Neural_decoding_DNN_random_spilt.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import required libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pip install seaborn\n",
    "# pip install xgboost\n",
    "# pip install keras\n",
    "# pip install sklearn\n",
    "# pip install pandas\n",
    "# pip install numpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  !pip install seaborn\n",
    "#  !pip install xgboost\n",
    "#  !pip install keras\n",
    "#  !pip install sklearn\n",
    "#  !pip install pandas\n",
    "#  !pip install numpy\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.callbacks import ModelCheckpoint\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Activation, Flatten\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import mean_absolute_error \n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sb\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings \n",
    "warnings.filterwarnings('ignore')\n",
    "warnings.filterwarnings('ignore', category=DeprecationWarning)\n",
    "from xgboost import XGBRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# help(pd.read_excel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_x = pd.read_excel('Research data sample.xlsx', sheet_name = 'Sheet1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_x = pd.read_excel('Research data.xlsx', sheet_name = 'Sheet1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Column1</th>\n",
       "      <th>Column2</th>\n",
       "      <th>Column3</th>\n",
       "      <th>Column4</th>\n",
       "      <th>Column5</th>\n",
       "      <th>Column6</th>\n",
       "      <th>Column7</th>\n",
       "      <th>Column8</th>\n",
       "      <th>Column9</th>\n",
       "      <th>Column10</th>\n",
       "      <th>...</th>\n",
       "      <th>Column12</th>\n",
       "      <th>Column13</th>\n",
       "      <th>Column14</th>\n",
       "      <th>Column15</th>\n",
       "      <th>Column16</th>\n",
       "      <th>Column17</th>\n",
       "      <th>Column18</th>\n",
       "      <th>Column19</th>\n",
       "      <th>Column20</th>\n",
       "      <th>Column21</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Column1  Column2  Column3  Column4  Column5  Column6  Column7  Column8  \\\n",
       "0        0        0        1        0        0        0        0        0   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "\n",
       "   Column9  Column10  ...  Column12  Column13  Column14  Column15  Column16  \\\n",
       "0        0         0  ...         0         0         0         0         0   \n",
       "1        0         0  ...         0         0         0         0         0   \n",
       "\n",
       "   Column17  Column18  Column19  Column20  Column21  \n",
       "0         0         0         0         0         0  \n",
       "1         0         0         0         0         1  \n",
       "\n",
       "[2 rows x 21 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_x.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Column1', 'Column2', 'Column3', 'Column4', 'Column5', 'Column6',\n",
       "       'Column7', 'Column8', 'Column9', 'Column10', 'Column11', 'Column12',\n",
       "       'Column13', 'Column14', 'Column15', 'Column16', 'Column17', 'Column18',\n",
       "       'Column19', 'Column20', 'Column21'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_x.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_y = pd.read_excel('Research data sample.xlsx', sheet_name = 'Sheet2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_y = pd.read_excel('Research data.xlsx', sheet_name = 'Sheet2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>Column1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>66.193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>66.193</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Column1\n",
       "0   66.193\n",
       "1   66.193"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_y.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_y = df_y.rename(columns={'Column1': 'Position'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Position</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>66.193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>66.193</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Position\n",
       "0    66.193\n",
       "1    66.193"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_y.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_all = df_x\n",
    "# df_all['Position'] = df_y['Position'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_all.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_kor = df.corr()\n",
    "plt.figure(figsize=(10,10))\n",
    "sns.heatmap(df_kor, vmin=-1, vmax=1, cmap=\"viridis\", annot=True, linewidth=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_x, test_x, train_y, test_y = train_test_split(df_x, df_y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Column1</th>\n",
       "      <th>Column2</th>\n",
       "      <th>Column3</th>\n",
       "      <th>Column4</th>\n",
       "      <th>Column5</th>\n",
       "      <th>Column6</th>\n",
       "      <th>Column7</th>\n",
       "      <th>Column8</th>\n",
       "      <th>Column9</th>\n",
       "      <th>Column10</th>\n",
       "      <th>...</th>\n",
       "      <th>Column12</th>\n",
       "      <th>Column13</th>\n",
       "      <th>Column14</th>\n",
       "      <th>Column15</th>\n",
       "      <th>Column16</th>\n",
       "      <th>Column17</th>\n",
       "      <th>Column18</th>\n",
       "      <th>Column19</th>\n",
       "      <th>Column20</th>\n",
       "      <th>Column21</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>339610</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>125963</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Column1  Column2  Column3  Column4  Column5  Column6  Column7  \\\n",
       "339610        0        0        0        0        0        1        0   \n",
       "125963        0        0        0        0        0        0        0   \n",
       "\n",
       "        Column8  Column9  Column10  ...  Column12  Column13  Column14  \\\n",
       "339610        0        0         0  ...         0         0         0   \n",
       "125963        0        0         0  ...         0         1         0   \n",
       "\n",
       "        Column15  Column16  Column17  Column18  Column19  Column20  Column21  \n",
       "339610         0         0         0         0         0         0         0  \n",
       "125963         0         0         0         0         0         0         0  \n",
       "\n",
       "[2 rows x 21 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "21"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.shape[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trying a Deep Neural Network"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Followed this blogpost: https://towardsdatascience.com/deep-neural-networks-for-regression-problems-81321897ca33"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_8 (Dense)              (None, 256)               5632      \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 128)               32896     \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 64)                8256      \n",
      "_________________________________________________________________\n",
      "dense_11 (Dense)             (None, 32)                2080      \n",
      "_________________________________________________________________\n",
      "dense_12 (Dense)             (None, 1)                 33        \n",
      "=================================================================\n",
      "Total params: 48,897\n",
      "Trainable params: 48,897\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "NN_model = Sequential()\n",
    "\n",
    "# The Input Layer :\n",
    "NN_model.add(Dense(256, kernel_initializer='normal',input_dim = train_x.shape[1], activation='relu'))\n",
    "\n",
    "# The Hidden Layers :\n",
    "NN_model.add(Dense(128, kernel_initializer='normal',activation='relu'))\n",
    "NN_model.add(Dense(64, kernel_initializer='normal',activation='relu'))\n",
    "NN_model.add(Dense(32, kernel_initializer='normal',activation='relu'))\n",
    "\n",
    "# The Output Layer :\n",
    "NN_model.add(Dense(1, kernel_initializer='normal',activation='linear'))\n",
    "\n",
    "# Compile the network :\n",
    "NN_model.compile(loss='mean_absolute_error', optimizer='adam', metrics=['mean_absolute_error'])\n",
    "NN_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "checkpoint_name = 'models/Weights-001--65.91199.hdf5' \n",
    "checkpoint = ModelCheckpoint(checkpoint_name, monitor='val_loss', verbose = 1, save_best_only = True, mode ='auto')\n",
    "callbacks_list = [checkpoint]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<keras.callbacks.callbacks.ModelCheckpoint at 0x20db89ddf08>]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "callbacks_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = train_x.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(583706, 21)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "target = train_y.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(583706, 1)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 466964 samples, validate on 116742 samples\n",
      "Epoch 1/500\n",
      "466964/466964 [==============================] - 24s 52us/step - loss: 67.0224 - mean_absolute_error: 67.0224 - val_loss: 65.9487 - val_mean_absolute_error: 65.9488\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 65.94874, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 2/500\n",
      "466964/466964 [==============================] - 25s 53us/step - loss: 65.8887 - mean_absolute_error: 65.8887 - val_loss: 65.8441 - val_mean_absolute_error: 65.8441\n",
      "\n",
      "Epoch 00002: val_loss improved from 65.94874 to 65.84411, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 3/500\n",
      "466964/466964 [==============================] - 26s 56us/step - loss: 65.8527 - mean_absolute_error: 65.8527 - val_loss: 65.8467 - val_mean_absolute_error: 65.8467\n",
      "\n",
      "Epoch 00003: val_loss did not improve from 65.84411\n",
      "Epoch 4/500\n",
      "466964/466964 [==============================] - 25s 53us/step - loss: 65.7934 - mean_absolute_error: 65.7933 - val_loss: 66.4936 - val_mean_absolute_error: 66.4936\n",
      "\n",
      "Epoch 00004: val_loss did not improve from 65.84411\n",
      "Epoch 5/500\n",
      "466964/466964 [==============================] - 27s 58us/step - loss: 65.7713 - mean_absolute_error: 65.7713 - val_loss: 66.0332 - val_mean_absolute_error: 66.0332\n",
      "\n",
      "Epoch 00005: val_loss did not improve from 65.84411\n",
      "Epoch 6/500\n",
      "466964/466964 [==============================] - 29s 63us/step - loss: 65.7486 - mean_absolute_error: 65.7486 - val_loss: 65.8643 - val_mean_absolute_error: 65.8643\n",
      "\n",
      "Epoch 00006: val_loss did not improve from 65.84411\n",
      "Epoch 7/500\n",
      "466964/466964 [==============================] - 32s 70us/step - loss: 65.7205 - mean_absolute_error: 65.7206 - val_loss: 66.8689 - val_mean_absolute_error: 66.8689\n",
      "\n",
      "Epoch 00007: val_loss did not improve from 65.84411\n",
      "Epoch 8/500\n",
      "466964/466964 [==============================] - 29s 63us/step - loss: 65.7166 - mean_absolute_error: 65.7166 - val_loss: 66.2067 - val_mean_absolute_error: 66.2067\n",
      "\n",
      "Epoch 00008: val_loss did not improve from 65.84411\n",
      "Epoch 9/500\n",
      "466964/466964 [==============================] - 23s 48us/step - loss: 65.7100 - mean_absolute_error: 65.7099 - val_loss: 65.7969 - val_mean_absolute_error: 65.7969\n",
      "\n",
      "Epoch 00009: val_loss improved from 65.84411 to 65.79694, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 10/500\n",
      "466964/466964 [==============================] - 23s 48us/step - loss: 65.6949 - mean_absolute_error: 65.6948 - val_loss: 65.9836 - val_mean_absolute_error: 65.9836\n",
      "\n",
      "Epoch 00010: val_loss did not improve from 65.79694\n",
      "Epoch 11/500\n",
      "466964/466964 [==============================] - 24s 50us/step - loss: 65.6725 - mean_absolute_error: 65.6725 - val_loss: 65.8531 - val_mean_absolute_error: 65.8531\n",
      "\n",
      "Epoch 00011: val_loss did not improve from 65.79694\n",
      "Epoch 12/500\n",
      "466964/466964 [==============================] - 26s 55us/step - loss: 65.6897 - mean_absolute_error: 65.6896 - val_loss: 66.1611 - val_mean_absolute_error: 66.1610\n",
      "\n",
      "Epoch 00012: val_loss did not improve from 65.79694\n",
      "Epoch 13/500\n",
      "466964/466964 [==============================] - 24s 52us/step - loss: 65.6765 - mean_absolute_error: 65.6765 - val_loss: 66.1927 - val_mean_absolute_error: 66.1927\n",
      "\n",
      "Epoch 00013: val_loss did not improve from 65.79694\n",
      "Epoch 14/500\n",
      "466964/466964 [==============================] - 24s 52us/step - loss: 65.6643 - mean_absolute_error: 65.6644 - val_loss: 65.9479 - val_mean_absolute_error: 65.9479\n",
      "\n",
      "Epoch 00014: val_loss did not improve from 65.79694\n",
      "Epoch 15/500\n",
      "466964/466964 [==============================] - 26s 56us/step - loss: 65.6574 - mean_absolute_error: 65.6573 - val_loss: 66.0138 - val_mean_absolute_error: 66.0139\n",
      "\n",
      "Epoch 00015: val_loss did not improve from 65.79694\n",
      "Epoch 16/500\n",
      "466964/466964 [==============================] - 25s 53us/step - loss: 65.6520 - mean_absolute_error: 65.6519 - val_loss: 65.7721 - val_mean_absolute_error: 65.7720\n",
      "\n",
      "Epoch 00016: val_loss improved from 65.79694 to 65.77208, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 17/500\n",
      "466964/466964 [==============================] - 27s 58us/step - loss: 65.6473 - mean_absolute_error: 65.6473 - val_loss: 65.8717 - val_mean_absolute_error: 65.8717\n",
      "\n",
      "Epoch 00017: val_loss did not improve from 65.77208\n",
      "Epoch 18/500\n",
      "466964/466964 [==============================] - 26s 56us/step - loss: 65.6402 - mean_absolute_error: 65.6403 - val_loss: 65.7507 - val_mean_absolute_error: 65.7507\n",
      "\n",
      "Epoch 00018: val_loss improved from 65.77208 to 65.75068, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 19/500\n",
      "466964/466964 [==============================] - 26s 56us/step - loss: 65.6402 - mean_absolute_error: 65.6402 - val_loss: 65.9546 - val_mean_absolute_error: 65.9546\n",
      "\n",
      "Epoch 00019: val_loss did not improve from 65.75068\n",
      "Epoch 20/500\n",
      "466964/466964 [==============================] - 25s 54us/step - loss: 65.6490 - mean_absolute_error: 65.6490 - val_loss: 65.8054 - val_mean_absolute_error: 65.8054\n",
      "\n",
      "Epoch 00020: val_loss did not improve from 65.75068\n",
      "Epoch 21/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6343 - mean_absolute_error: 65.6343 - val_loss: 65.8020 - val_mean_absolute_error: 65.8021\n",
      "\n",
      "Epoch 00021: val_loss did not improve from 65.75068\n",
      "Epoch 22/500\n",
      "466964/466964 [==============================] - 21s 45us/step - loss: 65.6228 - mean_absolute_error: 65.6228 - val_loss: 65.8828 - val_mean_absolute_error: 65.8828\n",
      "\n",
      "Epoch 00022: val_loss did not improve from 65.75068\n",
      "Epoch 23/500\n",
      "466964/466964 [==============================] - 21s 46us/step - loss: 65.6227 - mean_absolute_error: 65.6228 - val_loss: 65.9004 - val_mean_absolute_error: 65.9005\n",
      "\n",
      "Epoch 00023: val_loss did not improve from 65.75068\n",
      "Epoch 24/500\n",
      "466964/466964 [==============================] - 23s 48us/step - loss: 65.6210 - mean_absolute_error: 65.6210 - val_loss: 65.9292 - val_mean_absolute_error: 65.9292\n",
      "\n",
      "Epoch 00024: val_loss did not improve from 65.75068\n",
      "Epoch 25/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6163 - mean_absolute_error: 65.6162 - val_loss: 65.8030 - val_mean_absolute_error: 65.8030\n",
      "\n",
      "Epoch 00025: val_loss did not improve from 65.75068\n",
      "Epoch 26/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6131 - mean_absolute_error: 65.6131 - val_loss: 65.7574 - val_mean_absolute_error: 65.7574\n",
      "\n",
      "Epoch 00026: val_loss did not improve from 65.75068\n",
      "Epoch 27/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6145 - mean_absolute_error: 65.6145 - val_loss: 65.7558 - val_mean_absolute_error: 65.7559\n",
      "\n",
      "Epoch 00027: val_loss did not improve from 65.75068\n",
      "Epoch 28/500\n",
      "466964/466964 [==============================] - 22s 48us/step - loss: 65.6124 - mean_absolute_error: 65.6122 - val_loss: 66.0835 - val_mean_absolute_error: 66.0834\n",
      "\n",
      "Epoch 00028: val_loss did not improve from 65.75068\n",
      "Epoch 29/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6055 - mean_absolute_error: 65.6054 - val_loss: 65.7623 - val_mean_absolute_error: 65.7621\n",
      "\n",
      "Epoch 00029: val_loss did not improve from 65.75068\n",
      "Epoch 30/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.6022 - mean_absolute_error: 65.6022 - val_loss: 65.7569 - val_mean_absolute_error: 65.7569\n",
      "\n",
      "Epoch 00030: val_loss did not improve from 65.75068\n",
      "Epoch 31/500\n",
      "466964/466964 [==============================] - 24s 52us/step - loss: 65.6085 - mean_absolute_error: 65.6085 - val_loss: 65.7723 - val_mean_absolute_error: 65.7723\n",
      "\n",
      "Epoch 00031: val_loss did not improve from 65.75068\n",
      "Epoch 32/500\n",
      "466964/466964 [==============================] - 22s 48us/step - loss: 65.6090 - mean_absolute_error: 65.6089 - val_loss: 65.7616 - val_mean_absolute_error: 65.7616\n",
      "\n",
      "Epoch 00032: val_loss did not improve from 65.75068\n",
      "Epoch 33/500\n",
      "466964/466964 [==============================] - 26s 56us/step - loss: 65.6083 - mean_absolute_error: 65.6084 - val_loss: 65.8616 - val_mean_absolute_error: 65.8616\n",
      "\n",
      "Epoch 00033: val_loss did not improve from 65.75068\n",
      "Epoch 34/500\n",
      "466964/466964 [==============================] - 21s 46us/step - loss: 65.5928 - mean_absolute_error: 65.5928 - val_loss: 65.7919 - val_mean_absolute_error: 65.7919\n",
      "\n",
      "Epoch 00034: val_loss did not improve from 65.75068\n",
      "Epoch 35/500\n",
      "466964/466964 [==============================] - 21s 44us/step - loss: 65.6029 - mean_absolute_error: 65.6028 - val_loss: 65.7736 - val_mean_absolute_error: 65.7736\n",
      "\n",
      "Epoch 00035: val_loss did not improve from 65.75068\n",
      "Epoch 36/500\n",
      "466964/466964 [==============================] - 21s 45us/step - loss: 65.6036 - mean_absolute_error: 65.6037 - val_loss: 66.0675 - val_mean_absolute_error: 66.0675\n",
      "\n",
      "Epoch 00036: val_loss did not improve from 65.75068\n",
      "Epoch 37/500\n",
      "466964/466964 [==============================] - 22s 47us/step - loss: 65.5950 - mean_absolute_error: 65.5950 - val_loss: 65.8082 - val_mean_absolute_error: 65.8082\n",
      "\n",
      "Epoch 00037: val_loss did not improve from 65.75068\n",
      "Epoch 38/500\n",
      "466964/466964 [==============================] - 22s 48us/step - loss: 65.5925 - mean_absolute_error: 65.5925 - val_loss: 65.7449 - val_mean_absolute_error: 65.7449\n",
      "\n",
      "Epoch 00038: val_loss improved from 65.75068 to 65.74486, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 39/500\n",
      "466964/466964 [==============================] - 22s 46us/step - loss: 65.5966 - mean_absolute_error: 65.5968 - val_loss: 65.9304 - val_mean_absolute_error: 65.9304\n",
      "\n",
      "Epoch 00039: val_loss did not improve from 65.74486\n",
      "Epoch 40/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5837 - mean_absolute_error: 65.5836 - val_loss: 66.0021 - val_mean_absolute_error: 66.0021\n",
      "\n",
      "Epoch 00040: val_loss did not improve from 65.74486\n",
      "Epoch 41/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5817 - mean_absolute_error: 65.5815 - val_loss: 65.7531 - val_mean_absolute_error: 65.7531\n",
      "\n",
      "Epoch 00041: val_loss did not improve from 65.74486\n",
      "Epoch 42/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5859 - mean_absolute_error: 65.5859 - val_loss: 65.7565 - val_mean_absolute_error: 65.7564\n",
      "\n",
      "Epoch 00042: val_loss did not improve from 65.74486\n",
      "Epoch 43/500\n",
      "466964/466964 [==============================] - 22s 46us/step - loss: 65.5876 - mean_absolute_error: 65.5876 - val_loss: 65.7645 - val_mean_absolute_error: 65.7644\n",
      "\n",
      "Epoch 00043: val_loss did not improve from 65.74486\n",
      "Epoch 44/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5765 - mean_absolute_error: 65.5766 - val_loss: 65.8646 - val_mean_absolute_error: 65.8646\n",
      "\n",
      "Epoch 00044: val_loss did not improve from 65.74486\n",
      "Epoch 45/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5841 - mean_absolute_error: 65.5841 - val_loss: 65.7794 - val_mean_absolute_error: 65.7793\n",
      "\n",
      "Epoch 00045: val_loss did not improve from 65.74486\n",
      "Epoch 46/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5810 - mean_absolute_error: 65.5809 - val_loss: 65.7688 - val_mean_absolute_error: 65.7688\n",
      "\n",
      "Epoch 00046: val_loss did not improve from 65.74486\n",
      "Epoch 47/500\n",
      "466964/466964 [==============================] - 20s 42us/step - loss: 65.5840 - mean_absolute_error: 65.5841 - val_loss: 65.7492 - val_mean_absolute_error: 65.7492\n",
      "\n",
      "Epoch 00047: val_loss did not improve from 65.74486\n",
      "Epoch 48/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5827 - mean_absolute_error: 65.5827 - val_loss: 65.7431 - val_mean_absolute_error: 65.7431\n",
      "\n",
      "Epoch 00048: val_loss improved from 65.74486 to 65.74309, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 49/500\n",
      "466964/466964 [==============================] - 20s 42us/step - loss: 65.5738 - mean_absolute_error: 65.5736 - val_loss: 65.7422 - val_mean_absolute_error: 65.7422\n",
      "\n",
      "Epoch 00049: val_loss improved from 65.74309 to 65.74221, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 50/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5728 - mean_absolute_error: 65.5727 - val_loss: 65.7986 - val_mean_absolute_error: 65.7986\n",
      "\n",
      "Epoch 00050: val_loss did not improve from 65.74221\n",
      "Epoch 51/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5752 - mean_absolute_error: 65.5752 - val_loss: 65.7772 - val_mean_absolute_error: 65.7771\n",
      "\n",
      "Epoch 00051: val_loss did not improve from 65.74221\n",
      "Epoch 52/500\n",
      "466964/466964 [==============================] - 20s 42us/step - loss: 65.5722 - mean_absolute_error: 65.5723 - val_loss: 65.7455 - val_mean_absolute_error: 65.7455\n",
      "\n",
      "Epoch 00052: val_loss did not improve from 65.74221\n",
      "Epoch 53/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5730 - mean_absolute_error: 65.5731 - val_loss: 65.9148 - val_mean_absolute_error: 65.9148\n",
      "\n",
      "Epoch 00053: val_loss did not improve from 65.74221\n",
      "Epoch 54/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5662 - mean_absolute_error: 65.5662 - val_loss: 65.7668 - val_mean_absolute_error: 65.7668\n",
      "\n",
      "Epoch 00054: val_loss did not improve from 65.74221\n",
      "Epoch 55/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5607 - mean_absolute_error: 65.5607 - val_loss: 65.7653 - val_mean_absolute_error: 65.7652\n",
      "\n",
      "Epoch 00055: val_loss did not improve from 65.74221\n",
      "Epoch 56/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5596 - mean_absolute_error: 65.5596 - val_loss: 65.7939 - val_mean_absolute_error: 65.7939\n",
      "\n",
      "Epoch 00056: val_loss did not improve from 65.74221\n",
      "Epoch 57/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5617 - mean_absolute_error: 65.5618 - val_loss: 65.7628 - val_mean_absolute_error: 65.7627\n",
      "\n",
      "Epoch 00057: val_loss did not improve from 65.74221\n",
      "Epoch 58/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5652 - mean_absolute_error: 65.5651 - val_loss: 65.9699 - val_mean_absolute_error: 65.9699\n",
      "\n",
      "Epoch 00058: val_loss did not improve from 65.74221\n",
      "Epoch 59/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5685 - mean_absolute_error: 65.5685 - val_loss: 65.8897 - val_mean_absolute_error: 65.8896\n",
      "\n",
      "Epoch 00059: val_loss did not improve from 65.74221\n",
      "Epoch 60/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5637 - mean_absolute_error: 65.5637 - val_loss: 65.7541 - val_mean_absolute_error: 65.7541\n",
      "\n",
      "Epoch 00060: val_loss did not improve from 65.74221\n",
      "Epoch 61/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5638 - mean_absolute_error: 65.5638 - val_loss: 65.7708 - val_mean_absolute_error: 65.7708\n",
      "\n",
      "Epoch 00061: val_loss did not improve from 65.74221\n",
      "Epoch 62/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5704 - mean_absolute_error: 65.5704 - val_loss: 65.7869 - val_mean_absolute_error: 65.7868\n",
      "\n",
      "Epoch 00062: val_loss did not improve from 65.74221\n",
      "Epoch 63/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5598 - mean_absolute_error: 65.5598 - val_loss: 65.7551 - val_mean_absolute_error: 65.7551\n",
      "\n",
      "Epoch 00063: val_loss did not improve from 65.74221\n",
      "Epoch 64/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5616 - mean_absolute_error: 65.5615 - val_loss: 65.9850 - val_mean_absolute_error: 65.9850\n",
      "\n",
      "Epoch 00064: val_loss did not improve from 65.74221\n",
      "Epoch 65/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5604 - mean_absolute_error: 65.5605 - val_loss: 65.7548 - val_mean_absolute_error: 65.7548\n",
      "\n",
      "Epoch 00065: val_loss did not improve from 65.74221\n",
      "Epoch 66/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5650 - mean_absolute_error: 65.5651 - val_loss: 65.8198 - val_mean_absolute_error: 65.8198\n",
      "\n",
      "Epoch 00066: val_loss did not improve from 65.74221\n",
      "Epoch 67/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5539 - mean_absolute_error: 65.5538 - val_loss: 65.8114 - val_mean_absolute_error: 65.8115\n",
      "\n",
      "Epoch 00067: val_loss did not improve from 65.74221\n",
      "Epoch 68/500\n",
      "466964/466964 [==============================] - 20s 42us/step - loss: 65.5586 - mean_absolute_error: 65.5586 - val_loss: 65.7536 - val_mean_absolute_error: 65.7537\n",
      "\n",
      "Epoch 00068: val_loss did not improve from 65.74221\n",
      "Epoch 69/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5593 - mean_absolute_error: 65.5593 - val_loss: 65.7515 - val_mean_absolute_error: 65.7515\n",
      "\n",
      "Epoch 00069: val_loss did not improve from 65.74221\n",
      "Epoch 70/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5648 - mean_absolute_error: 65.5647 - val_loss: 65.7435 - val_mean_absolute_error: 65.7435\n",
      "\n",
      "Epoch 00070: val_loss did not improve from 65.74221\n",
      "Epoch 71/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5591 - mean_absolute_error: 65.5590 - val_loss: 65.7482 - val_mean_absolute_error: 65.7482\n",
      "\n",
      "Epoch 00071: val_loss did not improve from 65.74221\n",
      "Epoch 72/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5590 - mean_absolute_error: 65.5590 - val_loss: 65.7894 - val_mean_absolute_error: 65.7895\n",
      "\n",
      "Epoch 00072: val_loss did not improve from 65.74221\n",
      "Epoch 73/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5580 - mean_absolute_error: 65.5579 - val_loss: 65.7301 - val_mean_absolute_error: 65.7301\n",
      "\n",
      "Epoch 00073: val_loss improved from 65.74221 to 65.73015, saving model to models/Weights-001--65.91199.hdf5\n",
      "Epoch 74/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5585 - mean_absolute_error: 65.5583 - val_loss: 65.7562 - val_mean_absolute_error: 65.7561\n",
      "\n",
      "Epoch 00074: val_loss did not improve from 65.73015\n",
      "Epoch 75/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5590 - mean_absolute_error: 65.5590 - val_loss: 65.7606 - val_mean_absolute_error: 65.7607\n",
      "\n",
      "Epoch 00075: val_loss did not improve from 65.73015\n",
      "Epoch 76/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5545 - mean_absolute_error: 65.5546 - val_loss: 65.8277 - val_mean_absolute_error: 65.8277\n",
      "\n",
      "Epoch 00076: val_loss did not improve from 65.73015\n",
      "Epoch 77/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5539 - mean_absolute_error: 65.5539 - val_loss: 65.7683 - val_mean_absolute_error: 65.7683\n",
      "\n",
      "Epoch 00077: val_loss did not improve from 65.73015\n",
      "Epoch 78/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5530 - mean_absolute_error: 65.5529 - val_loss: 65.8126 - val_mean_absolute_error: 65.8126\n",
      "\n",
      "Epoch 00078: val_loss did not improve from 65.73015\n",
      "Epoch 79/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5549 - mean_absolute_error: 65.5550 - val_loss: 65.7608 - val_mean_absolute_error: 65.7608\n",
      "\n",
      "Epoch 00079: val_loss did not improve from 65.73015\n",
      "Epoch 80/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5547 - mean_absolute_error: 65.5546 - val_loss: 65.7633 - val_mean_absolute_error: 65.7632\n",
      "\n",
      "Epoch 00080: val_loss did not improve from 65.73015\n",
      "Epoch 81/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5593 - mean_absolute_error: 65.5592 - val_loss: 65.8130 - val_mean_absolute_error: 65.8130\n",
      "\n",
      "Epoch 00081: val_loss did not improve from 65.73015\n",
      "Epoch 82/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5547 - mean_absolute_error: 65.5547 - val_loss: 65.7564 - val_mean_absolute_error: 65.7564\n",
      "\n",
      "Epoch 00082: val_loss did not improve from 65.73015\n",
      "Epoch 83/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5561 - mean_absolute_error: 65.5561 - val_loss: 65.8256 - val_mean_absolute_error: 65.8256\n",
      "\n",
      "Epoch 00083: val_loss did not improve from 65.73015\n",
      "Epoch 84/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5566 - mean_absolute_error: 65.5566 - val_loss: 65.7685 - val_mean_absolute_error: 65.7685\n",
      "\n",
      "Epoch 00084: val_loss did not improve from 65.73015\n",
      "Epoch 85/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5542 - mean_absolute_error: 65.5542 - val_loss: 65.7898 - val_mean_absolute_error: 65.7898\n",
      "\n",
      "Epoch 00085: val_loss did not improve from 65.73015\n",
      "Epoch 86/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5537 - mean_absolute_error: 65.5538 - val_loss: 65.7398 - val_mean_absolute_error: 65.7397\n",
      "\n",
      "Epoch 00086: val_loss did not improve from 65.73015\n",
      "Epoch 87/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5483 - mean_absolute_error: 65.5484 - val_loss: 65.8022 - val_mean_absolute_error: 65.8022\n",
      "\n",
      "Epoch 00087: val_loss did not improve from 65.73015\n",
      "Epoch 88/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5533 - mean_absolute_error: 65.5533 - val_loss: 65.7418 - val_mean_absolute_error: 65.7418\n",
      "\n",
      "Epoch 00088: val_loss did not improve from 65.73015\n",
      "Epoch 89/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5524 - mean_absolute_error: 65.5524 - val_loss: 65.7925 - val_mean_absolute_error: 65.7925\n",
      "\n",
      "Epoch 00089: val_loss did not improve from 65.73015\n",
      "Epoch 90/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5513 - mean_absolute_error: 65.5514 - val_loss: 65.7534 - val_mean_absolute_error: 65.7534\n",
      "\n",
      "Epoch 00090: val_loss did not improve from 65.73015\n",
      "Epoch 91/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5499 - mean_absolute_error: 65.5499 - val_loss: 65.7594 - val_mean_absolute_error: 65.7594\n",
      "\n",
      "Epoch 00091: val_loss did not improve from 65.73015\n",
      "Epoch 92/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5494 - mean_absolute_error: 65.5493 - val_loss: 65.8281 - val_mean_absolute_error: 65.8281\n",
      "\n",
      "Epoch 00092: val_loss did not improve from 65.73015\n",
      "Epoch 93/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5553 - mean_absolute_error: 65.5554 - val_loss: 65.7820 - val_mean_absolute_error: 65.7820\n",
      "\n",
      "Epoch 00093: val_loss did not improve from 65.73015\n",
      "Epoch 94/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5515 - mean_absolute_error: 65.5515 - val_loss: 65.7420 - val_mean_absolute_error: 65.7421\n",
      "\n",
      "Epoch 00094: val_loss did not improve from 65.73015\n",
      "Epoch 95/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5445 - mean_absolute_error: 65.5444 - val_loss: 65.8519 - val_mean_absolute_error: 65.8518\n",
      "\n",
      "Epoch 00095: val_loss did not improve from 65.73015\n",
      "Epoch 96/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5502 - mean_absolute_error: 65.5501 - val_loss: 65.7857 - val_mean_absolute_error: 65.7857\n",
      "\n",
      "Epoch 00096: val_loss did not improve from 65.73015\n",
      "Epoch 97/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5431 - mean_absolute_error: 65.5429 - val_loss: 65.7333 - val_mean_absolute_error: 65.7334\n",
      "\n",
      "Epoch 00097: val_loss did not improve from 65.73015\n",
      "Epoch 98/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5457 - mean_absolute_error: 65.5457 - val_loss: 65.7373 - val_mean_absolute_error: 65.7373\n",
      "\n",
      "Epoch 00098: val_loss did not improve from 65.73015\n",
      "Epoch 99/500\n",
      "466964/466964 [==============================] - 18s 40us/step - loss: 65.5452 - mean_absolute_error: 65.5454 - val_loss: 65.7453 - val_mean_absolute_error: 65.7453\n",
      "\n",
      "Epoch 00099: val_loss did not improve from 65.73015\n",
      "Epoch 100/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5481 - mean_absolute_error: 65.5481 - val_loss: 65.7494 - val_mean_absolute_error: 65.7494\n",
      "\n",
      "Epoch 00100: val_loss did not improve from 65.73015\n",
      "Epoch 101/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5504 - mean_absolute_error: 65.5505 - val_loss: 65.7333 - val_mean_absolute_error: 65.7333\n",
      "\n",
      "Epoch 00101: val_loss did not improve from 65.73015\n",
      "Epoch 102/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5460 - mean_absolute_error: 65.5462 - val_loss: 65.7629 - val_mean_absolute_error: 65.7630\n",
      "\n",
      "Epoch 00102: val_loss did not improve from 65.73015\n",
      "Epoch 103/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5497 - mean_absolute_error: 65.5498 - val_loss: 65.7737 - val_mean_absolute_error: 65.7738\n",
      "\n",
      "Epoch 00103: val_loss did not improve from 65.73015\n",
      "Epoch 104/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5390 - mean_absolute_error: 65.5391 - val_loss: 65.7498 - val_mean_absolute_error: 65.7498\n",
      "\n",
      "Epoch 00104: val_loss did not improve from 65.73015\n",
      "Epoch 105/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5457 - mean_absolute_error: 65.5457 - val_loss: 65.7808 - val_mean_absolute_error: 65.7807\n",
      "\n",
      "Epoch 00105: val_loss did not improve from 65.73015\n",
      "Epoch 106/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5436 - mean_absolute_error: 65.5436 - val_loss: 65.7615 - val_mean_absolute_error: 65.7615\n",
      "\n",
      "Epoch 00106: val_loss did not improve from 65.73015\n",
      "Epoch 107/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5434 - mean_absolute_error: 65.5435 - val_loss: 66.1913 - val_mean_absolute_error: 66.1913\n",
      "\n",
      "Epoch 00107: val_loss did not improve from 65.73015\n",
      "Epoch 108/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5444 - mean_absolute_error: 65.5444 - val_loss: 65.7598 - val_mean_absolute_error: 65.7598\n",
      "\n",
      "Epoch 00108: val_loss did not improve from 65.73015\n",
      "Epoch 109/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5422 - mean_absolute_error: 65.5422 - val_loss: 65.7399 - val_mean_absolute_error: 65.7399\n",
      "\n",
      "Epoch 00109: val_loss did not improve from 65.73015\n",
      "Epoch 110/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5482 - mean_absolute_error: 65.5483 - val_loss: 65.7412 - val_mean_absolute_error: 65.7412\n",
      "\n",
      "Epoch 00110: val_loss did not improve from 65.73015\n",
      "Epoch 111/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5470 - mean_absolute_error: 65.5472 - val_loss: 65.7706 - val_mean_absolute_error: 65.7705\n",
      "\n",
      "Epoch 00111: val_loss did not improve from 65.73015\n",
      "Epoch 112/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5393 - mean_absolute_error: 65.5392 - val_loss: 65.7685 - val_mean_absolute_error: 65.7685\n",
      "\n",
      "Epoch 00112: val_loss did not improve from 65.73015\n",
      "Epoch 113/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5433 - mean_absolute_error: 65.5433 - val_loss: 65.7481 - val_mean_absolute_error: 65.7481\n",
      "\n",
      "Epoch 00113: val_loss did not improve from 65.73015\n",
      "Epoch 114/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5386 - mean_absolute_error: 65.5386 - val_loss: 65.8864 - val_mean_absolute_error: 65.8863\n",
      "\n",
      "Epoch 00114: val_loss did not improve from 65.73015\n",
      "Epoch 115/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5412 - mean_absolute_error: 65.5412 - val_loss: 65.7724 - val_mean_absolute_error: 65.7723\n",
      "\n",
      "Epoch 00115: val_loss did not improve from 65.73015\n",
      "Epoch 116/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5440 - mean_absolute_error: 65.5441 - val_loss: 65.8188 - val_mean_absolute_error: 65.8188\n",
      "\n",
      "Epoch 00116: val_loss did not improve from 65.73015\n",
      "Epoch 117/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5414 - mean_absolute_error: 65.5415 - val_loss: 65.7501 - val_mean_absolute_error: 65.7501\n",
      "\n",
      "Epoch 00117: val_loss did not improve from 65.73015\n",
      "Epoch 118/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5435 - mean_absolute_error: 65.5436 - val_loss: 65.7410 - val_mean_absolute_error: 65.7410\n",
      "\n",
      "Epoch 00118: val_loss did not improve from 65.73015\n",
      "Epoch 119/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5388 - mean_absolute_error: 65.5388 - val_loss: 65.7408 - val_mean_absolute_error: 65.7408\n",
      "\n",
      "Epoch 00119: val_loss did not improve from 65.73015\n",
      "Epoch 120/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5392 - mean_absolute_error: 65.5393 - val_loss: 65.7731 - val_mean_absolute_error: 65.7730\n",
      "\n",
      "Epoch 00120: val_loss did not improve from 65.73015\n",
      "Epoch 121/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5398 - mean_absolute_error: 65.5398 - val_loss: 65.7766 - val_mean_absolute_error: 65.7766\n",
      "\n",
      "Epoch 00121: val_loss did not improve from 65.73015\n",
      "Epoch 122/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5448 - mean_absolute_error: 65.5449 - val_loss: 65.7683 - val_mean_absolute_error: 65.7684\n",
      "\n",
      "Epoch 00122: val_loss did not improve from 65.73015\n",
      "Epoch 123/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5478 - mean_absolute_error: 65.5478 - val_loss: 65.7836 - val_mean_absolute_error: 65.7836\n",
      "\n",
      "Epoch 00123: val_loss did not improve from 65.73015\n",
      "Epoch 124/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5422 - mean_absolute_error: 65.5422 - val_loss: 65.7952 - val_mean_absolute_error: 65.7953\n",
      "\n",
      "Epoch 00124: val_loss did not improve from 65.73015\n",
      "Epoch 125/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5398 - mean_absolute_error: 65.5399 - val_loss: 65.7794 - val_mean_absolute_error: 65.7794\n",
      "\n",
      "Epoch 00125: val_loss did not improve from 65.73015\n",
      "Epoch 126/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5410 - mean_absolute_error: 65.5410 - val_loss: 65.7624 - val_mean_absolute_error: 65.7624\n",
      "\n",
      "Epoch 00126: val_loss did not improve from 65.73015\n",
      "Epoch 127/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5396 - mean_absolute_error: 65.5396 - val_loss: 65.7629 - val_mean_absolute_error: 65.7628\n",
      "\n",
      "Epoch 00127: val_loss did not improve from 65.73015\n",
      "Epoch 128/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5388 - mean_absolute_error: 65.5388 - val_loss: 65.7584 - val_mean_absolute_error: 65.7585\n",
      "\n",
      "Epoch 00128: val_loss did not improve from 65.73015\n",
      "Epoch 129/500\n",
      "466964/466964 [==============================] - 18s 40us/step - loss: 65.5405 - mean_absolute_error: 65.5405 - val_loss: 65.7496 - val_mean_absolute_error: 65.7496\n",
      "\n",
      "Epoch 00129: val_loss did not improve from 65.73015\n",
      "Epoch 130/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5362 - mean_absolute_error: 65.5362 - val_loss: 65.8030 - val_mean_absolute_error: 65.8030\n",
      "\n",
      "Epoch 00130: val_loss did not improve from 65.73015\n",
      "Epoch 131/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5332 - mean_absolute_error: 65.5333 - val_loss: 65.7953 - val_mean_absolute_error: 65.7952\n",
      "\n",
      "Epoch 00131: val_loss did not improve from 65.73015\n",
      "Epoch 132/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5355 - mean_absolute_error: 65.5355 - val_loss: 65.7866 - val_mean_absolute_error: 65.7866\n",
      "\n",
      "Epoch 00132: val_loss did not improve from 65.73015\n",
      "Epoch 133/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5335 - mean_absolute_error: 65.5335 - val_loss: 65.7898 - val_mean_absolute_error: 65.7897\n",
      "\n",
      "Epoch 00133: val_loss did not improve from 65.73015\n",
      "Epoch 134/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5346 - mean_absolute_error: 65.5347 - val_loss: 65.7499 - val_mean_absolute_error: 65.7500\n",
      "\n",
      "Epoch 00134: val_loss did not improve from 65.73015\n",
      "Epoch 135/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5288 - mean_absolute_error: 65.5288 - val_loss: 65.7373 - val_mean_absolute_error: 65.7373\n",
      "\n",
      "Epoch 00135: val_loss did not improve from 65.73015\n",
      "Epoch 136/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5321 - mean_absolute_error: 65.5322 - val_loss: 65.7432 - val_mean_absolute_error: 65.7432\n",
      "\n",
      "Epoch 00136: val_loss did not improve from 65.73015\n",
      "Epoch 137/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5340 - mean_absolute_error: 65.5339 - val_loss: 65.7800 - val_mean_absolute_error: 65.7800\n",
      "\n",
      "Epoch 00137: val_loss did not improve from 65.73015\n",
      "Epoch 138/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5312 - mean_absolute_error: 65.5313 - val_loss: 65.8113 - val_mean_absolute_error: 65.8113\n",
      "\n",
      "Epoch 00138: val_loss did not improve from 65.73015\n",
      "Epoch 139/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5395 - mean_absolute_error: 65.5395 - val_loss: 65.7528 - val_mean_absolute_error: 65.7528\n",
      "\n",
      "Epoch 00139: val_loss did not improve from 65.73015\n",
      "Epoch 140/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "466964/466964 [==============================] - 17s 37us/step - loss: 65.5294 - mean_absolute_error: 65.5294 - val_loss: 65.7551 - val_mean_absolute_error: 65.7551\n",
      "\n",
      "Epoch 00140: val_loss did not improve from 65.73015\n",
      "Epoch 141/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5368 - mean_absolute_error: 65.5369 - val_loss: 65.7408 - val_mean_absolute_error: 65.7409\n",
      "\n",
      "Epoch 00141: val_loss did not improve from 65.73015\n",
      "Epoch 142/500\n",
      "466964/466964 [==============================] - 17s 37us/step - loss: 65.5308 - mean_absolute_error: 65.5307 - val_loss: 65.7422 - val_mean_absolute_error: 65.7422\n",
      "\n",
      "Epoch 00142: val_loss did not improve from 65.73015\n",
      "Epoch 143/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5323 - mean_absolute_error: 65.5322 - val_loss: 65.8108 - val_mean_absolute_error: 65.8108\n",
      "\n",
      "Epoch 00143: val_loss did not improve from 65.73015\n",
      "Epoch 144/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5356 - mean_absolute_error: 65.5355 - val_loss: 65.7362 - val_mean_absolute_error: 65.7362\n",
      "\n",
      "Epoch 00144: val_loss did not improve from 65.73015\n",
      "Epoch 145/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5353 - mean_absolute_error: 65.5353 - val_loss: 65.7681 - val_mean_absolute_error: 65.7681\n",
      "\n",
      "Epoch 00145: val_loss did not improve from 65.73015\n",
      "Epoch 146/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5305 - mean_absolute_error: 65.5305 - val_loss: 65.8044 - val_mean_absolute_error: 65.8044\n",
      "\n",
      "Epoch 00146: val_loss did not improve from 65.73015\n",
      "Epoch 147/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5346 - mean_absolute_error: 65.5345 - val_loss: 65.7454 - val_mean_absolute_error: 65.7454\n",
      "\n",
      "Epoch 00147: val_loss did not improve from 65.73015\n",
      "Epoch 148/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5323 - mean_absolute_error: 65.5323 - val_loss: 65.7602 - val_mean_absolute_error: 65.7602\n",
      "\n",
      "Epoch 00148: val_loss did not improve from 65.73015\n",
      "Epoch 149/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5354 - mean_absolute_error: 65.5353 - val_loss: 65.7499 - val_mean_absolute_error: 65.7499\n",
      "\n",
      "Epoch 00149: val_loss did not improve from 65.73015\n",
      "Epoch 150/500\n",
      "466964/466964 [==============================] - 18s 37us/step - loss: 65.5364 - mean_absolute_error: 65.5364 - val_loss: 65.7324 - val_mean_absolute_error: 65.7325\n",
      "\n",
      "Epoch 00150: val_loss did not improve from 65.73015\n",
      "Epoch 151/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5329 - mean_absolute_error: 65.5330 - val_loss: 65.7309 - val_mean_absolute_error: 65.7309\n",
      "\n",
      "Epoch 00151: val_loss did not improve from 65.73015\n",
      "Epoch 152/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5282 - mean_absolute_error: 65.5282 - val_loss: 65.7383 - val_mean_absolute_error: 65.7382\n",
      "\n",
      "Epoch 00152: val_loss did not improve from 65.73015\n",
      "Epoch 153/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5387 - mean_absolute_error: 65.5388 - val_loss: 65.7397 - val_mean_absolute_error: 65.7397\n",
      "\n",
      "Epoch 00153: val_loss did not improve from 65.73015\n",
      "Epoch 154/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5355 - mean_absolute_error: 65.5354 - val_loss: 65.7343 - val_mean_absolute_error: 65.7343\n",
      "\n",
      "Epoch 00154: val_loss did not improve from 65.73015\n",
      "Epoch 155/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5366 - mean_absolute_error: 65.5365 - val_loss: 65.8046 - val_mean_absolute_error: 65.8046\n",
      "\n",
      "Epoch 00155: val_loss did not improve from 65.73015\n",
      "Epoch 156/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5310 - mean_absolute_error: 65.5310 - val_loss: 65.9101 - val_mean_absolute_error: 65.9101\n",
      "\n",
      "Epoch 00156: val_loss did not improve from 65.73015\n",
      "Epoch 157/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5352 - mean_absolute_error: 65.5354 - val_loss: 65.7494 - val_mean_absolute_error: 65.7495\n",
      "\n",
      "Epoch 00157: val_loss did not improve from 65.73015\n",
      "Epoch 158/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5327 - mean_absolute_error: 65.5327 - val_loss: 65.7583 - val_mean_absolute_error: 65.7583\n",
      "\n",
      "Epoch 00158: val_loss did not improve from 65.73015\n",
      "Epoch 159/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5294 - mean_absolute_error: 65.5294 - val_loss: 65.8120 - val_mean_absolute_error: 65.8119\n",
      "\n",
      "Epoch 00159: val_loss did not improve from 65.73015\n",
      "Epoch 160/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5368 - mean_absolute_error: 65.5368 - val_loss: 65.9704 - val_mean_absolute_error: 65.9704\n",
      "\n",
      "Epoch 00160: val_loss did not improve from 65.73015\n",
      "Epoch 161/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5317 - mean_absolute_error: 65.5317 - val_loss: 65.7381 - val_mean_absolute_error: 65.7381\n",
      "\n",
      "Epoch 00161: val_loss did not improve from 65.73015\n",
      "Epoch 162/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5278 - mean_absolute_error: 65.5278 - val_loss: 65.8125 - val_mean_absolute_error: 65.8125\n",
      "\n",
      "Epoch 00162: val_loss did not improve from 65.73015\n",
      "Epoch 163/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5343 - mean_absolute_error: 65.5343 - val_loss: 65.7973 - val_mean_absolute_error: 65.7973\n",
      "\n",
      "Epoch 00163: val_loss did not improve from 65.73015\n",
      "Epoch 164/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5296 - mean_absolute_error: 65.5296 - val_loss: 65.7546 - val_mean_absolute_error: 65.7546\n",
      "\n",
      "Epoch 00164: val_loss did not improve from 65.73015\n",
      "Epoch 165/500\n",
      "466964/466964 [==============================] - 542s 1ms/step - loss: 65.5307 - mean_absolute_error: 65.5308 - val_loss: 65.8098 - val_mean_absolute_error: 65.8099\n",
      "\n",
      "Epoch 00165: val_loss did not improve from 65.73015\n",
      "Epoch 166/500\n",
      "466964/466964 [==============================] - 29s 61us/step - loss: 65.5334 - mean_absolute_error: 65.5334 - val_loss: 65.7772 - val_mean_absolute_error: 65.7773\n",
      "\n",
      "Epoch 00166: val_loss did not improve from 65.73015\n",
      "Epoch 167/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5322 - mean_absolute_error: 65.5322 - val_loss: 65.7798 - val_mean_absolute_error: 65.7798\n",
      "\n",
      "Epoch 00167: val_loss did not improve from 65.73015\n",
      "Epoch 168/500\n",
      "466964/466964 [==============================] - 20s 42us/step - loss: 65.5246 - mean_absolute_error: 65.5246 - val_loss: 65.7709 - val_mean_absolute_error: 65.7710\n",
      "\n",
      "Epoch 00168: val_loss did not improve from 65.73015\n",
      "Epoch 169/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5294 - mean_absolute_error: 65.5294 - val_loss: 65.8381 - val_mean_absolute_error: 65.8381\n",
      "\n",
      "Epoch 00169: val_loss did not improve from 65.73015\n",
      "Epoch 170/500\n",
      "466964/466964 [==============================] - 19s 42us/step - loss: 65.5312 - mean_absolute_error: 65.5312 - val_loss: 65.7375 - val_mean_absolute_error: 65.7375\n",
      "\n",
      "Epoch 00170: val_loss did not improve from 65.73015\n",
      "Epoch 171/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5294 - mean_absolute_error: 65.5294 - val_loss: 65.7394 - val_mean_absolute_error: 65.7394\n",
      "\n",
      "Epoch 00171: val_loss did not improve from 65.73015\n",
      "Epoch 172/500\n",
      "466964/466964 [==============================] - 18s 40us/step - loss: 65.5295 - mean_absolute_error: 65.5295 - val_loss: 65.7582 - val_mean_absolute_error: 65.7582\n",
      "\n",
      "Epoch 00172: val_loss did not improve from 65.73015\n",
      "Epoch 173/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5317 - mean_absolute_error: 65.5318 - val_loss: 65.7735 - val_mean_absolute_error: 65.7735\n",
      "\n",
      "Epoch 00173: val_loss did not improve from 65.73015\n",
      "Epoch 174/500\n",
      "466964/466964 [==============================] - 18s 40us/step - loss: 65.5305 - mean_absolute_error: 65.5305 - val_loss: 65.7315 - val_mean_absolute_error: 65.7315\n",
      "\n",
      "Epoch 00174: val_loss did not improve from 65.73015\n",
      "Epoch 175/500\n",
      "466964/466964 [==============================] - 20s 44us/step - loss: 65.5235 - mean_absolute_error: 65.5235 - val_loss: 65.7383 - val_mean_absolute_error: 65.7383\n",
      "\n",
      "Epoch 00175: val_loss did not improve from 65.73015\n",
      "Epoch 176/500\n",
      "466964/466964 [==============================] - 18s 40us/step - loss: 65.5295 - mean_absolute_error: 65.5294 - val_loss: 65.7474 - val_mean_absolute_error: 65.7474\n",
      "\n",
      "Epoch 00176: val_loss did not improve from 65.73015\n",
      "Epoch 177/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5287 - mean_absolute_error: 65.5287 - val_loss: 65.7362 - val_mean_absolute_error: 65.7361\n",
      "\n",
      "Epoch 00177: val_loss did not improve from 65.73015\n",
      "Epoch 178/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5279 - mean_absolute_error: 65.5279 - val_loss: 65.8204 - val_mean_absolute_error: 65.8204\n",
      "\n",
      "Epoch 00178: val_loss did not improve from 65.73015\n",
      "Epoch 179/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5253 - mean_absolute_error: 65.5252 - val_loss: 65.7504 - val_mean_absolute_error: 65.7504\n",
      "\n",
      "Epoch 00179: val_loss did not improve from 65.73015\n",
      "Epoch 180/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5267 - mean_absolute_error: 65.5268 - val_loss: 65.7918 - val_mean_absolute_error: 65.7917\n",
      "\n",
      "Epoch 00180: val_loss did not improve from 65.73015\n",
      "Epoch 181/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5326 - mean_absolute_error: 65.5327 - val_loss: 65.7399 - val_mean_absolute_error: 65.7400\n",
      "\n",
      "Epoch 00181: val_loss did not improve from 65.73015\n",
      "Epoch 182/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5279 - mean_absolute_error: 65.5278 - val_loss: 65.7412 - val_mean_absolute_error: 65.7412\n",
      "\n",
      "Epoch 00182: val_loss did not improve from 65.73015\n",
      "Epoch 183/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5272 - mean_absolute_error: 65.5273 - val_loss: 65.7475 - val_mean_absolute_error: 65.7475\n",
      "\n",
      "Epoch 00183: val_loss did not improve from 65.73015\n",
      "Epoch 184/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5309 - mean_absolute_error: 65.5308 - val_loss: 65.7484 - val_mean_absolute_error: 65.7484\n",
      "\n",
      "Epoch 00184: val_loss did not improve from 65.73015\n",
      "Epoch 185/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5277 - mean_absolute_error: 65.5278 - val_loss: 65.8082 - val_mean_absolute_error: 65.8083\n",
      "\n",
      "Epoch 00185: val_loss did not improve from 65.73015\n",
      "Epoch 186/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5300 - mean_absolute_error: 65.5302 - val_loss: 65.7518 - val_mean_absolute_error: 65.7517\n",
      "\n",
      "Epoch 00186: val_loss did not improve from 65.73015\n",
      "Epoch 187/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5290 - mean_absolute_error: 65.5290 - val_loss: 65.7730 - val_mean_absolute_error: 65.7730\n",
      "\n",
      "Epoch 00187: val_loss did not improve from 65.73015\n",
      "Epoch 188/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5297 - mean_absolute_error: 65.5298 - val_loss: 65.7448 - val_mean_absolute_error: 65.7448\n",
      "\n",
      "Epoch 00188: val_loss did not improve from 65.73015\n",
      "Epoch 189/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5292 - mean_absolute_error: 65.5292 - val_loss: 65.7357 - val_mean_absolute_error: 65.7357\n",
      "\n",
      "Epoch 00189: val_loss did not improve from 65.73015\n",
      "Epoch 190/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5277 - mean_absolute_error: 65.5279 - val_loss: 65.8322 - val_mean_absolute_error: 65.8322\n",
      "\n",
      "Epoch 00190: val_loss did not improve from 65.73015\n",
      "Epoch 191/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5264 - mean_absolute_error: 65.5264 - val_loss: 65.7483 - val_mean_absolute_error: 65.7483\n",
      "\n",
      "Epoch 00191: val_loss did not improve from 65.73015\n",
      "Epoch 192/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5242 - mean_absolute_error: 65.5241 - val_loss: 65.7965 - val_mean_absolute_error: 65.7965\n",
      "\n",
      "Epoch 00192: val_loss did not improve from 65.73015\n",
      "Epoch 193/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5288 - mean_absolute_error: 65.5289 - val_loss: 65.7705 - val_mean_absolute_error: 65.7704\n",
      "\n",
      "Epoch 00193: val_loss did not improve from 65.73015\n",
      "Epoch 194/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5275 - mean_absolute_error: 65.5274 - val_loss: 65.7396 - val_mean_absolute_error: 65.7396\n",
      "\n",
      "Epoch 00194: val_loss did not improve from 65.73015\n",
      "Epoch 195/500\n",
      "466964/466964 [==============================] - 19s 40us/step - loss: 65.5227 - mean_absolute_error: 65.5228 - val_loss: 65.9041 - val_mean_absolute_error: 65.9040\n",
      "\n",
      "Epoch 00195: val_loss did not improve from 65.73015\n",
      "Epoch 196/500\n",
      "466964/466964 [==============================] - 18s 39us/step - loss: 65.5329 - mean_absolute_error: 65.5329 - val_loss: 65.7494 - val_mean_absolute_error: 65.7494\n",
      "\n",
      "Epoch 00196: val_loss did not improve from 65.73015\n",
      "Epoch 197/500\n",
      "466964/466964 [==============================] - 19s 42us/step - loss: 65.5272 - mean_absolute_error: 65.5273 - val_loss: 65.8306 - val_mean_absolute_error: 65.8306\n",
      "\n",
      "Epoch 00197: val_loss did not improve from 65.73015\n",
      "Epoch 198/500\n",
      "466964/466964 [==============================] - 20s 43us/step - loss: 65.5269 - mean_absolute_error: 65.5269 - val_loss: 65.8283 - val_mean_absolute_error: 65.8283\n",
      "\n",
      "Epoch 00198: val_loss did not improve from 65.73015\n",
      "Epoch 199/500\n",
      "466964/466964 [==============================] - 20s 44us/step - loss: 65.5307 - mean_absolute_error: 65.5308 - val_loss: 65.7754 - val_mean_absolute_error: 65.7755\n",
      "\n",
      "Epoch 00199: val_loss did not improve from 65.73015\n",
      "Epoch 200/500\n",
      "466964/466964 [==============================] - 21s 45us/step - loss: 65.5241 - mean_absolute_error: 65.5241 - val_loss: 65.7393 - val_mean_absolute_error: 65.7393\n",
      "\n",
      "Epoch 00200: val_loss did not improve from 65.73015\n",
      "Epoch 201/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5268 - mean_absolute_error: 65.5268 - val_loss: 65.7419 - val_mean_absolute_error: 65.7419\n",
      "\n",
      "Epoch 00201: val_loss did not improve from 65.73015\n",
      "Epoch 202/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5277 - mean_absolute_error: 65.5276 - val_loss: 65.7381 - val_mean_absolute_error: 65.7381\n",
      "\n",
      "Epoch 00202: val_loss did not improve from 65.73015\n",
      "Epoch 203/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5264 - mean_absolute_error: 65.5264 - val_loss: 65.7846 - val_mean_absolute_error: 65.7846\n",
      "\n",
      "Epoch 00203: val_loss did not improve from 65.73015\n",
      "Epoch 204/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5246 - mean_absolute_error: 65.5247 - val_loss: 65.7347 - val_mean_absolute_error: 65.7346\n",
      "\n",
      "Epoch 00204: val_loss did not improve from 65.73015\n",
      "Epoch 205/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5303 - mean_absolute_error: 65.5304 - val_loss: 65.8139 - val_mean_absolute_error: 65.8138\n",
      "\n",
      "Epoch 00205: val_loss did not improve from 65.73015\n",
      "Epoch 206/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5307 - mean_absolute_error: 65.5306 - val_loss: 65.7767 - val_mean_absolute_error: 65.7767\n",
      "\n",
      "Epoch 00206: val_loss did not improve from 65.73015\n",
      "Epoch 207/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5290 - mean_absolute_error: 65.5289 - val_loss: 65.7402 - val_mean_absolute_error: 65.7402\n",
      "\n",
      "Epoch 00207: val_loss did not improve from 65.73015\n",
      "Epoch 208/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5249 - mean_absolute_error: 65.5248 - val_loss: 65.7617 - val_mean_absolute_error: 65.7617\n",
      "\n",
      "Epoch 00208: val_loss did not improve from 65.73015\n",
      "Epoch 209/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5255 - mean_absolute_error: 65.5255 - val_loss: 65.7440 - val_mean_absolute_error: 65.7440\n",
      "\n",
      "Epoch 00209: val_loss did not improve from 65.73015\n",
      "Epoch 210/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5265 - mean_absolute_error: 65.5264 - val_loss: 65.7912 - val_mean_absolute_error: 65.7912\n",
      "\n",
      "Epoch 00210: val_loss did not improve from 65.73015\n",
      "Epoch 211/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5246 - mean_absolute_error: 65.5247 - val_loss: 65.7337 - val_mean_absolute_error: 65.7337\n",
      "\n",
      "Epoch 00211: val_loss did not improve from 65.73015\n",
      "Epoch 212/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5222 - mean_absolute_error: 65.5222 - val_loss: 65.7326 - val_mean_absolute_error: 65.7327\n",
      "\n",
      "Epoch 00212: val_loss did not improve from 65.73015\n",
      "Epoch 213/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5254 - mean_absolute_error: 65.5254 - val_loss: 65.7925 - val_mean_absolute_error: 65.7925\n",
      "\n",
      "Epoch 00213: val_loss did not improve from 65.73015\n",
      "Epoch 214/500\n",
      "466964/466964 [==============================] - 19s 41us/step - loss: 65.5223 - mean_absolute_error: 65.5222 - val_loss: 65.7415 - val_mean_absolute_error: 65.7415\n",
      "\n",
      "Epoch 00214: val_loss did not improve from 65.73015\n",
      "Epoch 215/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5190 - mean_absolute_error: 65.5190 - val_loss: 65.7776 - val_mean_absolute_error: 65.7777\n",
      "\n",
      "Epoch 00215: val_loss did not improve from 65.73015\n",
      "Epoch 216/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5237 - mean_absolute_error: 65.5236 - val_loss: 65.7838 - val_mean_absolute_error: 65.7838\n",
      "\n",
      "Epoch 00216: val_loss did not improve from 65.73015\n",
      "Epoch 217/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5232 - mean_absolute_error: 65.5234 - val_loss: 65.7316 - val_mean_absolute_error: 65.7316\n",
      "\n",
      "Epoch 00217: val_loss did not improve from 65.73015\n",
      "Epoch 218/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5257 - mean_absolute_error: 65.5257 - val_loss: 65.8069 - val_mean_absolute_error: 65.8068\n",
      "\n",
      "Epoch 00218: val_loss did not improve from 65.73015\n",
      "Epoch 219/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5284 - mean_absolute_error: 65.5282 - val_loss: 65.7363 - val_mean_absolute_error: 65.7363\n",
      "\n",
      "Epoch 00219: val_loss did not improve from 65.73015\n",
      "Epoch 220/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5215 - mean_absolute_error: 65.5214 - val_loss: 65.7646 - val_mean_absolute_error: 65.7646\n",
      "\n",
      "Epoch 00220: val_loss did not improve from 65.73015\n",
      "Epoch 221/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5315 - mean_absolute_error: 65.5314 - val_loss: 65.7332 - val_mean_absolute_error: 65.7332\n",
      "\n",
      "Epoch 00221: val_loss did not improve from 65.73015\n",
      "Epoch 222/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5275 - mean_absolute_error: 65.5275 - val_loss: 65.7449 - val_mean_absolute_error: 65.7449\n",
      "\n",
      "Epoch 00222: val_loss did not improve from 65.73015\n",
      "Epoch 223/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5252 - mean_absolute_error: 65.5252 - val_loss: 65.7936 - val_mean_absolute_error: 65.7937\n",
      "\n",
      "Epoch 00223: val_loss did not improve from 65.73015\n",
      "Epoch 224/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5255 - mean_absolute_error: 65.5256 - val_loss: 65.7620 - val_mean_absolute_error: 65.7620\n",
      "\n",
      "Epoch 00224: val_loss did not improve from 65.73015\n",
      "Epoch 225/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5219 - mean_absolute_error: 65.5219 - val_loss: 65.7946 - val_mean_absolute_error: 65.7945\n",
      "\n",
      "Epoch 00225: val_loss did not improve from 65.73015\n",
      "Epoch 226/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5230 - mean_absolute_error: 65.5229 - val_loss: 65.7321 - val_mean_absolute_error: 65.7321\n",
      "\n",
      "Epoch 00226: val_loss did not improve from 65.73015\n",
      "Epoch 227/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5255 - mean_absolute_error: 65.5255 - val_loss: 65.7448 - val_mean_absolute_error: 65.7448\n",
      "\n",
      "Epoch 00227: val_loss did not improve from 65.73015\n",
      "Epoch 228/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5217 - mean_absolute_error: 65.5218 - val_loss: 65.7702 - val_mean_absolute_error: 65.7702\n",
      "\n",
      "Epoch 00228: val_loss did not improve from 65.73015\n",
      "Epoch 229/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5252 - mean_absolute_error: 65.5253 - val_loss: 65.7410 - val_mean_absolute_error: 65.7410\n",
      "\n",
      "Epoch 00229: val_loss did not improve from 65.73015\n",
      "Epoch 230/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5221 - mean_absolute_error: 65.5221 - val_loss: 65.7403 - val_mean_absolute_error: 65.7403\n",
      "\n",
      "Epoch 00230: val_loss did not improve from 65.73015\n",
      "Epoch 231/500\n",
      "466964/466964 [==============================] - 18s 38us/step - loss: 65.5247 - mean_absolute_error: 65.5249 - val_loss: 65.7376 - val_mean_absolute_error: 65.7376\n",
      "\n",
      "Epoch 00231: val_loss did not improve from 65.73015\n",
      "Epoch 232/500\n",
      "465984/466964 [============================>.] - ETA: 18s - loss: 65.5256 - mean_absolute_error: 65.5256"
     ]
    }
   ],
   "source": [
    "NN_model.fit(train, target, epochs=500, batch_size=64, validation_split = 0.2, callbacks=callbacks_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "wights_file = 'models/Weights-001--65.91199.hdf5' # choose the best checkpoint \n",
    "NN_model.load_weights(wights_file) # load it\n",
    "NN_model.compile(loss='mean_absolute_error', optimizer='adam', metrics=['mean_absolute_error'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = test_x.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(145927, 21)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = NN_model.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([120.623795], dtype=float32)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "actual = test_y.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([126.46])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "actual[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### We will judge our model on Mean Absolute Error Metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Deep Neural Network validation MAE =  65.8824274131749\n"
     ]
    }
   ],
   "source": [
    "MAE = mean_absolute_error(actual , predictions)\n",
    "print('Deep Neural Network validation MAE = ', MAE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trying CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras.layers import Activation, Dropout, Flatten, Dense\n",
    "\n",
    "model = Sequential()\n",
    "# CNN Layers\n",
    "model.add(Conv2D(32, (3, 3), input_shape=(300, 300, 3)))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "# the model so far outputs 3D feature maps (height, width, features)\n",
    "\n",
    "model.add(Flatten())  # this converts our 3D feature maps to 1D feature vectors\n",
    "model.add(Dense(64))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1))\n",
    "model.add(Activation('sigmoid'))\n",
    "# COMPILE\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}