinSF1 BsmtFinType2  BsmtFinSF2  BsmtUnfSF  TotalBsmtSF  \\\n",
       "0          GLQ         706          Unf           0        150          856   \n",
       "1          ALQ         978          Unf           0        284         1262   \n",
       "2          GLQ         486          Unf           0        434          920   \n",
       "3          ALQ         216          Unf           0        540          756   \n",
       "4          GLQ         655          Unf           0        490         1145   \n",
       "\n",
       "  Heating HeatingQC CentralAir Electrical  1stFlrSF  2ndFlrSF  LowQualFinSF  \\\n",
       "0    GasA        Ex          Y      SBrkr       856       854             0   \n",
       "1    GasA        Ex          Y      SBrkr      1262         0             0   \n",
       "2    GasA        Ex          Y      SBrkr       920       866             0   \n",
       "3    GasA        Gd          Y      SBrkr       961       756             0   \n",
       "4    GasA        Ex          Y      SBrkr      1145      1053             0   \n",
       "\n",
       "   GrLivArea  BsmtFullBath  BsmtHalfBath  FullBath  HalfBath  BedroomAbvGr  \\\n",
       "0       1710             1             0         2         1             3   \n",
       "1       1262             0             1         2         0             3   \n",
       "2       1786             1             0         2         1             3   \n",
       "3       1717             1             0         1         0             3   \n",
       "4       2198             1             0         2         1             4   \n",
       "\n",
       "   KitchenAbvGr KitchenQual  TotRmsAbvGrd Functional  Fireplaces FireplaceQu  \\\n",
       "0             1          Gd             8        Typ           0         NaN   \n",
       "1             1          TA             6        Typ           1          TA   \n",
       "2             1          Gd             6        Typ           1          TA   \n",
       "3             1          Gd             7        Typ           1          Gd   \n",
       "4             1          Gd             9        Typ           1          TA   \n",
       "\n",
       "  GarageType  GarageYrBlt GarageFinish  GarageCars  GarageArea GarageQual  \\\n",
       "0     Attchd       2003.0          RFn           2         548         TA   \n",
       "1     Attchd       1976.0          RFn           2         460         TA   \n",
       "2     Attchd       2001.0          RFn           2         608         TA   \n",
       "3     Detchd       1998.0          Unf           3         642         TA   \n",
       "4     Attchd       2000.0          RFn           3         836         TA   \n",
       "\n",
       "  GarageCond PavedDrive  WoodDeckSF  OpenPorchSF  EnclosedPorch  3SsnPorch  \\\n",
       "0         TA          Y           0           61              0          0   \n",
       "1         TA          Y         298            0              0          0   \n",
       "2         TA          Y           0           42              0          0   \n",
       "3         TA          Y           0           35            272          0   \n",
       "4         TA          Y         192           84              0          0   \n",
       "\n",
       "   ScreenPorch  PoolArea PoolQC Fence MiscFeature  MiscVal  MoSold  YrSold  \\\n",
       "0            0         0    NaN   NaN         NaN        0       2    2008   \n",
       "1            0         0    NaN   NaN         NaN        0       5    2007   \n",
       "2            0         0    NaN   NaN         NaN        0       9    2008   \n",
       "3            0         0    NaN   NaN         NaN        0       2    2006   \n",
       "4            0         0    NaN   NaN         NaN        0      12    2008   \n",
       "\n",
       "  SaleType SaleCondition  SalePrice  \n",
       "0       WD        Normal     208500  \n",
       "1       WD        Normal     181500  \n",
       "2       WD        Normal     223500  \n",
       "3       WD       Abnorml     140000  \n",
       "4       WD        Normal     250000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train = pd.read_csv('.\\\\data\\\\HPR\\\\train\\\\train.csv')\n",
    "data_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{dtype('int64'): Index(['Id', 'MSSubClass', 'LotArea', 'OverallQual', 'OverallCond',\n",
       "        'YearBuilt', 'YearRemodAdd', 'BsmtFinSF1', 'BsmtFinSF2', 'BsmtUnfSF',\n",
       "        'TotalBsmtSF', '1stFlrSF', '2ndFlrSF', 'LowQualFinSF', 'GrLivArea',\n",
       "        'BsmtFullBath', 'BsmtHalfBath', 'FullBath', 'HalfBath', 'BedroomAbvGr',\n",
       "        'KitchenAbvGr', 'TotRmsAbvGrd', 'Fireplaces', 'GarageCars',\n",
       "        'GarageArea', 'WoodDeckSF', 'OpenPorchSF', 'EnclosedPorch', '3SsnPorch',\n",
       "        'ScreenPorch', 'PoolArea', 'MiscVal', 'MoSold', 'YrSold', 'SalePrice'],\n",
       "       dtype='object'),\n",
       " dtype('float64'): Index(['LotFrontage', 'MasVnrArea', 'GarageYrBlt'], dtype='object'),\n",
       " dtype('O'): Index(['MSZoning', 'Street', 'Alley', 'LotShape', 'LandContour', 'Utilities',\n",
       "        'LotConfig', 'LandSlope', 'Neighborhood', 'Condition1', 'Condition2',\n",
       "        'BldgType', 'HouseStyle', 'RoofStyle', 'RoofMatl', 'Exterior1st',\n",
       "        'Exterior2nd', 'MasVnrType', 'ExterQual', 'ExterCond', 'Foundation',\n",
       "        'BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinType2',\n",
       "        'Heating', 'HeatingQC', 'CentralAir', 'Electrical', 'KitchenQual',\n",
       "        'Functional', 'FireplaceQu', 'GarageType', 'GarageFinish', 'GarageQual',\n",
       "        'GarageCond', 'PavedDrive', 'PoolQC', 'Fence', 'MiscFeature',\n",
       "        'SaleType', 'SaleCondition'],\n",
       "       dtype='object')}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.columns.to_series().groupby(data_train.dtypes).groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LotFrontage      259\n",
       "Alley           1369\n",
       "MasVnrType         8\n",
       "MasVnrArea         8\n",
       "BsmtQual          37\n",
       "BsmtCond          37\n",
       "BsmtExposure      38\n",
       "BsmtFinType1      37\n",
       "BsmtFinType2      38\n",
       "Electrical         1\n",
       "FireplaceQu      690\n",
       "GarageType        81\n",
       "GarageYrBlt       81\n",
       "GarageFinish      81\n",
       "GarageQual        81\n",
       "GarageCond        81\n",
       "PoolQC          1453\n",
       "Fence           1179\n",
       "MiscFeature     1406\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = data_train.isna().sum()\n",
    "df[df > 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1460, 81)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#pre processing data\n",
    "def pre_process(st, label, one, data):\n",
    "    data['MSSubClass'] = data['MSSubClass'] / 190\n",
    "    data['LotFrontage'].fillna(0, inplace=True)\n",
    "    data['Alley'].fillna('NA', inplace=True)\n",
    "    data['OverallQual'] = data['OverallQual'] / 10\n",
    "    data['OverallCond'] = data['OverallCond'] / 10\n",
    "    data['YearBuilt'] = data['YearBuilt'] / 1000\n",
    "    data['YearRemodAdd'] = data['YearRemodAdd'] /1000\n",
    "    data['MasVnrType'].fillna('NA', inplace=True)\n",
    "    data['MasVnrArea'].fillna(0, inplace=True)\n",
    "    data['ExterQual'] = data['ExterQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['ExterCond'] = data['ExterCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtQual'] = data['BsmtQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtQual'].fillna(0, inplace=True)\n",
    "    data['BsmtCond'] = data['BsmtCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtCond'].fillna(0, inplace=True)\n",
    "    data['BsmtExposure'] = data['BsmtExposure'].replace(['Gd', 'Av', 'Mn', 'No'], [1, 0.67, 0.34, 0]).astype(pd.np.float)\n",
    "    data['BsmtExposure'].fillna(0, inplace=True)\n",
    "    data['BsmtFinType1'] = data['BsmtFinType1'].replace(['GLQ', 'ALQ', 'BLQ', 'Rec', 'LwQ', 'Unf'], [1 ,0.80, 0.64, 0.48, 0.32, 0.16])\n",
    "    data['BsmtFinType1'].fillna(0, inplace=True)\n",
    "    data['BsmtFinType2'] = data['BsmtFinType2'].replace(['GLQ', 'ALQ', 'BLQ', 'Rec', 'LwQ', 'Unf'], [1 ,0.80, 0.64, 0.48, 0.32, 0.16])\n",
    "    data['BsmtFinType2'].fillna(0, inplace=True)\n",
    "    data['HeatingQC'] = data['HeatingQC'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['Electrical'].fillna('Mix', inplace=True)\n",
    "    data['KitchenQual'] = data['KitchenQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['Functional'] = data['Functional'].replace(['Typ', 'Min1', 'Min2', 'Mod', 'Maj1', 'Maj2', 'Sev', 'Sal'],\n",
    "                                                    [1, 0.875, 0.75, 0.625, 0.50, 0.325, 0.2, 0]).astype(pd.np.float)\n",
    "    data['FireplaceQu'] = data['FireplaceQu'].replace(['Ex', 'Gd', 'TA', 'Fa','Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['FireplaceQu'].fillna(0, inplace=True)\n",
    "    data['GarageType'].fillna('NA', inplace=True)\n",
    "    data['GarageYrBlt'] = data['GarageYrBlt'] / 1000\n",
    "    data['GarageFinish'] = data['GarageFinish'].replace(['Fin', 'RFn', 'Unf'], [1, 0.6, 0.3]).astype(pd.np.float)\n",
    "    data['GarageFinish'].fillna(0, inplace=True)\n",
    "    data['GarageQual'] = data['GarageQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['GarageQual'].fillna(0, inplace=True)\n",
    "    data['GarageCond'] = data['GarageCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['GarageCond'].fillna(0, inplace=True)\n",
    "    data['PavedDrive'] = data['PavedDrive'].replace(['Y', 'P', 'N'], [1, 0.5, 0]).astype(pd.np.float)\n",
    "    data['PoolQC'] = data['PoolQC'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['PoolQC'].fillna(0, inplace=True)\n",
    "    data['Fence'] = data['Fence'].replace(['GdPrv', 'MnPrv', 'GdWo', 'MnWw'], [1, 0.75, 0.50, 0.25]).astype(pd.np.float)\n",
    "    data['Fence'].fillna(0, inplace=True)\n",
    "    data['MiscFeature'].fillna('NA', inplace=True)\n",
    "    data_num = data[['MSSubClass', 'LotFrontage', 'LotArea', 'OverallQual', 'OverallCond',\n",
    "                     'YearBuilt', 'YearRemodAdd', 'MasVnrArea', 'ExterQual', 'ExterCond',\n",
    "                     'BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinSF1', 'BsmtFinType2',\n",
    "                     'BsmtFinSF2', 'BsmtUnfSF', 'TotalBsmtSF', '1stFlrSF', '2ndFlrSF', 'LowQualFinSF', 'GrLivArea',\n",
    "                     'BsmtFullBath', 'BsmtHalfBath', 'FullBath', 'HalfBath', 'BedroomAbvGr', 'KitchenAbvGr', 'KitchenQual',\n",
    "                     'TotRmsAbvGrd', 'Functional', 'Fireplaces', 'FireplaceQu', 'GarageFinish', 'GarageCars',\n",
    "                     'GarageArea', 'GarageQual', 'GarageCond', 'PavedDrive', 'WoodDeckSF', 'OpenPorchSF', 'EnclosedPorch',\n",
    "                     '3SsnPorch', 'ScreenPorch', 'PoolArea', 'PoolQC', 'Fence', 'MiscVal']]\n",
    "    data_cat = data[['MSZoning', 'Street', 'Alley', 'LotShape', 'LandContour', 'Utilities',\n",
    "                     'LotConfig', 'LandSlope', 'Neighborhood', 'Condition1', 'Condition2',\n",
    "                     'BldgType', 'HouseStyle', 'RoofStyle', 'RoofMatl', 'Exterior1st', 'MasVnrType', 'Foundation',\n",
    "                     'Heating', 'CentralAir', 'Electrical', 'GarageType', 'MiscFeature', 'SaleType', 'SaleCondition']]\n",
    "    \n",
    "    label_fit = data_cat.apply(lambda x: label_dict[x.name].fit_transform(x.astype(str)))\n",
    "    data_cat = one.fit_transform(label_fit)\n",
    "    return st.fit_transform(np.append(data_num.values, data_cat, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\ProgramData\\Anaconda3\\envs\\sam_new\\lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:371: FutureWarning: The handling of integer data will change in version 0.22. Currently, the categories are determined based on the range [0, max(values)], while in the future they will be determined based on the unique values.\n",
      "If you want the future behaviour and silence this warning, you can specify \"categories='auto'\".\n",
      "In case you used a LabelEncoder before this OneHotEncoder to convert the categories to integers, then you can now use the OneHotEncoder directly.\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "st = StandardScaler()\n",
    "label_dict = defaultdict(LabelEncoder)\n",
    "# Encoding the variable\n",
    "one = OneHotEncoder(sparse=False)\n",
    "data = pre_process(st, label_dict, one, data_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1460, 212)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\ProgramData\\Anaconda3\\envs\\sam_new\\lib\\site-packages\\sklearn\\utils\\validation.py:595: DataConversionWarning: Data with input dtype int64 was converted to float64 by StandardScaler.\n",
      "  warnings.warn(msg, DataConversionWarning)\n",
      "D:\\ProgramData\\Anaconda3\\envs\\sam_new\\lib\\site-packages\\sklearn\\utils\\validation.py:595: DataConversionWarning: Data with input dtype int64 was converted to float64 by StandardScaler.\n",
      "  warnings.warn(msg, DataConversionWarning)\n"
     ]
    }
   ],
   "source": [
    "sy = StandardScaler()\n",
    "data_y = sy.fit_transform(data_train['SalePrice'].values.reshape(data_train.shape[0], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_model(): \n",
    "    inputs = Input(shape=(212,))\n",
    "    x = Dense(1024, activation='tanh')(inputs)\n",
    "    x = Dense(512, activation='tanh')(x)\n",
    "    predictions = Dense(1)(x)\n",
    "    model = Model(inputs=inputs, outputs=predictions)\n",
    "    model.compile(optimizer='adam',\n",
    "            loss='mean_squared_error',\n",
    "            metrics=['mean_squared_error'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From D:\\ProgramData\\Anaconda3\\envs\\sam_new\\lib\\site-packages\\tensorflow\\python\\framework\\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From D:\\ProgramData\\Anaconda3\\envs\\sam_new\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 3s 2ms/step - loss: 1.2386 - mean_squared_error: 1.2386\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 266us/step - loss: 0.4347 - mean_squared_error: 0.4347\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 184us/step - loss: 0.3184 - mean_squared_error: 0.3184\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 187us/step - loss: 0.1943 - mean_squared_error: 0.1943\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.1243 - mean_squared_error: 0.1243\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 249us/step - loss: 0.0973 - mean_squared_error: 0.0973\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 239us/step - loss: 0.0762 - mean_squared_error: 0.0762\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 240us/step - loss: 0.0684 - mean_squared_error: 0.0684\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 263us/step - loss: 0.0801 - mean_squared_error: 0.0801\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 244us/step - loss: 0.0604 - mean_squared_error: 0.0604\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 239us/step - loss: 0.0476 - mean_squared_error: 0.0476\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 209us/step - loss: 0.0417 - mean_squared_error: 0.0417\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0408 - mean_squared_error: 0.0408\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 247us/step - loss: 0.0650 - mean_squared_error: 0.0650\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 238us/step - loss: 0.0348 - mean_squared_error: 0.0348\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 209us/step - loss: 0.0258 - mean_squared_error: 0.0258\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 236us/step - loss: 0.0273 - mean_squared_error: 0.0273\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 193us/step - loss: 0.0331 - mean_squared_error: 0.0331\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0326 - mean_squared_error: 0.0326\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 1s 441us/step - loss: 0.0308 - mean_squared_error: 0.0308\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 1s 436us/step - loss: 0.0524 - mean_squared_error: 0.05240s - loss: 0.0491 - mean_squared_e\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 1s 438us/step - loss: 0.0415 - mean_squared_error: 0.0415\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 1s 402us/step - loss: 0.0428 - mean_squared_error: 0.0428\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 1s 435us/step - loss: 0.1145 - mean_squared_error: 0.1145\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 1s 474us/step - loss: 0.0474 - mean_squared_error: 0.0474\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 1s 461us/step - loss: 0.0340 - mean_squared_error: 0.03400s - loss: 0.0358 - mean_squared_error: \n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 1s 421us/step - loss: 0.0222 - mean_squared_error: 0.0222\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 1s 450us/step - loss: 0.0188 - mean_squared_error: 0.0188\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 1s 419us/step - loss: 0.0141 - mean_squared_error: 0.0141\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 1s 461us/step - loss: 0.0135 - mean_squared_error: 0.0135\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 1s 466us/step - loss: 0.0156 - mean_squared_error: 0.0156\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 1s 459us/step - loss: 0.0155 - mean_squared_error: 0.01550s - loss: 0.0153 - mean_squared_error: 0.01\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 1s 441us/step - loss: 0.0165 - mean_squared_error: 0.0165\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 1s 479us/step - loss: 0.0121 - mean_squared_error: 0.0121\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 1s 415us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 1s 409us/step - loss: 0.0208 - mean_squared_error: 0.0208\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 1s 413us/step - loss: 0.0286 - mean_squared_error: 0.0286\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 375us/step - loss: 0.0340 - mean_squared_error: 0.0340\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 368us/step - loss: 0.0299 - mean_squared_error: 0.0299\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0205 - mean_squared_error: 0.0205\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 371us/step - loss: 0.0126 - mean_squared_error: 0.0126\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 359us/step - loss: 0.0138 - mean_squared_error: 0.0138\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 368us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0114 - mean_squared_error: 0.0114\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 1s 408us/step - loss: 0.0108 - mean_squared_error: 0.0108\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 1s 476us/step - loss: 0.0217 - mean_squared_error: 0.0217\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 1s 408us/step - loss: 0.0588 - mean_squared_error: 0.0588\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 1s 414us/step - loss: 0.0374 - mean_squared_error: 0.0374\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 1s 457us/step - loss: 0.0288 - mean_squared_error: 0.02880s - loss: 0.0288 - mean_squared_error: 0.02\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 326us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0216 - mean_squared_error: 0.0216\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 330us/step - loss: 0.0156 - mean_squared_error: 0.0156\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0121 - mean_squared_error: 0.0121\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0093 - mean_squared_error: 0.0093\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0090 - mean_squared_error: 0.00900s - loss: 0.0106 - mean_squared_e\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 1s 392us/step - loss: 0.0095 - mean_squared_error: 0.0095\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.0121 - mean_squared_error: 0.0121\n",
      "Epoch 65/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 370us/step - loss: 0.0171 - mean_squared_error: 0.0171\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 268us/step - loss: 0.0210 - mean_squared_error: 0.0210\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 314us/step - loss: 0.0297 - mean_squared_error: 0.0297\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0284 - mean_squared_error: 0.02840s - loss: 0.0326 - mean_squared\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0538 - mean_squared_error: 0.0538\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.0359 - mean_squared_error: 0.0359\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 1s 393us/step - loss: 0.0250 - mean_squared_error: 0.0250\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 1s 413us/step - loss: 0.0158 - mean_squared_error: 0.01580s - loss: 0.0163 - mean_squared_e\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 1s 406us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 1s 382us/step - loss: 0.0193 - mean_squared_error: 0.0193\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0163 - mean_squared_error: 0.0163\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0206 - mean_squared_error: 0.0206\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.0218 - mean_squared_error: 0.0218\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0193 - mean_squared_error: 0.0193\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 334us/step - loss: 0.0240 - mean_squared_error: 0.0240\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 290us/step - loss: 0.0210 - mean_squared_error: 0.0210\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 1s 389us/step - loss: 0.0188 - mean_squared_error: 0.0188\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 1s 401us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0199 - mean_squared_error: 0.0199\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0321 - mean_squared_error: 0.0321\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.0360 - mean_squared_error: 0.0360\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 1s 395us/step - loss: 0.0371 - mean_squared_error: 0.0371\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 369us/step - loss: 0.2320 - mean_squared_error: 0.2320\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.1434 - mean_squared_error: 0.1434\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 375us/step - loss: 0.0737 - mean_squared_error: 0.0737\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 1s 451us/step - loss: 0.0468 - mean_squared_error: 0.0468\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0271 - mean_squared_error: 0.0271\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.0569 - mean_squared_error: 0.0569\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 1s 405us/step - loss: 0.0263 - mean_squared_error: 0.0263\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0205 - mean_squared_error: 0.02050s - loss: 0.0185 - mean_squared_err\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 366us/step - loss: 0.0222 - mean_squared_error: 0.0222\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 319us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "146/146 [==============================] - 0s 301us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 0s 306us/step - loss: 1.2342 - mean_squared_error: 1.2342\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 166us/step - loss: 0.4694 - mean_squared_error: 0.4694\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 164us/step - loss: 0.2169 - mean_squared_error: 0.2169\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 156us/step - loss: 0.1723 - mean_squared_error: 0.1723\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.1510 - mean_squared_error: 0.1510\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 158us/step - loss: 0.1271 - mean_squared_error: 0.1271\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 162us/step - loss: 0.1029 - mean_squared_error: 0.1029\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 167us/step - loss: 0.0651 - mean_squared_error: 0.0651\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 156us/step - loss: 0.0587 - mean_squared_error: 0.0587\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.0405 - mean_squared_error: 0.0405\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 157us/step - loss: 0.0356 - mean_squared_error: 0.0356\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 167us/step - loss: 0.0301 - mean_squared_error: 0.0301\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 160us/step - loss: 0.0308 - mean_squared_error: 0.0308\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 165us/step - loss: 0.0384 - mean_squared_error: 0.0384\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 162us/step - loss: 0.0301 - mean_squared_error: 0.0301\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 159us/step - loss: 0.0292 - mean_squared_error: 0.0292\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0300 - mean_squared_error: 0.0300\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 158us/step - loss: 0.0251 - mean_squared_error: 0.0251\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.0323 - mean_squared_error: 0.0323\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 159us/step - loss: 0.0256 - mean_squared_error: 0.0256\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 164us/step - loss: 0.0326 - mean_squared_error: 0.0326\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 246us/step - loss: 0.0286 - mean_squared_error: 0.0286\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 1s 383us/step - loss: 0.0249 - mean_squared_error: 0.0249\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0227 - mean_squared_error: 0.0227\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 1s 410us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 1s 424us/step - loss: 0.0208 - mean_squared_error: 0.0208\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 1s 463us/step - loss: 0.0329 - mean_squared_error: 0.0329\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 1s 441us/step - loss: 0.0242 - mean_squared_error: 0.0242\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 1s 450us/step - loss: 0.0291 - mean_squared_error: 0.0291\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 1s 400us/step - loss: 0.0334 - mean_squared_error: 0.0334\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0219 - mean_squared_error: 0.0219\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 1s 406us/step - loss: 0.0620 - mean_squared_error: 0.0620\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 1s 478us/step - loss: 0.0516 - mean_squared_error: 0.0516\n",
      "Epoch 34/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 1s 452us/step - loss: 0.0279 - mean_squared_error: 0.0279\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 1s 433us/step - loss: 0.0164 - mean_squared_error: 0.0164\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 1s 429us/step - loss: 0.0116 - mean_squared_error: 0.0116\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 1s 464us/step - loss: 0.0099 - mean_squared_error: 0.00990s - loss: 0.0102 - mean_squared_err\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 188us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 200us/step - loss: 0.0126 - mean_squared_error: 0.0126\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 218us/step - loss: 0.0124 - mean_squared_error: 0.0124\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 198us/step - loss: 0.0214 - mean_squared_error: 0.0214\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 177us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 180us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 175us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 175us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.0187 - mean_squared_error: 0.0187\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 172us/step - loss: 0.0995 - mean_squared_error: 0.0995\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0811 - mean_squared_error: 0.0811\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 171us/step - loss: 0.0395 - mean_squared_error: 0.0395\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 166us/step - loss: 0.0240 - mean_squared_error: 0.0240\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 175us/step - loss: 0.0187 - mean_squared_error: 0.0187\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 180us/step - loss: 0.0143 - mean_squared_error: 0.0143\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 189us/step - loss: 0.0106 - mean_squared_error: 0.0106\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 194us/step - loss: 0.0097 - mean_squared_error: 0.0097\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 186us/step - loss: 0.0104 - mean_squared_error: 0.0104\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 169us/step - loss: 0.0136 - mean_squared_error: 0.0136\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 175us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0606 - mean_squared_error: 0.0606\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 206us/step - loss: 0.0410 - mean_squared_error: 0.0410\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 1s 417us/step - loss: 0.0373 - mean_squared_error: 0.0373\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 1s 414us/step - loss: 0.0328 - mean_squared_error: 0.0328\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 1s 411us/step - loss: 0.0211 - mean_squared_error: 0.0211\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0237 - mean_squared_error: 0.0237\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 1s 383us/step - loss: 0.0271 - mean_squared_error: 0.0271\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 1s 402us/step - loss: 0.0268 - mean_squared_error: 0.0268\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 1s 418us/step - loss: 0.0153 - mean_squared_error: 0.0153\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0228 - mean_squared_error: 0.0228\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 1s 398us/step - loss: 0.0223 - mean_squared_error: 0.0223\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 369us/step - loss: 0.0361 - mean_squared_error: 0.0361\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 314us/step - loss: 0.0238 - mean_squared_error: 0.0238\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 248us/step - loss: 0.0192 - mean_squared_error: 0.0192\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 329us/step - loss: 0.0121 - mean_squared_error: 0.0121\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 227us/step - loss: 0.0107 - mean_squared_error: 0.0107\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0101 - mean_squared_error: 0.0101\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 262us/step - loss: 0.0100 - mean_squared_error: 0.0100\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 305us/step - loss: 0.0101 - mean_squared_error: 0.0101\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0106 - mean_squared_error: 0.0106\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 317us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 332us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 335us/step - loss: 0.0163 - mean_squared_error: 0.0163\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0212 - mean_squared_error: 0.0212\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0246 - mean_squared_error: 0.0246\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0158 - mean_squared_error: 0.0158\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.0175 - mean_squared_error: 0.0175\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 291us/step - loss: 0.0281 - mean_squared_error: 0.0281\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 321us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0207 - mean_squared_error: 0.0207\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 1s 411us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 263us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 222us/step - loss: 0.0129 - mean_squared_error: 0.0129\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 278us/step - loss: 0.0211 - mean_squared_error: 0.0211\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 331us/step - loss: 0.0228 - mean_squared_error: 0.0228\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0256 - mean_squared_error: 0.0256\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 324us/step - loss: 0.0202 - mean_squared_error: 0.0202\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 283us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 238us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 316us/step - loss: 0.0126 - mean_squared_error: 0.0126\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 236us/step - loss: 0.0195 - mean_squared_error: 0.0195\n",
      "146/146 [==============================] - 0s 806us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 599us/step - loss: 1.1868 - mean_squared_error: 1.1868\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.5802 - mean_squared_error: 0.5802\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 321us/step - loss: 0.2846 - mean_squared_error: 0.2846\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 238us/step - loss: 0.1760 - mean_squared_error: 0.1760\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.1426 - mean_squared_error: 0.1426\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.1126 - mean_squared_error: 0.1126\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0842 - mean_squared_error: 0.0842\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0592 - mean_squared_error: 0.0592\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0560 - mean_squared_error: 0.0560\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.1786 - mean_squared_error: 0.1786\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0784 - mean_squared_error: 0.0784\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 1s 408us/step - loss: 0.0476 - mean_squared_error: 0.0476\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0476 - mean_squared_error: 0.0476\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0369 - mean_squared_error: 0.0369\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 338us/step - loss: 0.0356 - mean_squared_error: 0.0356\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0243 - mean_squared_error: 0.0243\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0328 - mean_squared_error: 0.0328\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 1s 410us/step - loss: 0.0358 - mean_squared_error: 0.0358\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 1s 392us/step - loss: 0.0317 - mean_squared_error: 0.0317\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0293 - mean_squared_error: 0.0293\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0418 - mean_squared_error: 0.0418\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0250 - mean_squared_error: 0.0250\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.0215 - mean_squared_error: 0.0215\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0206 - mean_squared_error: 0.0206\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0274 - mean_squared_error: 0.0274\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0283 - mean_squared_error: 0.0283\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 225us/step - loss: 0.0224 - mean_squared_error: 0.0224\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 266us/step - loss: 0.0173 - mean_squared_error: 0.0173\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 305us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 211us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 331us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0145 - mean_squared_error: 0.0145\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0113 - mean_squared_error: 0.0113\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 335us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0165 - mean_squared_error: 0.0165\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0212 - mean_squared_error: 0.0212\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0216 - mean_squared_error: 0.0216\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0195 - mean_squared_error: 0.0195\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0143 - mean_squared_error: 0.0143\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 272us/step - loss: 0.0481 - mean_squared_error: 0.0481\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0383 - mean_squared_error: 0.0383\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 233us/step - loss: 0.0658 - mean_squared_error: 0.0658\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 289us/step - loss: 0.0443 - mean_squared_error: 0.0443\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0232 - mean_squared_error: 0.0232\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 288us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 265us/step - loss: 0.0129 - mean_squared_error: 0.0129\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 316us/step - loss: 0.0130 - mean_squared_error: 0.0130\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 1s 402us/step - loss: 0.0101 - mean_squared_error: 0.0101\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 1s 405us/step - loss: 0.0107 - mean_squared_error: 0.0107\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 1s 395us/step - loss: 0.0165 - mean_squared_error: 0.0165\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 380us/step - loss: 0.0166 - mean_squared_error: 0.0166\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 368us/step - loss: 0.0131 - mean_squared_error: 0.0131\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 372us/step - loss: 0.0143 - mean_squared_error: 0.0143\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0121 - mean_squared_error: 0.0121\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0147 - mean_squared_error: 0.0147\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 282us/step - loss: 0.0117 - mean_squared_error: 0.0117\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 269us/step - loss: 0.0101 - mean_squared_error: 0.0101\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0147 - mean_squared_error: 0.0147\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0111 - mean_squared_error: 0.0111\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 212us/step - loss: 0.0170 - mean_squared_error: 0.0170\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 312us/step - loss: 0.0173 - mean_squared_error: 0.0173\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 232us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0326 - mean_squared_error: 0.0326\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 219us/step - loss: 0.0317 - mean_squared_error: 0.0317\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 287us/step - loss: 0.0215 - mean_squared_error: 0.0215\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 257us/step - loss: 0.0213 - mean_squared_error: 0.0213\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 337us/step - loss: 0.0216 - mean_squared_error: 0.0216\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 260us/step - loss: 0.0258 - mean_squared_error: 0.0258\n",
      "Epoch 74/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 277us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 191us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 277us/step - loss: 0.0124 - mean_squared_error: 0.0124\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0101 - mean_squared_error: 0.0101\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0127 - mean_squared_error: 0.0127\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0320 - mean_squared_error: 0.0320\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0275 - mean_squared_error: 0.0275\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0525 - mean_squared_error: 0.0525\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0367 - mean_squared_error: 0.0367\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0442 - mean_squared_error: 0.0442\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0328 - mean_squared_error: 0.0328\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 363us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 369us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.0169 - mean_squared_error: 0.0169\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 374us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 284us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 303us/step - loss: 0.0134 - mean_squared_error: 0.0134\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 276us/step - loss: 0.0106 - mean_squared_error: 0.0106\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0159 - mean_squared_error: 0.0159\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 267us/step - loss: 0.0125 - mean_squared_error: 0.0125\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 336us/step - loss: 0.0109 - mean_squared_error: 0.0109\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 328us/step - loss: 0.0104 - mean_squared_error: 0.0104\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 287us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "146/146 [==============================] - 0s 833us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 405us/step - loss: 1.4180 - mean_squared_error: 1.4180\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 212us/step - loss: 0.4860 - mean_squared_error: 0.4860\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 301us/step - loss: 0.2483 - mean_squared_error: 0.2483\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.1611 - mean_squared_error: 0.1611\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.1231 - mean_squared_error: 0.1231\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 332us/step - loss: 0.1167 - mean_squared_error: 0.1167\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.1263 - mean_squared_error: 0.1263\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 340us/step - loss: 0.1492 - mean_squared_error: 0.1492\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0567 - mean_squared_error: 0.0567\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 337us/step - loss: 0.0609 - mean_squared_error: 0.0609\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0477 - mean_squared_error: 0.0477\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0724 - mean_squared_error: 0.0724\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.0419 - mean_squared_error: 0.0419\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0273 - mean_squared_error: 0.0273\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0269 - mean_squared_error: 0.0269\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0227 - mean_squared_error: 0.0227\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 338us/step - loss: 0.0304 - mean_squared_error: 0.0304\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 358us/step - loss: 0.0270 - mean_squared_error: 0.0270\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0452 - mean_squared_error: 0.0452\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0336 - mean_squared_error: 0.0336\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0294 - mean_squared_error: 0.0294\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0279 - mean_squared_error: 0.0279\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0262 - mean_squared_error: 0.0262\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 298us/step - loss: 0.0301 - mean_squared_error: 0.0301\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 312us/step - loss: 0.0249 - mean_squared_error: 0.0249\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 298us/step - loss: 0.0164 - mean_squared_error: 0.0164\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 319us/step - loss: 0.0218 - mean_squared_error: 0.0218\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 301us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0354 - mean_squared_error: 0.0354\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 340us/step - loss: 0.0341 - mean_squared_error: 0.0341\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 281us/step - loss: 0.0369 - mean_squared_error: 0.0369\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 135us/step - loss: 0.0262 - mean_squared_error: 0.0262\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 137us/step - loss: 0.0188 - mean_squared_error: 0.0188\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 138us/step - loss: 0.0181 - mean_squared_error: 0.0181\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0220 - mean_squared_error: 0.0220\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0152 - mean_squared_error: 0.0152\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0230 - mean_squared_error: 0.0230\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0166 - mean_squared_error: 0.0166\n",
      "Epoch 44/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0253 - mean_squared_error: 0.0253\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0237 - mean_squared_error: 0.0237\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0219 - mean_squared_error: 0.0219\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0201 - mean_squared_error: 0.0201\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0163 - mean_squared_error: 0.0163\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0155 - mean_squared_error: 0.0155\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0228 - mean_squared_error: 0.0228\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0176 - mean_squared_error: 0.0176\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0176 - mean_squared_error: 0.0176\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 177us/step - loss: 0.0201 - mean_squared_error: 0.0201\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0208 - mean_squared_error: 0.0208\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 343us/step - loss: 0.0232 - mean_squared_error: 0.0232\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0172 - mean_squared_error: 0.0172\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 266us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0123 - mean_squared_error: 0.0123\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0169 - mean_squared_error: 0.0169\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0159 - mean_squared_error: 0.0159\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0237 - mean_squared_error: 0.0237\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 287us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0232 - mean_squared_error: 0.0232\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 273us/step - loss: 0.0230 - mean_squared_error: 0.0230\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 289us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 269us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 292us/step - loss: 0.0130 - mean_squared_error: 0.0130\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 257us/step - loss: 0.0105 - mean_squared_error: 0.0105\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0093 - mean_squared_error: 0.0093\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 162us/step - loss: 0.0109 - mean_squared_error: 0.0109\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 163us/step - loss: 0.0117 - mean_squared_error: 0.0117\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 162us/step - loss: 0.0112 - mean_squared_error: 0.0112\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0412 - mean_squared_error: 0.0412\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0422 - mean_squared_error: 0.0422\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0316 - mean_squared_error: 0.0316\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0305 - mean_squared_error: 0.0305\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0269 - mean_squared_error: 0.0269\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0348 - mean_squared_error: 0.0348\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0379 - mean_squared_error: 0.0379\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0267 - mean_squared_error: 0.0267\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0173 - mean_squared_error: 0.0173\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0135 - mean_squared_error: 0.0135\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0220 - mean_squared_error: 0.0220\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0228 - mean_squared_error: 0.0228\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0244 - mean_squared_error: 0.0244\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "146/146 [==============================] - 0s 492us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 565us/step - loss: 1.3241 - mean_squared_error: 1.3241\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.5985 - mean_squared_error: 0.5985\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.2587 - mean_squared_error: 0.2587\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.1775 - mean_squared_error: 0.1775\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.1552 - mean_squared_error: 0.1552\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 276us/step - loss: 0.0805 - mean_squared_error: 0.0805\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - ETA: 0s - loss: 0.0821 - mean_squared_error: 0.08 - 0s 344us/step - loss: 0.0819 - mean_squared_error: 0.0819\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0621 - mean_squared_error: 0.0621\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0598 - mean_squared_error: 0.0598\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0491 - mean_squared_error: 0.0491\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0447 - mean_squared_error: 0.0447\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 166us/step - loss: 0.0351 - mean_squared_error: 0.0351\n",
      "Epoch 13/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 283us/step - loss: 0.0347 - mean_squared_error: 0.0347\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 258us/step - loss: 0.0455 - mean_squared_error: 0.0455\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 289us/step - loss: 0.0417 - mean_squared_error: 0.0417\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 305us/step - loss: 0.0332 - mean_squared_error: 0.0332\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0278 - mean_squared_error: 0.0278\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 169us/step - loss: 0.0307 - mean_squared_error: 0.0307\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 135us/step - loss: 0.0274 - mean_squared_error: 0.0274\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 136us/step - loss: 0.0333 - mean_squared_error: 0.0333\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0277 - mean_squared_error: 0.0277\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0250 - mean_squared_error: 0.0250\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0200 - mean_squared_error: 0.0200\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0235 - mean_squared_error: 0.0235\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0256 - mean_squared_error: 0.0256\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 163us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 145us/step - loss: 0.0224 - mean_squared_error: 0.0224\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0207 - mean_squared_error: 0.0207\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0297 - mean_squared_error: 0.0297\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0241 - mean_squared_error: 0.0241\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0180 - mean_squared_error: 0.0180\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0199 - mean_squared_error: 0.0199\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0326 - mean_squared_error: 0.0326\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 161us/step - loss: 0.0272 - mean_squared_error: 0.0272\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 173us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 164us/step - loss: 0.0220 - mean_squared_error: 0.0220\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 169us/step - loss: 0.0286 - mean_squared_error: 0.0286\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 163us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 169us/step - loss: 0.0294 - mean_squared_error: 0.0294\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0250 - mean_squared_error: 0.0250\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0289 - mean_squared_error: 0.0289\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0231 - mean_squared_error: 0.0231\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0175 - mean_squared_error: 0.0175\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0166 - mean_squared_error: 0.0166\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 202us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0124 - mean_squared_error: 0.0124\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0117 - mean_squared_error: 0.0117\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0120 - mean_squared_error: 0.0120\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 191us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 144us/step - loss: 0.0126 - mean_squared_error: 0.0126\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 134us/step - loss: 0.0114 - mean_squared_error: 0.0114\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0127 - mean_squared_error: 0.0127\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0127 - mean_squared_error: 0.0127\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0141 - mean_squared_error: 0.0141\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 163us/step - loss: 0.0127 - mean_squared_error: 0.0127\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0215 - mean_squared_error: 0.0215\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0195 - mean_squared_error: 0.0195\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0238 - mean_squared_error: 0.0238\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0165 - mean_squared_error: 0.0165\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 186us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0117 - mean_squared_error: 0.0117\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0156 - mean_squared_error: 0.0156\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0142 - mean_squared_error: 0.0142\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 154us/step - loss: 0.0114 - mean_squared_error: 0.0114\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0474 - mean_squared_error: 0.0474\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0476 - mean_squared_error: 0.0476\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0395 - mean_squared_error: 0.0395\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0311 - mean_squared_error: 0.0311\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0277 - mean_squared_error: 0.0277\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 208us/step - loss: 0.0257 - mean_squared_error: 0.0257\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0223 - mean_squared_error: 0.0223\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 83/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0164 - mean_squared_error: 0.0164\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0513 - mean_squared_error: 0.0513\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0397 - mean_squared_error: 0.0397\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0339 - mean_squared_error: 0.0339\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0535 - mean_squared_error: 0.0535\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0460 - mean_squared_error: 0.0460\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - ETA: 0s - loss: 0.0305 - mean_squared_error: 0.03 - 0s 358us/step - loss: 0.0305 - mean_squared_error: 0.0305\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0191 - mean_squared_error: 0.0191\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0189 - mean_squared_error: 0.0189\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0124 - mean_squared_error: 0.0124\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 1s 381us/step - loss: 0.0120 - mean_squared_error: 0.0120\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.0080 - mean_squared_error: 0.0080\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0087 - mean_squared_error: 0.0087\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0091 - mean_squared_error: 0.0091\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 1s 411us/step - loss: 0.0123 - mean_squared_error: 0.0123\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 1s 381us/step - loss: 0.0107 - mean_squared_error: 0.0107\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0094 - mean_squared_error: 0.0094\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0134 - mean_squared_error: 0.0134\n",
      "146/146 [==============================] - 0s 471us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 526us/step - loss: 1.1067 - mean_squared_error: 1.1067\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 297us/step - loss: 0.5712 - mean_squared_error: 0.5712\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 187us/step - loss: 0.2518 - mean_squared_error: 0.2518\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 304us/step - loss: 0.1408 - mean_squared_error: 0.1408\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.1069 - mean_squared_error: 0.1069\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 326us/step - loss: 0.1035 - mean_squared_error: 0.1035\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 310us/step - loss: 0.0810 - mean_squared_error: 0.0810\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 285us/step - loss: 0.0764 - mean_squared_error: 0.0764\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0499 - mean_squared_error: 0.0499\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 228us/step - loss: 0.0618 - mean_squared_error: 0.0618\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 203us/step - loss: 0.0597 - mean_squared_error: 0.0597\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0449 - mean_squared_error: 0.0449\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0478 - mean_squared_error: 0.0478\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 315us/step - loss: 0.0510 - mean_squared_error: 0.0510\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0348 - mean_squared_error: 0.0348\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0315 - mean_squared_error: 0.0315\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0500 - mean_squared_error: 0.0500\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0331 - mean_squared_error: 0.0331\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0335 - mean_squared_error: 0.0335\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0303 - mean_squared_error: 0.0303\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0207 - mean_squared_error: 0.0207\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0191 - mean_squared_error: 0.0191\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0221 - mean_squared_error: 0.0221\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 295us/step - loss: 0.0306 - mean_squared_error: 0.0306\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 316us/step - loss: 0.0477 - mean_squared_error: 0.0477\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 331us/step - loss: 0.0260 - mean_squared_error: 0.0260\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 282us/step - loss: 0.0188 - mean_squared_error: 0.0188\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 324us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 315us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0137 - mean_squared_error: 0.0137\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 302us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 299us/step - loss: 0.0153 - mean_squared_error: 0.0153\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 310us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 340us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 288us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 313us/step - loss: 0.0180 - mean_squared_error: 0.0180\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 300us/step - loss: 0.0353 - mean_squared_error: 0.0353\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 261us/step - loss: 0.0260 - mean_squared_error: 0.0260\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 250us/step - loss: 0.0238 - mean_squared_error: 0.0238\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0206 - mean_squared_error: 0.0206\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 328us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 315us/step - loss: 0.0187 - mean_squared_error: 0.0187\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0379 - mean_squared_error: 0.0379\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0361 - mean_squared_error: 0.0361\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 257us/step - loss: 0.0267 - mean_squared_error: 0.0267\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 299us/step - loss: 0.0418 - mean_squared_error: 0.0418\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 276us/step - loss: 0.0329 - mean_squared_error: 0.0329\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 216us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 52/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0259 - mean_squared_error: 0.0259\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0267 - mean_squared_error: 0.0267\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 293us/step - loss: 0.0130 - mean_squared_error: 0.0130\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 268us/step - loss: 0.0137 - mean_squared_error: 0.0137\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0154 - mean_squared_error: 0.0154\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0437 - mean_squared_error: 0.0437\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 216us/step - loss: 0.0231 - mean_squared_error: 0.0231\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0181 - mean_squared_error: 0.0181\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0179 - mean_squared_error: 0.0179\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 260us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 134us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 137us/step - loss: 0.0100 - mean_squared_error: 0.0100\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 140us/step - loss: 0.0107 - mean_squared_error: 0.0107\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0252 - mean_squared_error: 0.0252\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0188 - mean_squared_error: 0.0188\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 154us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 176us/step - loss: 0.0138 - mean_squared_error: 0.0138\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 169us/step - loss: 0.0173 - mean_squared_error: 0.0173\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 165us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 166us/step - loss: 0.0145 - mean_squared_error: 0.0145\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 164us/step - loss: 0.0179 - mean_squared_error: 0.0179\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 156us/step - loss: 0.0293 - mean_squared_error: 0.0293\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0306 - mean_squared_error: 0.0306\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0424 - mean_squared_error: 0.0424\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0319 - mean_squared_error: 0.0319\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0267 - mean_squared_error: 0.0267\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0173 - mean_squared_error: 0.0173\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0153 - mean_squared_error: 0.0153\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0222 - mean_squared_error: 0.0222\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0230 - mean_squared_error: 0.0230\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0242 - mean_squared_error: 0.0242\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0212 - mean_squared_error: 0.0212\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0261 - mean_squared_error: 0.0261\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0176 - mean_squared_error: 0.0176\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0384 - mean_squared_error: 0.0384\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 147us/step - loss: 0.0280 - mean_squared_error: 0.0280\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0250 - mean_squared_error: 0.0250\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0221 - mean_squared_error: 0.0221\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0236 - mean_squared_error: 0.0236\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 282us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0191 - mean_squared_error: 0.0191\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0155 - mean_squared_error: 0.0155\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0154 - mean_squared_error: 0.0154\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "146/146 [==============================] - 0s 621us/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 723us/step - loss: 1.2024 - mean_squared_error: 1.2024\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 1s 417us/step - loss: 0.4801 - mean_squared_error: 0.4801\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 1s 412us/step - loss: 0.3123 - mean_squared_error: 0.3123\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.1836 - mean_squared_error: 0.1836\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.1256 - mean_squared_error: 0.1256\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 358us/step - loss: 0.0888 - mean_squared_error: 0.0888\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0631 - mean_squared_error: 0.0631\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0553 - mean_squared_error: 0.05530s - loss: 0.0528 - mean_squared_e\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0510 - mean_squared_error: 0.0510\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0499 - mean_squared_error: 0.0499\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0525 - mean_squared_error: 0.0525\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0463 - mean_squared_error: 0.0463\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0448 - mean_squared_error: 0.0448\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 363us/step - loss: 0.0833 - mean_squared_error: 0.0833\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0756 - mean_squared_error: 0.0756\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 1s 385us/step - loss: 0.0411 - mean_squared_error: 0.0411\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 326us/step - loss: 0.0399 - mean_squared_error: 0.0399\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 333us/step - loss: 0.0273 - mean_squared_error: 0.0273\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 304us/step - loss: 0.0461 - mean_squared_error: 0.0461\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 310us/step - loss: 0.0254 - mean_squared_error: 0.0254\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 287us/step - loss: 0.0270 - mean_squared_error: 0.0270\n",
      "Epoch 22/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0257 - mean_squared_error: 0.0257\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 241us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 282us/step - loss: 0.0210 - mean_squared_error: 0.0210\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0285 - mean_squared_error: 0.0285\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 335us/step - loss: 0.0217 - mean_squared_error: 0.0217\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0189 - mean_squared_error: 0.0189\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 1s 420us/step - loss: 0.0280 - mean_squared_error: 0.0280\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 1s 408us/step - loss: 0.0258 - mean_squared_error: 0.0258\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 370us/step - loss: 0.0333 - mean_squared_error: 0.0333\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 377us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 366us/step - loss: 0.0184 - mean_squared_error: 0.0184\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 363us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 267us/step - loss: 0.0134 - mean_squared_error: 0.0134\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 341us/step - loss: 0.0172 - mean_squared_error: 0.0172\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0252 - mean_squared_error: 0.0252\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0119 - mean_squared_error: 0.0119\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0112 - mean_squared_error: 0.0112\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0111 - mean_squared_error: 0.0111\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0123 - mean_squared_error: 0.0123\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0116 - mean_squared_error: 0.0116\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 338us/step - loss: 0.0245 - mean_squared_error: 0.0245\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0333 - mean_squared_error: 0.0333\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 304us/step - loss: 0.0296 - mean_squared_error: 0.0296\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 250us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 334us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 263us/step - loss: 0.0281 - mean_squared_error: 0.0281\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 326us/step - loss: 0.0217 - mean_squared_error: 0.0217\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0234 - mean_squared_error: 0.0234\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 322us/step - loss: 0.0310 - mean_squared_error: 0.0310\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 292us/step - loss: 0.0256 - mean_squared_error: 0.0256\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 278us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 285us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 290us/step - loss: 0.0172 - mean_squared_error: 0.0172\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 269us/step - loss: 0.0134 - mean_squared_error: 0.0134\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 275us/step - loss: 0.0120 - mean_squared_error: 0.0120\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 287us/step - loss: 0.0104 - mean_squared_error: 0.0104\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 275us/step - loss: 0.0106 - mean_squared_error: 0.0106\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 330us/step - loss: 0.0120 - mean_squared_error: 0.0120\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 329us/step - loss: 0.0241 - mean_squared_error: 0.0241\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 316us/step - loss: 0.0300 - mean_squared_error: 0.0300\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0224 - mean_squared_error: 0.0224\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 210us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0293 - mean_squared_error: 0.0293\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 268us/step - loss: 0.0225 - mean_squared_error: 0.0225\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0617 - mean_squared_error: 0.0617\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 364us/step - loss: 0.0318 - mean_squared_error: 0.0318\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 269us/step - loss: 0.0566 - mean_squared_error: 0.0566\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 208us/step - loss: 0.0395 - mean_squared_error: 0.0395\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0287 - mean_squared_error: 0.0287\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0227 - mean_squared_error: 0.0227\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0300 - mean_squared_error: 0.0300\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0242 - mean_squared_error: 0.0242\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0187 - mean_squared_error: 0.0187\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 370us/step - loss: 0.0117 - mean_squared_error: 0.0117\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 359us/step - loss: 0.0088 - mean_squared_error: 0.0088\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 372us/step - loss: 0.0085 - mean_squared_error: 0.0085\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0073 - mean_squared_error: 0.0073\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0089 - mean_squared_error: 0.0089\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 181us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 376us/step - loss: 0.0125 - mean_squared_error: 0.0125\n",
      "Epoch 92/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 370us/step - loss: 0.0165 - mean_squared_error: 0.0165\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 1s 392us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 1s 394us/step - loss: 0.0245 - mean_squared_error: 0.0245\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 1s 386us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0131 - mean_squared_error: 0.0131\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0185 - mean_squared_error: 0.01850s - loss: 0.0194 - mean_squared_e\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0137 - mean_squared_error: 0.0137\n",
      "146/146 [==============================] - 0s 2ms/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 730us/step - loss: 1.3167 - mean_squared_error: 1.3167\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.4315 - mean_squared_error: 0.4315\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.2554 - mean_squared_error: 0.2554\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.3552 - mean_squared_error: 0.3552\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 358us/step - loss: 0.1319 - mean_squared_error: 0.1319\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.1489 - mean_squared_error: 0.1489\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0618 - mean_squared_error: 0.0618\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0486 - mean_squared_error: 0.0486\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0474 - mean_squared_error: 0.0474\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 303us/step - loss: 0.0576 - mean_squared_error: 0.0576\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 295us/step - loss: 0.0466 - mean_squared_error: 0.0466\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 204us/step - loss: 0.0369 - mean_squared_error: 0.0369\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 283us/step - loss: 0.0378 - mean_squared_error: 0.0378\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0387 - mean_squared_error: 0.0387\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 253us/step - loss: 0.0281 - mean_squared_error: 0.0281\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0457 - mean_squared_error: 0.0457\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0315 - mean_squared_error: 0.0315\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 359us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 295us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0257 - mean_squared_error: 0.0257\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 285us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 297us/step - loss: 0.0245 - mean_squared_error: 0.0245\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0261 - mean_squared_error: 0.0261\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 359us/step - loss: 0.0258 - mean_squared_error: 0.0258\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0255 - mean_squared_error: 0.0255\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 337us/step - loss: 0.0206 - mean_squared_error: 0.0206\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 298us/step - loss: 0.0446 - mean_squared_error: 0.0446\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0381 - mean_squared_error: 0.0381\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0265 - mean_squared_error: 0.0265\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0309 - mean_squared_error: 0.0309\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0354 - mean_squared_error: 0.0354\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0247 - mean_squared_error: 0.0247\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0179 - mean_squared_error: 0.0179\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0182 - mean_squared_error: 0.0182\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0132 - mean_squared_error: 0.0132\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0136 - mean_squared_error: 0.0136\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0228 - mean_squared_error: 0.0228\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 321us/step - loss: 0.0237 - mean_squared_error: 0.0237\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0234 - mean_squared_error: 0.0234\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0172 - mean_squared_error: 0.0172\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0297 - mean_squared_error: 0.0297\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0291 - mean_squared_error: 0.0291\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0201 - mean_squared_error: 0.0201\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0129 - mean_squared_error: 0.0129\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0254 - mean_squared_error: 0.0254\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 1s 382us/step - loss: 0.0356 - mean_squared_error: 0.0356\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0330 - mean_squared_error: 0.0330\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 272us/step - loss: 0.0230 - mean_squared_error: 0.0230\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0162 - mean_squared_error: 0.0162\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 241us/step - loss: 0.0132 - mean_squared_error: 0.0132\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 1s 425us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - ETA: 0s - loss: 0.0157 - mean_squared_error: 0.01 - 1s 405us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 61/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 372us/step - loss: 0.0205 - mean_squared_error: 0.0205\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0257 - mean_squared_error: 0.0257\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 366us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0289 - mean_squared_error: 0.0289\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0248 - mean_squared_error: 0.0248\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0194 - mean_squared_error: 0.0194\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0151 - mean_squared_error: 0.0151\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 292us/step - loss: 0.0149 - mean_squared_error: 0.0149\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 338us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 344us/step - loss: 0.0091 - mean_squared_error: 0.0091\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0090 - mean_squared_error: 0.0090\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0130 - mean_squared_error: 0.0130\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 260us/step - loss: 0.0156 - mean_squared_error: 0.0156\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0174 - mean_squared_error: 0.0174\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0158 - mean_squared_error: 0.0158\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 312us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 333us/step - loss: 0.0353 - mean_squared_error: 0.0353\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 276us/step - loss: 0.0306 - mean_squared_error: 0.0306\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 277us/step - loss: 0.0486 - mean_squared_error: 0.0486\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 272us/step - loss: 0.0272 - mean_squared_error: 0.0272\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 295us/step - loss: 0.0204 - mean_squared_error: 0.0204\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 346us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 258us/step - loss: 0.0198 - mean_squared_error: 0.0198\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0219 - mean_squared_error: 0.0219\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0141 - mean_squared_error: 0.0141\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0201 - mean_squared_error: 0.0201\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0232 - mean_squared_error: 0.0232\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0285 - mean_squared_error: 0.0285\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0273 - mean_squared_error: 0.0273\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0343 - mean_squared_error: 0.0343\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 371us/step - loss: 0.0363 - mean_squared_error: 0.0363\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 365us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 370us/step - loss: 0.0370 - mean_squared_error: 0.0370\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 335us/step - loss: 0.0258 - mean_squared_error: 0.0258\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0191 - mean_squared_error: 0.0191\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 301us/step - loss: 0.0146 - mean_squared_error: 0.01460s - loss: 0.0142 - mean_squared_error\n",
      "146/146 [==============================] - 0s 1ms/step\n",
      "Epoch 1/100\n",
      "1314/1314 [==============================] - 1s 515us/step - loss: 1.1522 - mean_squared_error: 1.1522\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.3660 - mean_squared_error: 0.3660\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.4085 - mean_squared_error: 0.4085\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 321us/step - loss: 0.1629 - mean_squared_error: 0.1629\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.1062 - mean_squared_error: 0.1062\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.1064 - mean_squared_error: 0.1064\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0686 - mean_squared_error: 0.0686\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0606 - mean_squared_error: 0.0606\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0603 - mean_squared_error: 0.0603\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0521 - mean_squared_error: 0.0521\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 348us/step - loss: 0.0394 - mean_squared_error: 0.0394\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0417 - mean_squared_error: 0.0417\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0433 - mean_squared_error: 0.0433\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0332 - mean_squared_error: 0.0332\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0383 - mean_squared_error: 0.0383\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0315 - mean_squared_error: 0.0315\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0332 - mean_squared_error: 0.0332\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0263 - mean_squared_error: 0.0263\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0426 - mean_squared_error: 0.0426\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 332us/step - loss: 0.0383 - mean_squared_error: 0.0383\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0312 - mean_squared_error: 0.0312\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0234 - mean_squared_error: 0.0234\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 320us/step - loss: 0.0260 - mean_squared_error: 0.0260\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 272us/step - loss: 0.0268 - mean_squared_error: 0.0268\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 247us/step - loss: 0.0254 - mean_squared_error: 0.0254\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 282us/step - loss: 0.0274 - mean_squared_error: 0.0274\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 146us/step - loss: 0.0303 - mean_squared_error: 0.0303\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 135us/step - loss: 0.0251 - mean_squared_error: 0.0251\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 144us/step - loss: 0.0312 - mean_squared_error: 0.0312\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0229 - mean_squared_error: 0.0229\n",
      "Epoch 31/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0193 - mean_squared_error: 0.0193\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0111 - mean_squared_error: 0.0111\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0103 - mean_squared_error: 0.0103\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0216 - mean_squared_error: 0.0216\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 173us/step - loss: 0.0141 - mean_squared_error: 0.0141\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 166us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 168us/step - loss: 0.0171 - mean_squared_error: 0.0171\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 160us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0153 - mean_squared_error: 0.0153\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0123 - mean_squared_error: 0.0123\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0158 - mean_squared_error: 0.0158\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0151 - mean_squared_error: 0.0151\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0241 - mean_squared_error: 0.0241\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0155 - mean_squared_error: 0.0155\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0129 - mean_squared_error: 0.0129\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 229us/step - loss: 0.0124 - mean_squared_error: 0.0124\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 358us/step - loss: 0.0257 - mean_squared_error: 0.0257\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0168 - mean_squared_error: 0.0168\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0122 - mean_squared_error: 0.0122\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0106 - mean_squared_error: 0.0106\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0129 - mean_squared_error: 0.0129\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0238 - mean_squared_error: 0.0238\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0145 - mean_squared_error: 0.0145\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 324us/step - loss: 0.0260 - mean_squared_error: 0.0260\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 316us/step - loss: 0.0190 - mean_squared_error: 0.0190\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 265us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 265us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 257us/step - loss: 0.0122 - mean_squared_error: 0.0122\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 323us/step - loss: 0.0102 - mean_squared_error: 0.0102\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0104 - mean_squared_error: 0.0104\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 238us/step - loss: 0.0096 - mean_squared_error: 0.0096\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0225 - mean_squared_error: 0.0225\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 206us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 71/100\n",
      "1314/1314 [==============================] - 0s 303us/step - loss: 0.0161 - mean_squared_error: 0.0161\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0152 - mean_squared_error: 0.0152\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 299us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 249us/step - loss: 0.0135 - mean_squared_error: 0.0135\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 312us/step - loss: 0.0201 - mean_squared_error: 0.0201\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0200 - mean_squared_error: 0.0200\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0157 - mean_squared_error: 0.0157\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 219us/step - loss: 0.0260 - mean_squared_error: 0.0260\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 318us/step - loss: 0.0237 - mean_squared_error: 0.0237\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 300us/step - loss: 0.0184 - mean_squared_error: 0.0184\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 189us/step - loss: 0.0164 - mean_squared_error: 0.0164\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0142 - mean_squared_error: 0.0142\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 1s 398us/step - loss: 0.0142 - mean_squared_error: 0.0142\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 371us/step - loss: 0.0198 - mean_squared_error: 0.0198\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 375us/step - loss: 0.0181 - mean_squared_error: 0.0181\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 375us/step - loss: 0.0209 - mean_squared_error: 0.0209\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0185 - mean_squared_error: 0.0185\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 366us/step - loss: 0.0171 - mean_squared_error: 0.0171\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 347us/step - loss: 0.0146 - mean_squared_error: 0.0146\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0182 - mean_squared_error: 0.0182\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0141 - mean_squared_error: 0.0141\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 329us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0155 - mean_squared_error: 0.0155\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0135 - mean_squared_error: 0.0135\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0203 - mean_squared_error: 0.0203\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 328us/step - loss: 0.0279 - mean_squared_error: 0.0279\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 301us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "146/146 [==============================] - 0s 1ms/step\n",
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 1s 614us/step - loss: 1.1758 - mean_squared_error: 1.1758\n",
      "Epoch 2/100\n",
      "1314/1314 [==============================] - 0s 298us/step - loss: 0.4589 - mean_squared_error: 0.4589\n",
      "Epoch 3/100\n",
      "1314/1314 [==============================] - 0s 367us/step - loss: 0.2527 - mean_squared_error: 0.2527\n",
      "Epoch 4/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.2135 - mean_squared_error: 0.2135\n",
      "Epoch 5/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.1281 - mean_squared_error: 0.1281\n",
      "Epoch 6/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0873 - mean_squared_error: 0.0873\n",
      "Epoch 7/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0789 - mean_squared_error: 0.0789\n",
      "Epoch 8/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.1050 - mean_squared_error: 0.1050\n",
      "Epoch 9/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0537 - mean_squared_error: 0.0537\n",
      "Epoch 10/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0431 - mean_squared_error: 0.0431\n",
      "Epoch 11/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0383 - mean_squared_error: 0.0383\n",
      "Epoch 12/100\n",
      "1314/1314 [==============================] - 0s 361us/step - loss: 0.0429 - mean_squared_error: 0.0429\n",
      "Epoch 13/100\n",
      "1314/1314 [==============================] - 0s 244us/step - loss: 0.0396 - mean_squared_error: 0.0396\n",
      "Epoch 14/100\n",
      "1314/1314 [==============================] - 0s 253us/step - loss: 0.0367 - mean_squared_error: 0.0367\n",
      "Epoch 15/100\n",
      "1314/1314 [==============================] - 0s 293us/step - loss: 0.0659 - mean_squared_error: 0.0659\n",
      "Epoch 16/100\n",
      "1314/1314 [==============================] - 0s 257us/step - loss: 0.0486 - mean_squared_error: 0.0486\n",
      "Epoch 17/100\n",
      "1314/1314 [==============================] - 0s 338us/step - loss: 0.0387 - mean_squared_error: 0.0387\n",
      "Epoch 18/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0412 - mean_squared_error: 0.0412\n",
      "Epoch 19/100\n",
      "1314/1314 [==============================] - 0s 277us/step - loss: 0.0446 - mean_squared_error: 0.0446\n",
      "Epoch 20/100\n",
      "1314/1314 [==============================] - 0s 332us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 21/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0212 - mean_squared_error: 0.0212\n",
      "Epoch 22/100\n",
      "1314/1314 [==============================] - 0s 238us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 23/100\n",
      "1314/1314 [==============================] - 0s 280us/step - loss: 0.0253 - mean_squared_error: 0.0253\n",
      "Epoch 24/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0244 - mean_squared_error: 0.0244\n",
      "Epoch 25/100\n",
      "1314/1314 [==============================] - 0s 307us/step - loss: 0.0235 - mean_squared_error: 0.0235\n",
      "Epoch 26/100\n",
      "1314/1314 [==============================] - 0s 326us/step - loss: 0.0272 - mean_squared_error: 0.0272\n",
      "Epoch 27/100\n",
      "1314/1314 [==============================] - 0s 322us/step - loss: 0.0214 - mean_squared_error: 0.0214\n",
      "Epoch 28/100\n",
      "1314/1314 [==============================] - 0s 276us/step - loss: 0.0295 - mean_squared_error: 0.0295\n",
      "Epoch 29/100\n",
      "1314/1314 [==============================] - 0s 240us/step - loss: 0.0330 - mean_squared_error: 0.0330\n",
      "Epoch 30/100\n",
      "1314/1314 [==============================] - 0s 309us/step - loss: 0.0319 - mean_squared_error: 0.0319\n",
      "Epoch 31/100\n",
      "1314/1314 [==============================] - 0s 293us/step - loss: 0.0405 - mean_squared_error: 0.0405\n",
      "Epoch 32/100\n",
      "1314/1314 [==============================] - 0s 333us/step - loss: 0.0298 - mean_squared_error: 0.0298\n",
      "Epoch 33/100\n",
      "1314/1314 [==============================] - 0s 245us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 34/100\n",
      "1314/1314 [==============================] - 0s 244us/step - loss: 0.0148 - mean_squared_error: 0.0148\n",
      "Epoch 35/100\n",
      "1314/1314 [==============================] - 0s 273us/step - loss: 0.0135 - mean_squared_error: 0.0135\n",
      "Epoch 36/100\n",
      "1314/1314 [==============================] - 0s 315us/step - loss: 0.0115 - mean_squared_error: 0.0115\n",
      "Epoch 37/100\n",
      "1314/1314 [==============================] - 0s 279us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "Epoch 38/100\n",
      "1314/1314 [==============================] - 0s 295us/step - loss: 0.0130 - mean_squared_error: 0.0130\n",
      "Epoch 39/100\n",
      "1314/1314 [==============================] - 0s 328us/step - loss: 0.0118 - mean_squared_error: 0.0118\n",
      "Epoch 40/100\n",
      "1314/1314 [==============================] - 0s 187us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 41/100\n",
      "1314/1314 [==============================] - 0s 137us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 42/100\n",
      "1314/1314 [==============================] - 0s 136us/step - loss: 0.0299 - mean_squared_error: 0.0299\n",
      "Epoch 43/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0345 - mean_squared_error: 0.0345\n",
      "Epoch 44/100\n",
      "1314/1314 [==============================] - 0s 159us/step - loss: 0.0364 - mean_squared_error: 0.0364\n",
      "Epoch 45/100\n",
      "1314/1314 [==============================] - 0s 156us/step - loss: 0.0196 - mean_squared_error: 0.0196\n",
      "Epoch 46/100\n",
      "1314/1314 [==============================] - 0s 148us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 47/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0204 - mean_squared_error: 0.0204\n",
      "Epoch 48/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0198 - mean_squared_error: 0.0198\n",
      "Epoch 49/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0757 - mean_squared_error: 0.0757\n",
      "Epoch 50/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0347 - mean_squared_error: 0.0347\n",
      "Epoch 51/100\n",
      "1314/1314 [==============================] - 0s 149us/step - loss: 0.0239 - mean_squared_error: 0.0239\n",
      "Epoch 52/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0217 - mean_squared_error: 0.0217\n",
      "Epoch 53/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0176 - mean_squared_error: 0.0176\n",
      "Epoch 54/100\n",
      "1314/1314 [==============================] - 0s 152us/step - loss: 0.0150 - mean_squared_error: 0.0150\n",
      "Epoch 55/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0233 - mean_squared_error: 0.0233\n",
      "Epoch 56/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0160 - mean_squared_error: 0.0160\n",
      "Epoch 57/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0139 - mean_squared_error: 0.0139\n",
      "Epoch 58/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0167 - mean_squared_error: 0.0167\n",
      "Epoch 59/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0186 - mean_squared_error: 0.0186\n",
      "Epoch 60/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0140 - mean_squared_error: 0.0140\n",
      "Epoch 61/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0133 - mean_squared_error: 0.0133\n",
      "Epoch 62/100\n",
      "1314/1314 [==============================] - 0s 151us/step - loss: 0.0109 - mean_squared_error: 0.0109\n",
      "Epoch 63/100\n",
      "1314/1314 [==============================] - 0s 150us/step - loss: 0.0087 - mean_squared_error: 0.0087\n",
      "Epoch 64/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0098 - mean_squared_error: 0.0098\n",
      "Epoch 65/100\n",
      "1314/1314 [==============================] - 0s 153us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 66/100\n",
      "1314/1314 [==============================] - 0s 198us/step - loss: 0.0163 - mean_squared_error: 0.0163\n",
      "Epoch 67/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0214 - mean_squared_error: 0.0214\n",
      "Epoch 68/100\n",
      "1314/1314 [==============================] - 0s 358us/step - loss: 0.0175 - mean_squared_error: 0.0175\n",
      "Epoch 69/100\n",
      "1314/1314 [==============================] - 0s 362us/step - loss: 0.0144 - mean_squared_error: 0.0144\n",
      "Epoch 70/100\n",
      "1314/1314 [==============================] - 0s 372us/step - loss: 0.0169 - mean_squared_error: 0.0169\n",
      "Epoch 71/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1314/1314 [==============================] - 0s 373us/step - loss: 0.0169 - mean_squared_error: 0.0169\n",
      "Epoch 72/100\n",
      "1314/1314 [==============================] - 1s 392us/step - loss: 0.0204 - mean_squared_error: 0.0204\n",
      "Epoch 73/100\n",
      "1314/1314 [==============================] - 0s 371us/step - loss: 0.0183 - mean_squared_error: 0.0183\n",
      "Epoch 74/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0238 - mean_squared_error: 0.0238\n",
      "Epoch 75/100\n",
      "1314/1314 [==============================] - 0s 359us/step - loss: 0.0191 - mean_squared_error: 0.0191\n",
      "Epoch 76/100\n",
      "1314/1314 [==============================] - 0s 363us/step - loss: 0.0189 - mean_squared_error: 0.0189\n",
      "Epoch 77/100\n",
      "1314/1314 [==============================] - 0s 353us/step - loss: 0.0287 - mean_squared_error: 0.0287\n",
      "Epoch 78/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0252 - mean_squared_error: 0.0252\n",
      "Epoch 79/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0232 - mean_squared_error: 0.0232\n",
      "Epoch 80/100\n",
      "1314/1314 [==============================] - 0s 342us/step - loss: 0.0190 - mean_squared_error: 0.0190\n",
      "Epoch 81/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0158 - mean_squared_error: 0.0158\n",
      "Epoch 82/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0177 - mean_squared_error: 0.0177\n",
      "Epoch 83/100\n",
      "1314/1314 [==============================] - 0s 369us/step - loss: 0.0302 - mean_squared_error: 0.0302\n",
      "Epoch 84/100\n",
      "1314/1314 [==============================] - 0s 371us/step - loss: 0.0264 - mean_squared_error: 0.0264\n",
      "Epoch 85/100\n",
      "1314/1314 [==============================] - 0s 284us/step - loss: 0.0234 - mean_squared_error: 0.0234\n",
      "Epoch 86/100\n",
      "1314/1314 [==============================] - 0s 339us/step - loss: 0.0193 - mean_squared_error: 0.0193\n",
      "Epoch 87/100\n",
      "1314/1314 [==============================] - 0s 263us/step - loss: 0.0197 - mean_squared_error: 0.0197\n",
      "Epoch 88/100\n",
      "1314/1314 [==============================] - 0s 251us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "Epoch 89/100\n",
      "1314/1314 [==============================] - 0s 325us/step - loss: 0.0217 - mean_squared_error: 0.0217\n",
      "Epoch 90/100\n",
      "1314/1314 [==============================] - 0s 354us/step - loss: 0.0226 - mean_squared_error: 0.0226\n",
      "Epoch 91/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0178 - mean_squared_error: 0.0178\n",
      "Epoch 92/100\n",
      "1314/1314 [==============================] - 0s 360us/step - loss: 0.0190 - mean_squared_error: 0.0190\n",
      "Epoch 93/100\n",
      "1314/1314 [==============================] - 0s 352us/step - loss: 0.0223 - mean_squared_error: 0.0223\n",
      "Epoch 94/100\n",
      "1314/1314 [==============================] - 0s 355us/step - loss: 0.0181 - mean_squared_error: 0.0181\n",
      "Epoch 95/100\n",
      "1314/1314 [==============================] - 0s 356us/step - loss: 0.0179 - mean_squared_error: 0.0179\n",
      "Epoch 96/100\n",
      "1314/1314 [==============================] - 0s 357us/step - loss: 0.0156 - mean_squared_error: 0.0156\n",
      "Epoch 97/100\n",
      "1314/1314 [==============================] - 0s 345us/step - loss: 0.0151 - mean_squared_error: 0.0151\n",
      "Epoch 98/100\n",
      "1314/1314 [==============================] - 0s 349us/step - loss: 0.0275 - mean_squared_error: 0.0275\n",
      "Epoch 99/100\n",
      "1314/1314 [==============================] - 0s 351us/step - loss: 0.0279 - mean_squared_error: 0.0279\n",
      "Epoch 100/100\n",
      "1314/1314 [==============================] - 0s 350us/step - loss: 0.0261 - mean_squared_error: 0.0261\n",
      "146/146 [==============================] - 0s 2ms/step\n",
      "Standardized: -0.19 (0.10) MSE\n"
     ]
    }
   ],
   "source": [
    "regressor = KerasRegressor(build_fn=get_model, epochs=100, verbose=1)\n",
    "kfold = KFold(n_splits=10)\n",
    "results = cross_val_score(regressor, data, data_y, cv=kfold)\n",
    "print(\"Standardized: %.2f (%.2f) MSE\" % (results.mean(), results.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1168 samples, validate on 292 samples\n",
      "Epoch 1/100\n",
      "1168/1168 [==============================] - 1s 885us/step - loss: 1.3620 - mean_squared_error: 1.3620 - val_loss: 0.7326 - val_mean_squared_error: 0.7326\n",
      "Epoch 2/100\n",
      "1168/1168 [==============================] - 0s 250us/step - loss: 0.4860 - mean_squared_error: 0.4860 - val_loss: 0.4198 - val_mean_squared_error: 0.4198\n",
      "Epoch 3/100\n",
      "1168/1168 [==============================] - 0s 348us/step - loss: 0.2821 - mean_squared_error: 0.2821 - val_loss: 0.3691 - val_mean_squared_error: 0.3691\n",
      "Epoch 4/100\n",
      "1168/1168 [==============================] - 0s 374us/step - loss: 0.1725 - mean_squared_error: 0.1725 - val_loss: 0.2806 - val_mean_squared_error: 0.2806\n",
      "Epoch 5/100\n",
      "1168/1168 [==============================] - 0s 343us/step - loss: 0.1186 - mean_squared_error: 0.1186 - val_loss: 0.3006 - val_mean_squared_error: 0.3006\n",
      "Epoch 6/100\n",
      "1168/1168 [==============================] - 0s 341us/step - loss: 0.0899 - mean_squared_error: 0.0899 - val_loss: 0.2876 - val_mean_squared_error: 0.2876\n",
      "Epoch 7/100\n",
      "1168/1168 [==============================] - 0s 210us/step - loss: 0.0743 - mean_squared_error: 0.0743 - val_loss: 0.2484 - val_mean_squared_error: 0.2484\n",
      "Epoch 8/100\n",
      "1168/1168 [==============================] - 0s 144us/step - loss: 0.0575 - mean_squared_error: 0.0575 - val_loss: 0.2967 - val_mean_squared_error: 0.2967\n",
      "Epoch 9/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0506 - mean_squared_error: 0.0506 - val_loss: 0.2854 - val_mean_squared_error: 0.2854\n",
      "Epoch 10/100\n",
      "1168/1168 [==============================] - 0s 158us/step - loss: 0.0464 - mean_squared_error: 0.0464 - val_loss: 0.2516 - val_mean_squared_error: 0.2516\n",
      "Epoch 11/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0398 - mean_squared_error: 0.0398 - val_loss: 0.2665 - val_mean_squared_error: 0.2665\n",
      "Epoch 12/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0365 - mean_squared_error: 0.0365 - val_loss: 0.2535 - val_mean_squared_error: 0.2535\n",
      "Epoch 13/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0331 - mean_squared_error: 0.0331 - val_loss: 0.2531 - val_mean_squared_error: 0.2531\n",
      "Epoch 14/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0349 - mean_squared_error: 0.0349 - val_loss: 0.2478 - val_mean_squared_error: 0.2478\n",
      "Epoch 15/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0317 - mean_squared_error: 0.0317 - val_loss: 0.2654 - val_mean_squared_error: 0.2654\n",
      "Epoch 16/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0354 - mean_squared_error: 0.0354 - val_loss: 0.2648 - val_mean_squared_error: 0.2648\n",
      "Epoch 17/100\n",
      "1168/1168 [==============================] - 0s 163us/step - loss: 0.0340 - mean_squared_error: 0.0340 - val_loss: 0.2387 - val_mean_squared_error: 0.2387\n",
      "Epoch 18/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0317 - mean_squared_error: 0.0317 - val_loss: 0.2423 - val_mean_squared_error: 0.2423\n",
      "Epoch 19/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0316 - mean_squared_error: 0.0316 - val_loss: 0.2808 - val_mean_squared_error: 0.2808\n",
      "Epoch 20/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0281 - mean_squared_error: 0.0281 - val_loss: 0.2399 - val_mean_squared_error: 0.2399\n",
      "Epoch 21/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0228 - mean_squared_error: 0.0228 - val_loss: 0.2695 - val_mean_squared_error: 0.2695\n",
      "Epoch 22/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0232 - mean_squared_error: 0.0232 - val_loss: 0.2305 - val_mean_squared_error: 0.2305\n",
      "Epoch 23/100\n",
      "1168/1168 [==============================] - 0s 164us/step - loss: 0.0192 - mean_squared_error: 0.0192 - val_loss: 0.2470 - val_mean_squared_error: 0.2470\n",
      "Epoch 24/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0166 - mean_squared_error: 0.0166 - val_loss: 0.2382 - val_mean_squared_error: 0.2382\n",
      "Epoch 25/100\n",
      "1168/1168 [==============================] - 0s 165us/step - loss: 0.0203 - mean_squared_error: 0.0203 - val_loss: 0.2445 - val_mean_squared_error: 0.2445\n",
      "Epoch 26/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0198 - mean_squared_error: 0.0198 - val_loss: 0.2526 - val_mean_squared_error: 0.2526\n",
      "Epoch 27/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0175 - mean_squared_error: 0.0175 - val_loss: 0.2482 - val_mean_squared_error: 0.2482\n",
      "Epoch 28/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0230 - mean_squared_error: 0.0230 - val_loss: 0.2553 - val_mean_squared_error: 0.2553\n",
      "Epoch 29/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0185 - mean_squared_error: 0.0185 - val_loss: 0.2366 - val_mean_squared_error: 0.2366\n",
      "Epoch 30/100\n",
      "1168/1168 [==============================] - 0s 166us/step - loss: 0.0157 - mean_squared_error: 0.0157 - val_loss: 0.2447 - val_mean_squared_error: 0.2447\n",
      "Epoch 31/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0182 - mean_squared_error: 0.0182 - val_loss: 0.2315 - val_mean_squared_error: 0.2315\n",
      "Epoch 32/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0159 - mean_squared_error: 0.0159 - val_loss: 0.2431 - val_mean_squared_error: 0.2431\n",
      "Epoch 33/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0218 - mean_squared_error: 0.0218 - val_loss: 0.2450 - val_mean_squared_error: 0.2450\n",
      "Epoch 34/100\n",
      "1168/1168 [==============================] - 0s 369us/step - loss: 0.0174 - mean_squared_error: 0.0174 - val_loss: 0.2439 - val_mean_squared_error: 0.2439\n",
      "Epoch 35/100\n",
      "1168/1168 [==============================] - 0s 373us/step - loss: 0.0196 - mean_squared_error: 0.0196 - val_loss: 0.2455 - val_mean_squared_error: 0.2455\n",
      "Epoch 36/100\n",
      "1168/1168 [==============================] - 0s 378us/step - loss: 0.0179 - mean_squared_error: 0.0179 - val_loss: 0.2470 - val_mean_squared_error: 0.2470\n",
      "Epoch 37/100\n",
      "1168/1168 [==============================] - 0s 389us/step - loss: 0.0189 - mean_squared_error: 0.0189 - val_loss: 0.2390 - val_mean_squared_error: 0.2390\n",
      "Epoch 38/100\n",
      "1168/1168 [==============================] - 0s 379us/step - loss: 0.0179 - mean_squared_error: 0.0179 - val_loss: 0.2498 - val_mean_squared_error: 0.2498\n",
      "Epoch 39/100\n",
      "1168/1168 [==============================] - 0s 383us/step - loss: 0.0182 - mean_squared_error: 0.0182 - val_loss: 0.2441 - val_mean_squared_error: 0.2441\n",
      "Epoch 40/100\n",
      "1168/1168 [==============================] - 0s 301us/step - loss: 0.0173 - mean_squared_error: 0.0173 - val_loss: 0.2428 - val_mean_squared_error: 0.2428\n",
      "Epoch 41/100\n",
      "1168/1168 [==============================] - 0s 146us/step - loss: 0.0165 - mean_squared_error: 0.0165 - val_loss: 0.2445 - val_mean_squared_error: 0.2445\n",
      "Epoch 42/100\n",
      "1168/1168 [==============================] - 0s 150us/step - loss: 0.0198 - mean_squared_error: 0.0198 - val_loss: 0.2345 - val_mean_squared_error: 0.2345\n",
      "Epoch 43/100\n",
      "1168/1168 [==============================] - 0s 151us/step - loss: 0.0175 - mean_squared_error: 0.0175 - val_loss: 0.2312 - val_mean_squared_error: 0.2312\n",
      "Epoch 44/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0150 - mean_squared_error: 0.0150 - val_loss: 0.2434 - val_mean_squared_error: 0.2434\n",
      "Epoch 45/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0143 - mean_squared_error: 0.0143 - val_loss: 0.2332 - val_mean_squared_error: 0.2332\n",
      "Epoch 46/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0134 - mean_squared_error: 0.0134 - val_loss: 0.2486 - val_mean_squared_error: 0.2486\n",
      "Epoch 47/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0111 - mean_squared_error: 0.0111 - val_loss: 0.2338 - val_mean_squared_error: 0.2338\n",
      "Epoch 48/100\n",
      "1168/1168 [==============================] - 0s 165us/step - loss: 0.0131 - mean_squared_error: 0.0131 - val_loss: 0.2293 - val_mean_squared_error: 0.2293\n",
      "Epoch 49/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1168/1168 [==============================] - 0s 163us/step - loss: 0.0137 - mean_squared_error: 0.0137 - val_loss: 0.2237 - val_mean_squared_error: 0.2237\n",
      "Epoch 50/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0148 - mean_squared_error: 0.0148 - val_loss: 0.2501 - val_mean_squared_error: 0.2501\n",
      "Epoch 51/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0129 - mean_squared_error: 0.0129 - val_loss: 0.2438 - val_mean_squared_error: 0.2438\n",
      "Epoch 52/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0152 - mean_squared_error: 0.0152 - val_loss: 0.2349 - val_mean_squared_error: 0.2349\n",
      "Epoch 53/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0152 - mean_squared_error: 0.0152 - val_loss: 0.2439 - val_mean_squared_error: 0.2439\n",
      "Epoch 54/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0121 - mean_squared_error: 0.0121 - val_loss: 0.2444 - val_mean_squared_error: 0.2444\n",
      "Epoch 55/100\n",
      "1168/1168 [==============================] - 0s 166us/step - loss: 0.0100 - mean_squared_error: 0.0100 - val_loss: 0.2506 - val_mean_squared_error: 0.2506\n",
      "Epoch 56/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0115 - mean_squared_error: 0.0115 - val_loss: 0.2507 - val_mean_squared_error: 0.2507\n",
      "Epoch 57/100\n",
      "1168/1168 [==============================] - 0s 160us/step - loss: 0.0129 - mean_squared_error: 0.0129 - val_loss: 0.2509 - val_mean_squared_error: 0.2509\n",
      "Epoch 58/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0147 - mean_squared_error: 0.0147 - val_loss: 0.2502 - val_mean_squared_error: 0.2502\n",
      "Epoch 59/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0126 - mean_squared_error: 0.0126 - val_loss: 0.2407 - val_mean_squared_error: 0.2407\n",
      "Epoch 60/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0151 - mean_squared_error: 0.0151 - val_loss: 0.2613 - val_mean_squared_error: 0.2613\n",
      "Epoch 61/100\n",
      "1168/1168 [==============================] - 0s 165us/step - loss: 0.0113 - mean_squared_error: 0.0113 - val_loss: 0.2427 - val_mean_squared_error: 0.2427\n",
      "Epoch 62/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0109 - mean_squared_error: 0.0109 - val_loss: 0.2461 - val_mean_squared_error: 0.2461\n",
      "Epoch 63/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0090 - mean_squared_error: 0.0090 - val_loss: 0.2551 - val_mean_squared_error: 0.2551\n",
      "Epoch 64/100\n",
      "1168/1168 [==============================] - 0s 165us/step - loss: 0.0094 - mean_squared_error: 0.0094 - val_loss: 0.2341 - val_mean_squared_error: 0.2341\n",
      "Epoch 65/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0086 - mean_squared_error: 0.0086 - val_loss: 0.2474 - val_mean_squared_error: 0.2474\n",
      "Epoch 66/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0111 - mean_squared_error: 0.0111 - val_loss: 0.2375 - val_mean_squared_error: 0.2375\n",
      "Epoch 67/100\n",
      "1168/1168 [==============================] - 0s 194us/step - loss: 0.0108 - mean_squared_error: 0.0108 - val_loss: 0.2417 - val_mean_squared_error: 0.2417\n",
      "Epoch 68/100\n",
      "1168/1168 [==============================] - 0s 382us/step - loss: 0.0094 - mean_squared_error: 0.0094 - val_loss: 0.2394 - val_mean_squared_error: 0.2394\n",
      "Epoch 69/100\n",
      "1168/1168 [==============================] - 0s 380us/step - loss: 0.0108 - mean_squared_error: 0.0108 - val_loss: 0.2462 - val_mean_squared_error: 0.2462\n",
      "Epoch 70/100\n",
      "1168/1168 [==============================] - 0s 394us/step - loss: 0.0136 - mean_squared_error: 0.0136 - val_loss: 0.2534 - val_mean_squared_error: 0.2534\n",
      "Epoch 71/100\n",
      "1168/1168 [==============================] - 1s 432us/step - loss: 0.0124 - mean_squared_error: 0.0124 - val_loss: 0.2454 - val_mean_squared_error: 0.2454\n",
      "Epoch 72/100\n",
      "1168/1168 [==============================] - 0s 402us/step - loss: 0.0117 - mean_squared_error: 0.0117 - val_loss: 0.2475 - val_mean_squared_error: 0.2475\n",
      "Epoch 73/100\n",
      "1168/1168 [==============================] - 0s 400us/step - loss: 0.0112 - mean_squared_error: 0.0112 - val_loss: 0.2434 - val_mean_squared_error: 0.2434\n",
      "Epoch 74/100\n",
      "1168/1168 [==============================] - 0s 412us/step - loss: 0.0116 - mean_squared_error: 0.0116 - val_loss: 0.2381 - val_mean_squared_error: 0.2381\n",
      "Epoch 75/100\n",
      "1168/1168 [==============================] - 0s 390us/step - loss: 0.0139 - mean_squared_error: 0.0139 - val_loss: 0.2396 - val_mean_squared_error: 0.2396\n",
      "Epoch 76/100\n",
      "1168/1168 [==============================] - 0s 393us/step - loss: 0.0146 - mean_squared_error: 0.0146 - val_loss: 0.2375 - val_mean_squared_error: 0.2375\n",
      "Epoch 77/100\n",
      "1168/1168 [==============================] - 0s 380us/step - loss: 0.0146 - mean_squared_error: 0.0146 - val_loss: 0.2422 - val_mean_squared_error: 0.2422\n",
      "Epoch 78/100\n",
      "1168/1168 [==============================] - 0s 376us/step - loss: 0.0178 - mean_squared_error: 0.0178 - val_loss: 0.2538 - val_mean_squared_error: 0.2538\n",
      "Epoch 79/100\n",
      "1168/1168 [==============================] - 0s 357us/step - loss: 0.0233 - mean_squared_error: 0.0233 - val_loss: 0.2565 - val_mean_squared_error: 0.2565\n",
      "Epoch 80/100\n",
      "1168/1168 [==============================] - 0s 377us/step - loss: 0.0245 - mean_squared_error: 0.0245 - val_loss: 0.2561 - val_mean_squared_error: 0.2561\n",
      "Epoch 81/100\n",
      "1168/1168 [==============================] - 0s 283us/step - loss: 0.0187 - mean_squared_error: 0.0187 - val_loss: 0.2437 - val_mean_squared_error: 0.2437\n",
      "Epoch 82/100\n",
      "1168/1168 [==============================] - 0s 342us/step - loss: 0.0177 - mean_squared_error: 0.0177 - val_loss: 0.2514 - val_mean_squared_error: 0.2514\n",
      "Epoch 83/100\n",
      "1168/1168 [==============================] - 0s 355us/step - loss: 0.0137 - mean_squared_error: 0.0137 - val_loss: 0.2406 - val_mean_squared_error: 0.2406\n",
      "Epoch 84/100\n",
      "1168/1168 [==============================] - 0s 296us/step - loss: 0.0105 - mean_squared_error: 0.0105 - val_loss: 0.2535 - val_mean_squared_error: 0.2535\n",
      "Epoch 85/100\n",
      "1168/1168 [==============================] - 0s 267us/step - loss: 0.0103 - mean_squared_error: 0.0103 - val_loss: 0.2396 - val_mean_squared_error: 0.2396\n",
      "Epoch 86/100\n",
      "1168/1168 [==============================] - 0s 301us/step - loss: 0.0124 - mean_squared_error: 0.0124 - val_loss: 0.2370 - val_mean_squared_error: 0.2370\n",
      "Epoch 87/100\n",
      "1168/1168 [==============================] - 0s 352us/step - loss: 0.0159 - mean_squared_error: 0.0159 - val_loss: 0.2347 - val_mean_squared_error: 0.2347\n",
      "Epoch 88/100\n",
      "1168/1168 [==============================] - 0s 143us/step - loss: 0.0166 - mean_squared_error: 0.0166 - val_loss: 0.2366 - val_mean_squared_error: 0.2366\n",
      "Epoch 89/100\n",
      "1168/1168 [==============================] - 0s 145us/step - loss: 0.0179 - mean_squared_error: 0.0179 - val_loss: 0.2399 - val_mean_squared_error: 0.2399\n",
      "Epoch 90/100\n",
      "1168/1168 [==============================] - 0s 149us/step - loss: 0.0154 - mean_squared_error: 0.0154 - val_loss: 0.2387 - val_mean_squared_error: 0.2387\n",
      "Epoch 91/100\n",
      "1168/1168 [==============================] - 0s 164us/step - loss: 0.0156 - mean_squared_error: 0.0156 - val_loss: 0.2393 - val_mean_squared_error: 0.2393\n",
      "Epoch 92/100\n",
      "1168/1168 [==============================] - 0s 164us/step - loss: 0.0148 - mean_squared_error: 0.0148 - val_loss: 0.2521 - val_mean_squared_error: 0.2521\n",
      "Epoch 93/100\n",
      "1168/1168 [==============================] - 0s 161us/step - loss: 0.0122 - mean_squared_error: 0.0122 - val_loss: 0.2356 - val_mean_squared_error: 0.2356\n",
      "Epoch 94/100\n",
      "1168/1168 [==============================] - 0s 167us/step - loss: 0.0147 - mean_squared_error: 0.0147 - val_loss: 0.2578 - val_mean_squared_error: 0.2578\n",
      "Epoch 95/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0142 - mean_squared_error: 0.0142 - val_loss: 0.2462 - val_mean_squared_error: 0.2462\n",
      "Epoch 96/100\n",
      "1168/1168 [==============================] - 0s 168us/step - loss: 0.0129 - mean_squared_error: 0.0129 - val_loss: 0.2381 - val_mean_squared_error: 0.2381\n",
      "Epoch 97/100\n",
      "1168/1168 [==============================] - 0s 164us/step - loss: 0.0122 - mean_squared_error: 0.0122 - val_loss: 0.2474 - val_mean_squared_error: 0.2474\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 98/100\n",
      "1168/1168 [==============================] - 0s 159us/step - loss: 0.0147 - mean_squared_error: 0.0147 - val_loss: 0.2445 - val_mean_squared_error: 0.2445\n",
      "Epoch 99/100\n",
      "1168/1168 [==============================] - 0s 163us/step - loss: 0.0131 - mean_squared_error: 0.0131 - val_loss: 0.2479 - val_mean_squared_error: 0.2479\n",
      "Epoch 100/100\n",
      "1168/1168 [==============================] - 0s 162us/step - loss: 0.0148 - mean_squared_error: 0.0148 - val_loss: 0.2418 - val_mean_squared_error: 0.2418\n"
     ]
    }
   ],
   "source": [
    "regressor = KerasRegressor(build_fn=get_model, epochs=100, verbose=1)\n",
    "h = regressor.fit(data, data_y, epochs=100,  verbose=1, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1710c6020b8>]"
      ]
     },
     "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": [
    "plt.plot(h.history['mean_squared_error'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1460/1460 [==============================] - 0s 125us/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3600x2880 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(50, 40))\n",
    "predictions = regressor.predict(data)\n",
    "reality = data_y\n",
    "plt.style.use('ggplot')\n",
    "plt.plot(predictions, reality, 'ro')\n",
    "plt.xlabel('Predictions', fontsize = 30)\n",
    "plt.ylabel('Reality', fontsize = 30)\n",
    "plt.title('Predictions x Reality on dataset Test', fontsize = 30)\n",
    "ax.plot([reality.min(), reality.max()], [reality.min(), reality.max()], 'k--', lw=4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.13549667387248993\n",
      "0.061987082987087\n",
      "0.938012917012913\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n",
    "\n",
    "print(mean_absolute_error(data_y,predictions))\n",
    "print(mean_squared_error(data_y,predictions))\n",
    "print(r2_score(data_y,predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>LotConfig</th>\n",
       "      <th>LandSlope</th>\n",
       "      <th>Neighborhood</th>\n",
       "      <th>Condition1</th>\n",
       "      <th>Condition2</th>\n",
       "      <th>BldgType</th>\n",
       "      <th>HouseStyle</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>RoofStyle</th>\n",
       "      <th>RoofMatl</th>\n",
       "      <th>Exterior1st</th>\n",
       "      <th>Exterior2nd</th>\n",
       "      <th>MasVnrType</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>ExterCond</th>\n",
       "      <th>Foundation</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>BsmtCond</th>\n",
       "      <th>BsmtExposure</th>\n",
       "      <th>BsmtFinType1</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinType2</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <th>Heating</th>\n",
       "      <th>HeatingQC</th>\n",
       "      <th>CentralAir</th>\n",
       "      <th>Electrical</th>\n",
       "      <th>1stFlrSF</th>\n",
       "      <th>2ndFlrSF</th>\n",
       "      <th>LowQualFinSF</th>\n",
       "      <th>GrLivArea</th>\n",
       "      <th>BsmtFullBath</th>\n",
       "      <th>BsmtHalfBath</th>\n",
       "      <th>FullBath</th>\n",
       "      <th>HalfBath</th>\n",
       "      <th>BedroomAbvGr</th>\n",
       "      <th>KitchenAbvGr</th>\n",
       "      <th>KitchenQual</th>\n",
       "      <th>TotRmsAbvGrd</th>\n",
       "      <th>Functional</th>\n",
       "      <th>Fireplaces</th>\n",
       "      <th>FireplaceQu</th>\n",
       "      <th>GarageType</th>\n",
       "      <th>GarageYrBlt</th>\n",
       "      <th>GarageFinish</th>\n",
       "      <th>GarageCars</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>GarageQual</th>\n",
       "      <th>GarageCond</th>\n",
       "      <th>PavedDrive</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1461</td>\n",
       "      <td>20</td>\n",
       "      <td>RH</td>\n",
       "      <td>80.0</td>\n",
       "      <td>11622</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>NAmes</td>\n",
       "      <td>Feedr</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>1Story</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>1961</td>\n",
       "      <td>1961</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>CBlock</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>Rec</td>\n",
       "      <td>468.0</td>\n",
       "      <td>LwQ</td>\n",
       "      <td>144.0</td>\n",
       "      <td>270.0</td>\n",
       "      <td>882.0</td>\n",
       "      <td>GasA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>896</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>896</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>5</td>\n",
       "      <td>Typ</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1961.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>1.0</td>\n",
       "      <td>730.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>140</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MnPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1462</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>81.0</td>\n",
       "      <td>14267</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Corner</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>NAmes</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>1Story</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>1958</td>\n",
       "      <td>1958</td>\n",
       "      <td>Hip</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>Wd Sdng</td>\n",
       "      <td>Wd Sdng</td>\n",
       "      <td>BrkFace</td>\n",
       "      <td>108.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>CBlock</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>ALQ</td>\n",
       "      <td>923.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0.0</td>\n",
       "      <td>406.0</td>\n",
       "      <td>1329.0</td>\n",
       "      <td>GasA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>1329</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1329</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>6</td>\n",
       "      <td>Typ</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1958.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>1.0</td>\n",
       "      <td>312.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>393</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Gar2</td>\n",
       "      <td>12500</td>\n",
       "      <td>6</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1463</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>74.0</td>\n",
       "      <td>13830</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>Gilbert</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1997</td>\n",
       "      <td>1998</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>GLQ</td>\n",
       "      <td>791.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0.0</td>\n",
       "      <td>137.0</td>\n",
       "      <td>928.0</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Gd</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>928</td>\n",
       "      <td>701</td>\n",
       "      <td>0</td>\n",
       "      <td>1629</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>6</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>TA</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1997.0</td>\n",
       "      <td>Fin</td>\n",
       "      <td>2.0</td>\n",
       "      <td>482.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>212</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MnPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1464</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>78.0</td>\n",
       "      <td>9978</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>Gilbert</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>2Story</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>1998</td>\n",
       "      <td>1998</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>BrkFace</td>\n",
       "      <td>20.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>GLQ</td>\n",
       "      <td>602.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0.0</td>\n",
       "      <td>324.0</td>\n",
       "      <td>926.0</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>926</td>\n",
       "      <td>678</td>\n",
       "      <td>0</td>\n",
       "      <td>1604</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>7</td>\n",
       "      <td>Typ</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1998.0</td>\n",
       "      <td>Fin</td>\n",
       "      <td>2.0</td>\n",
       "      <td>470.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>360</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1465</td>\n",
       "      <td>120</td>\n",
       "      <td>RL</td>\n",
       "      <td>43.0</td>\n",
       "      <td>5005</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>HLS</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>StoneBr</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>TwnhsE</td>\n",
       "      <td>1Story</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>1992</td>\n",
       "      <td>1992</td>\n",
       "      <td>Gable</td>\n",
       "      <td>CompShg</td>\n",
       "      <td>HdBoard</td>\n",
       "      <td>HdBoard</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>PConc</td>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>No</td>\n",
       "      <td>ALQ</td>\n",
       "      <td>263.0</td>\n",
       "      <td>Unf</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1017.0</td>\n",
       "      <td>1280.0</td>\n",
       "      <td>GasA</td>\n",
       "      <td>Ex</td>\n",
       "      <td>Y</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>1280</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1280</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>Gd</td>\n",
       "      <td>5</td>\n",
       "      <td>Typ</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>1992.0</td>\n",
       "      <td>RFn</td>\n",
       "      <td>2.0</td>\n",
       "      <td>506.0</td>\n",
       "      <td>TA</td>\n",
       "      <td>TA</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>144</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Id  MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0  1461          20       RH         80.0    11622   Pave   NaN      Reg   \n",
       "1  1462          20       RL         81.0    14267   Pave   NaN      IR1   \n",
       "2  1463          60       RL         74.0    13830   Pave   NaN      IR1   \n",
       "3  1464          60       RL         78.0     9978   Pave   NaN      IR1   \n",
       "4  1465         120       RL         43.0     5005   Pave   NaN      IR1   \n",
       "\n",
       "  LandContour Utilities LotConfig LandSlope Neighborhood Condition1  \\\n",
       "0         Lvl    AllPub    Inside       Gtl        NAmes      Feedr   \n",
       "1         Lvl    AllPub    Corner       Gtl        NAmes       Norm   \n",
       "2         Lvl    AllPub    Inside       Gtl      Gilbert       Norm   \n",
       "3         Lvl    AllPub    Inside       Gtl      Gilbert       Norm   \n",
       "4         HLS    AllPub    Inside       Gtl      StoneBr       Norm   \n",
       "\n",
       "  Condition2 BldgType HouseStyle  OverallQual  OverallCond  YearBuilt  \\\n",
       "0       Norm     1Fam     1Story            5            6       1961   \n",
       "1       Norm     1Fam     1Story            6            6       1958   \n",
       "2       Norm     1Fam     2Story            5            5       1997   \n",
       "3       Norm     1Fam     2Story            6            6       1998   \n",
       "4       Norm   TwnhsE     1Story            8            5       1992   \n",
       "\n",
       "   YearRemodAdd RoofStyle RoofMatl Exterior1st Exterior2nd MasVnrType  \\\n",
       "0          1961     Gable  CompShg     VinylSd     VinylSd       None   \n",
       "1          1958       Hip  CompShg     Wd Sdng     Wd Sdng    BrkFace   \n",
       "2          1998     Gable  CompShg     VinylSd     VinylSd       None   \n",
       "3          1998     Gable  CompShg     VinylSd     VinylSd    BrkFace   \n",
       "4          1992     Gable  CompShg     HdBoard     HdBoard       None   \n",
       "\n",
       "   MasVnrArea ExterQual ExterCond Foundation BsmtQual BsmtCond BsmtExposure  \\\n",
       "0         0.0        TA        TA     CBlock       TA       TA           No   \n",
       "1       108.0        TA        TA     CBlock       TA       TA           No   \n",
       "2         0.0        TA        TA      PConc       Gd       TA           No   \n",
       "3        20.0        TA        TA      PConc       TA       TA           No   \n",
       "4         0.0        Gd        TA      PConc       Gd       TA           No   \n",
       "\n",
       "  BsmtFinType1  BsmtFinSF1 BsmtFinType2  BsmtFinSF2  BsmtUnfSF  TotalBsmtSF  \\\n",
       "0          Rec       468.0          LwQ       144.0      270.0        882.0   \n",
       "1          ALQ       923.0          Unf         0.0      406.0       1329.0   \n",
       "2          GLQ       791.0          Unf         0.0      137.0        928.0   \n",
       "3          GLQ       602.0          Unf         0.0      324.0        926.0   \n",
       "4          ALQ       263.0          Unf         0.0     1017.0       1280.0   \n",
       "\n",
       "  Heating HeatingQC CentralAir Electrical  1stFlrSF  2ndFlrSF  LowQualFinSF  \\\n",
       "0    GasA        TA          Y      SBrkr       896         0             0   \n",
       "1    GasA        TA          Y      SBrkr      1329         0             0   \n",
       "2    GasA        Gd          Y      SBrkr       928       701             0   \n",
       "3    GasA        Ex          Y      SBrkr       926       678             0   \n",
       "4    GasA        Ex          Y      SBrkr      1280         0             0   \n",
       "\n",
       "   GrLivArea  BsmtFullBath  BsmtHalfBath  FullBath  HalfBath  BedroomAbvGr  \\\n",
       "0        896           0.0           0.0         1         0             2   \n",
       "1       1329           0.0           0.0         1         1             3   \n",
       "2       1629           0.0           0.0         2         1             3   \n",
       "3       1604           0.0           0.0         2         1             3   \n",
       "4       1280           0.0           0.0         2         0             2   \n",
       "\n",
       "   KitchenAbvGr KitchenQual  TotRmsAbvGrd Functional  Fireplaces FireplaceQu  \\\n",
       "0             1          TA             5        Typ           0         NaN   \n",
       "1             1          Gd             6        Typ           0         NaN   \n",
       "2             1          TA             6        Typ           1          TA   \n",
       "3             1          Gd             7        Typ           1          Gd   \n",
       "4             1          Gd             5        Typ           0         NaN   \n",
       "\n",
       "  GarageType  GarageYrBlt GarageFinish  GarageCars  GarageArea GarageQual  \\\n",
       "0     Attchd       1961.0          Unf         1.0       730.0         TA   \n",
       "1     Attchd       1958.0          Unf         1.0       312.0         TA   \n",
       "2     Attchd       1997.0          Fin         2.0       482.0         TA   \n",
       "3     Attchd       1998.0          Fin         2.0       470.0         TA   \n",
       "4     Attchd       1992.0          RFn         2.0       506.0         TA   \n",
       "\n",
       "  GarageCond PavedDrive  WoodDeckSF  OpenPorchSF  EnclosedPorch  3SsnPorch  \\\n",
       "0         TA          Y         140            0              0          0   \n",
       "1         TA          Y         393           36              0          0   \n",
       "2         TA          Y         212           34              0          0   \n",
       "3         TA          Y         360           36              0          0   \n",
       "4         TA          Y           0           82              0          0   \n",
       "\n",
       "   ScreenPorch  PoolArea PoolQC  Fence MiscFeature  MiscVal  MoSold  YrSold  \\\n",
       "0          120         0    NaN  MnPrv         NaN        0       6    2010   \n",
       "1            0         0    NaN    NaN        Gar2    12500       6    2010   \n",
       "2            0         0    NaN  MnPrv         NaN        0       3    2010   \n",
       "3            0         0    NaN    NaN         NaN        0       6    2010   \n",
       "4          144         0    NaN    NaN         NaN        0       1    2010   \n",
       "\n",
       "  SaleType SaleCondition  \n",
       "0       WD        Normal  \n",
       "1       WD        Normal  \n",
       "2       WD        Normal  \n",
       "3       WD        Normal  \n",
       "4       WD        Normal  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_test = pd.read_csv('.\\\\data\\\\6\\\\test\\\\test.csv')\n",
    "data_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MSZoning           4\n",
       "LotFrontage      227\n",
       "Alley           1352\n",
       "Utilities          2\n",
       "Exterior1st        1\n",
       "Exterior2nd        1\n",
       "MasVnrType        16\n",
       "MasVnrArea        15\n",
       "BsmtQual          44\n",
       "BsmtCond          45\n",
       "BsmtExposure      44\n",
       "BsmtFinType1      42\n",
       "BsmtFinSF1         1\n",
       "BsmtFinType2      42\n",
       "BsmtFinSF2         1\n",
       "BsmtUnfSF          1\n",
       "TotalBsmtSF        1\n",
       "BsmtFullBath       2\n",
       "BsmtHalfBath       2\n",
       "KitchenQual        1\n",
       "Functional         2\n",
       "FireplaceQu      730\n",
       "GarageType        76\n",
       "GarageYrBlt       78\n",
       "GarageFinish      78\n",
       "GarageCars         1\n",
       "GarageArea         1\n",
       "GarageQual        78\n",
       "GarageCond        78\n",
       "PoolQC          1456\n",
       "Fence           1169\n",
       "MiscFeature     1408\n",
       "SaleType           1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = data_test.isna().sum()\n",
    "df[df > 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pre_process_test(st, label, one, data):\n",
    "    data['MSSubClass'] = data['MSSubClass'] / 190\n",
    "    data['LotFrontage'].fillna(0, inplace=True)\n",
    "    data['Alley'].fillna('NA', inplace=True)\n",
    "    data['OverallQual'] = data['OverallQual'] / 10\n",
    "    data['OverallCond'] = data['OverallCond'] / 10\n",
    "    data['YearBuilt'] = data['YearBuilt'] / 1000\n",
    "    data['YearRemodAdd'] = data['YearRemodAdd'] /1000\n",
    "    data['MasVnrType'].fillna('NA', inplace=True)\n",
    "    data['MasVnrArea'].fillna(0, inplace=True)\n",
    "    data['ExterQual'] = data['ExterQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['ExterCond'] = data['ExterCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtQual'] = data['BsmtQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtQual'].fillna(0, inplace=True)\n",
    "    data['BsmtCond'] = data['BsmtCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['BsmtCond'].fillna(0, inplace=True)\n",
    "    data['BsmtExposure'] = data['BsmtExposure'].replace(['Gd', 'Av', 'Mn', 'No'], [1, 0.67, 0.34, 0]).astype(pd.np.float)\n",
    "    data['BsmtExposure'].fillna(0, inplace=True)\n",
    "    data['BsmtFinType1'] = data['BsmtFinType1'].replace(['GLQ', 'ALQ', 'BLQ', 'Rec', 'LwQ', 'Unf'], [1 ,0.80, 0.64, 0.48, 0.32, 0.16])\n",
    "    data['BsmtFinType1'].fillna(0, inplace=True)\n",
    "    data['BsmtFinType2'] = data['BsmtFinType2'].replace(['GLQ', 'ALQ', 'BLQ', 'Rec', 'LwQ', 'Unf'], [1 ,0.80, 0.64, 0.48, 0.32, 0.16])\n",
    "    data['BsmtFinType2'].fillna(0, inplace=True)\n",
    "    data['HeatingQC'] = data['HeatingQC'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['Electrical'].fillna('Mix', inplace=True)\n",
    "    data['KitchenQual'] = data['KitchenQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['Functional'] = data['Functional'].replace(['Typ', 'Min1', 'Min2', 'Mod', 'Maj1', 'Maj2', 'Sev', 'Sal'],\n",
    "                                                    [1, 0.875, 0.75, 0.625, 0.50, 0.325, 0.2, 0]).astype(pd.np.float)\n",
    "    data['FireplaceQu'] = data['FireplaceQu'].replace(['Ex', 'Gd', 'TA', 'Fa','Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['FireplaceQu'].fillna(0, inplace=True)\n",
    "    data['GarageType'].fillna('NA', inplace=True)\n",
    "    data['GarageYrBlt'] = data['GarageYrBlt'] / 1000\n",
    "    data['GarageFinish'] = data['GarageFinish'].replace(['Fin', 'RFn', 'Unf'], [1, 0.6, 0.3]).astype(pd.np.float)\n",
    "    data['GarageFinish'].fillna(0, inplace=True)\n",
    "    data['GarageQual'] = data['GarageQual'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['GarageQual'].fillna(0, inplace=True)\n",
    "    data['GarageCond'] = data['GarageCond'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['GarageCond'].fillna(0, inplace=True)\n",
    "    data['PavedDrive'] = data['PavedDrive'].replace(['Y', 'P', 'N'], [1, 0.5, 0]).astype(pd.np.float)\n",
    "    data['PoolQC'] = data['PoolQC'].replace(['Ex', 'Gd', 'TA', 'Fa', 'Po'], [1, 0.8, 0.6, 0.4, 0.2]).astype(pd.np.float)\n",
    "    data['PoolQC'].fillna(0, inplace=True)\n",
    "    data['Fence'] = data['Fence'].replace(['GdPrv', 'MnPrv', 'GdWo', 'MnWw'], [1, 0.75, 0.50, 0.25]).astype(pd.np.float)\n",
    "    data['Fence'].fillna(0, inplace=True)\n",
    "    data['MiscFeature'].fillna('NA', inplace=True)\n",
    "    #Extra\n",
    "    data['MSZoning'].fillna(data['MSZoning'].mode()[0], inplace=True)\n",
    "    data['Utilities'].fillna(data['Utilities'].mode()[0], inplace=True)\n",
    "    data['Exterior1st'].fillna(data['Exterior1st'].mode()[0], inplace=True)\n",
    "    data['Exterior2nd'].fillna(data['Exterior2nd'].mode()[0], inplace=True)\n",
    "    data['BsmtFinSF1'].fillna(0, inplace=True)\n",
    "    data['BsmtFinSF2'].fillna(0, inplace=True)\n",
    "    data['BsmtUnfSF'].fillna(0, inplace=True)\n",
    "    data['TotalBsmtSF'].fillna(0, inplace=True)\n",
    "    data['BsmtFullBath'].fillna(0, inplace=True)\n",
    "    data['BsmtHalfBath'].fillna(0, inplace=True)\n",
    "    data['KitchenQual'].fillna(0.6, inplace=True)\n",
    "    data['Functional'].fillna(1, inplace=True)\n",
    "    data['GarageCars'].fillna(0, inplace=True)\n",
    "    data['GarageArea'].fillna(0, inplace=True)\n",
    "    data['SaleType'].fillna(data['SaleType'].mode()[0], inplace=True)\n",
    "    data_num = data[['MSSubClass', 'LotFrontage', 'LotArea', 'OverallQual', 'OverallCond',\n",
    "                     'YearBuilt', 'YearRemodAdd', 'MasVnrArea', 'ExterQual', 'ExterCond',\n",
    "                     'BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinSF1', 'BsmtFinType2',\n",
    "                     'BsmtFinSF2', 'BsmtUnfSF', 'TotalBsmtSF', '1stFlrSF', '2ndFlrSF', 'LowQualFinSF', 'GrLivArea',\n",
    "                     'BsmtFullBath', 'BsmtHalfBath', 'FullBath', 'HalfBath', 'BedroomAbvGr', 'KitchenAbvGr', 'KitchenQual',\n",
    "                     'TotRmsAbvGrd', 'Functional', 'Fireplaces', 'FireplaceQu', 'GarageFinish', 'GarageCars',\n",
    "                     'GarageArea', 'GarageQual', 'GarageCond', 'PavedDrive', 'WoodDeckSF', 'OpenPorchSF', 'EnclosedPorch',\n",
    "                     '3SsnPorch', 'ScreenPorch', 'PoolArea', 'PoolQC', 'Fence', 'MiscVal']]\n",
    "    data_cat = data[['MSZoning', 'Street', 'Alley', 'LotShape', 'LandContour', 'Utilities',\n",
    "                     'LotConfig', 'LandSlope', 'Neighborhood', 'Condition1', 'Condition2',\n",
    "                     'BldgType', 'HouseStyle', 'RoofStyle', 'RoofMatl', 'Exterior1st', 'MasVnrType', 'Foundation',\n",
    "                     'Heating', 'CentralAir', 'Electrical', 'GarageType', 'MiscFeature', 'SaleType', 'SaleCondition']]\n",
    "    \n",
    "    label_fit = data_cat.apply(lambda x: label_dict[x.name].transform(x.astype(str)))\n",
    "    data_cat = one.transform(label_fit)\n",
    "    return st.transform(np.append(data_num.values, data_cat, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_x = pre_process_test(st, label_dict, one, data_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1459/1459 [==============================] - 0s 36us/step\n"
     ]
    }
   ],
   "source": [
    "data_y_test = regressor.predict(data_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data=sy.inverse_transform(data_y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Id</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1461</td>\n",
       "      <td>123295.37500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1462</td>\n",
       "      <td>170846.96875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1463</td>\n",
       "      <td>203811.78125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1464</td>\n",
       "      <td>206804.15625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1465</td>\n",
       "      <td>199481.50000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Id     SalePrice\n",
       "0  1461  123295.37500\n",
       "1  1462  170846.96875\n",
       "2  1463  203811.78125\n",
       "3  1464  206804.15625\n",
       "4  1465  199481.50000"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_test_i = data_test[['Id']]\n",
    "data_test_ans = pd.concat([data_test_i, df], axis=1)\n",
    "data_test_ans.columns = ['Id', 'SalePrice']\n",
    "data_test_ans.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_test_ans.to_csv(\"sub.csv\", index=False)"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}