Yield Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Climate and Soil Data to Predict Crop Yeilds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from keras import Sequential\n",
    "from keras.layers import Dense\n",
    "\n",
    "from sklearn.preprocessing import Normalizer\n",
    "from sklearn.metrics import r2_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The first kind of data used is Yield Data collected from the USDA NASS QuickStat \n",
    "corn = pd.read_csv('data/corn.csv').set_index(['Year','ANSI'])\n",
    "corn.columns = ['corn']\n",
    "soy = pd.read_csv('data/soy.csv').set_index(['Year','ANSI'])\n",
    "soy.columns = ['soy']\n",
    "wheat = pd.read_csv('data/wheat.csv').set_index(['Year','ANSI'])\n",
    "wheat.columns = ['wheat']\n",
    "crops = corn.join(soy).join(wheat)\n",
    "crops = crops.stack().reset_index()\n",
    "crops = crops.join(pd.get_dummies(temp['level_2']))\n",
    "crops = crops.set_index(['Year','ANSI'])[[0,'corn','soy','wheat']]\n",
    "\n",
    "# The second kind of data is Soil Data from Conservation Science Partners via Google Earth Engine(https://developers.google.com/earth-engine/datasets/catalog/CSP_ERGo_1_0_US_lithology)\n",
    "soil = pd.read_csv('data/soilclust.csv')\n",
    "soil['Clust']=soil['Clust'].apply(int)\n",
    "\n",
    "# The third and final data set is climate (temp and rain) data from NASA ORNL DAAC at Oak Ridge National Laboratory collected via Google Earth Engine (https://developers.google.com/earth-engine/datasets/catalog/NASA_ORNL_DAYMET_V3)\n",
    "climate = pd.read_csv('data/climate.csv')\n",
    "climate['Date']=pd.to_datetime(climate['Date'])\n",
    "climate['Month']=climate['Date'].apply(lambda x:x.month)\n",
    "climate['Year']=climate['Date'].apply(lambda x:x.year)\n",
    "climate = climate.set_index(['Clust','Year','Month'])[['prcp','tmax','tmin']].stack().reset_index()\n",
    "climate['label'] = climate['level_3']+'_'+climate['Month'].apply(str)\n",
    "climate['Clust']=climate['Clust'].apply(int)\n",
    "climate=climate.set_index(['Clust','Year','label'])[0]\n",
    "climate=climate.unstack()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x000001C71F515EC8>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 197,
     "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We are looking to predict the yields\n",
    "wheat.plot.hist(bins=100)\n",
    "corn.plot.hist(bins=100)\n",
    "soy.hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1c71ec4b308>"
      ]
     },
     "execution_count": 200,
     "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": [
    "# As shown below, clusters were selected to be geographicly adjacent to each other while at the same time having similar compostions of soils\n",
    "colors = pd.DataFrame(np.random.rand(3,100))\n",
    "colors = soil['Clust'].apply(lambda x:colors[x].values)\n",
    "soil.plot.scatter('INTPTLON','INTPTLAT',s=5,c=colors,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 351,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Now we create a data frame where we can train\n",
    "#first we join crop data with soil data\n",
    "df = crops.reset_index().set_index('ANSI').join(soil.set_index('fips').drop(columns=['INTPTLAT','INTPTLON'])).reset_index()\n",
    "\n",
    "#Next we join the dataframe with climate data\n",
    "columns = df.columns.tolist()\n",
    "columns[-1]='Clust'\n",
    "df.columns = columns\n",
    "df['Clust']=df['Clust']#.apply(str)\n",
    "df=df.set_index(['Clust','Year'])\n",
    "\n",
    "# Join the climate data from that year\n",
    "df = df.join(climate)\n",
    "\n",
    "# Join the climate data from the previous year\n",
    "df = df.join(climate.reset_index().set_index(['Year','Clust']).unstack().shift(1).stack(),rsuffix='_-1')\n",
    "df=df.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "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></th>\n",
       "      <th>index</th>\n",
       "      <th>0</th>\n",
       "      <th>corn</th>\n",
       "      <th>soy</th>\n",
       "      <th>wheat</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>...</th>\n",
       "      <th>tmin_11_-1</th>\n",
       "      <th>tmin_12_-1</th>\n",
       "      <th>tmin_2_-1</th>\n",
       "      <th>tmin_3_-1</th>\n",
       "      <th>tmin_4_-1</th>\n",
       "      <th>tmin_5_-1</th>\n",
       "      <th>tmin_6_-1</th>\n",
       "      <th>tmin_7_-1</th>\n",
       "      <th>tmin_8_-1</th>\n",
       "      <th>tmin_9_-1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Clust</th>\n",
       "      <th>Year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">0</th>\n",
       "      <th>2000</th>\n",
       "      <td>20019</td>\n",
       "      <td>115.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.929101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>20019</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.929101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>20019</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.929101</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>20035</td>\n",
       "      <td>93.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.780487</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>20035</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.780487</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">99</th>\n",
       "      <th>2019</th>\n",
       "      <td>1069</td>\n",
       "      <td>46.2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.712407</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.145205</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.173806</td>\n",
       "      <td>-3.162866</td>\n",
       "      <td>-5.661959</td>\n",
       "      <td>1.276372</td>\n",
       "      <td>1.728747</td>\n",
       "      <td>15.607889</td>\n",
       "      <td>19.334206</td>\n",
       "      <td>20.205403</td>\n",
       "      <td>19.001041</td>\n",
       "      <td>16.536122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019</th>\n",
       "      <td>1099</td>\n",
       "      <td>124.1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.095540</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.602053</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.173806</td>\n",
       "      <td>-3.162866</td>\n",
       "      <td>-5.661959</td>\n",
       "      <td>1.276372</td>\n",
       "      <td>1.728747</td>\n",
       "      <td>15.607889</td>\n",
       "      <td>19.334206</td>\n",
       "      <td>20.205403</td>\n",
       "      <td>19.001041</td>\n",
       "      <td>16.536122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019</th>\n",
       "      <td>13099</td>\n",
       "      <td>202.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.869239</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.120371</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.173806</td>\n",
       "      <td>-3.162866</td>\n",
       "      <td>-5.661959</td>\n",
       "      <td>1.276372</td>\n",
       "      <td>1.728747</td>\n",
       "      <td>15.607889</td>\n",
       "      <td>19.334206</td>\n",
       "      <td>20.205403</td>\n",
       "      <td>19.001041</td>\n",
       "      <td>16.536122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019</th>\n",
       "      <td>13243</td>\n",
       "      <td>213.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.425947</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.574053</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.173806</td>\n",
       "      <td>-3.162866</td>\n",
       "      <td>-5.661959</td>\n",
       "      <td>1.276372</td>\n",
       "      <td>1.728747</td>\n",
       "      <td>15.607889</td>\n",
       "      <td>19.334206</td>\n",
       "      <td>20.205403</td>\n",
       "      <td>19.001041</td>\n",
       "      <td>16.536122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019</th>\n",
       "      <td>13253</td>\n",
       "      <td>213.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.925197</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.173806</td>\n",
       "      <td>-3.162866</td>\n",
       "      <td>-5.661959</td>\n",
       "      <td>1.276372</td>\n",
       "      <td>1.728747</td>\n",
       "      <td>15.607889</td>\n",
       "      <td>19.334206</td>\n",
       "      <td>20.205403</td>\n",
       "      <td>19.001041</td>\n",
       "      <td>16.536122</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>82147 rows × 96 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            index      0  corn  soy  wheat         1    2         3    4    5  \\\n",
       "Clust Year                                                                      \n",
       "0     2000  20019  115.0     1    0      0  0.000000  0.0  0.929101  0.0  0.0   \n",
       "      2000  20019    9.0     0    1      0  0.000000  0.0  0.929101  0.0  0.0   \n",
       "      2000  20019   32.0     0    0      1  0.000000  0.0  0.929101  0.0  0.0   \n",
       "      2000  20035   93.0     1    0      0  0.000000  0.0  0.780487  0.0  0.0   \n",
       "      2000  20035   22.0     0    1      0  0.000000  0.0  0.780487  0.0  0.0   \n",
       "...           ...    ...   ...  ...    ...       ...  ...       ...  ...  ...   \n",
       "99    2019   1069   46.2     0    1      0  0.712407  0.0  0.145205  0.0  0.0   \n",
       "      2019   1099  124.1     1    0      0  0.095540  0.0  0.602053  0.0  0.0   \n",
       "      2019  13099  202.0     1    0      0  0.869239  0.0  0.120371  0.0  0.0   \n",
       "      2019  13243  213.0     1    0      0  0.425947  0.0  0.574053  0.0  0.0   \n",
       "      2019  13253  213.0     1    0      0  0.925197  0.0  0.000000  0.0  0.0   \n",
       "\n",
       "            ...  tmin_11_-1  tmin_12_-1  tmin_2_-1  tmin_3_-1  tmin_4_-1  \\\n",
       "Clust Year  ...                                                            \n",
       "0     2000  ...         NaN         NaN        NaN        NaN        NaN   \n",
       "      2000  ...         NaN         NaN        NaN        NaN        NaN   \n",
       "      2000  ...         NaN         NaN        NaN        NaN        NaN   \n",
       "      2000  ...         NaN         NaN        NaN        NaN        NaN   \n",
       "      2000  ...         NaN         NaN        NaN        NaN        NaN   \n",
       "...         ...         ...         ...        ...        ...        ...   \n",
       "99    2019  ...   -2.173806   -3.162866  -5.661959   1.276372   1.728747   \n",
       "      2019  ...   -2.173806   -3.162866  -5.661959   1.276372   1.728747   \n",
       "      2019  ...   -2.173806   -3.162866  -5.661959   1.276372   1.728747   \n",
       "      2019  ...   -2.173806   -3.162866  -5.661959   1.276372   1.728747   \n",
       "      2019  ...   -2.173806   -3.162866  -5.661959   1.276372   1.728747   \n",
       "\n",
       "            tmin_5_-1  tmin_6_-1  tmin_7_-1  tmin_8_-1  tmin_9_-1  \n",
       "Clust Year                                                         \n",
       "0     2000        NaN        NaN        NaN        NaN        NaN  \n",
       "      2000        NaN        NaN        NaN        NaN        NaN  \n",
       "      2000        NaN        NaN        NaN        NaN        NaN  \n",
       "      2000        NaN        NaN        NaN        NaN        NaN  \n",
       "      2000        NaN        NaN        NaN        NaN        NaN  \n",
       "...               ...        ...        ...        ...        ...  \n",
       "99    2019  15.607889  19.334206  20.205403  19.001041  16.536122  \n",
       "      2019  15.607889  19.334206  20.205403  19.001041  16.536122  \n",
       "      2019  15.607889  19.334206  20.205403  19.001041  16.536122  \n",
       "      2019  15.607889  19.334206  20.205403  19.001041  16.536122  \n",
       "      2019  15.607889  19.334206  20.205403  19.001041  16.536122  \n",
       "\n",
       "[82147 rows x 96 columns]"
      ]
     },
     "execution_count": 350,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 358,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Now we split the data into test and training\n",
    "# The last three years(17,18,and 19) will be reserved for testing while the rest is training\n",
    "train = df[df['Year']<2017].reset_index().dropna()\n",
    "test = df[df['Year']>=2017].reset_index()\n",
    "\n",
    "transformer = Normalizer()\n",
    "\n",
    "trainy = train.iloc[:,4]\n",
    "trainx = train.iloc[:,5:]\n",
    "\n",
    "\n",
    "testy = test.iloc[:,4]\n",
    "testx = test.iloc[:,5:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 363,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_20\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_49 (Dense)             (None, 75)                7125      \n",
      "_________________________________________________________________\n",
      "dense_50 (Dense)             (None, 50)                3800      \n",
      "_________________________________________________________________\n",
      "dense_51 (Dense)             (None, 25)                1275      \n",
      "_________________________________________________________________\n",
      "dense_52 (Dense)             (None, 5)                 130       \n",
      "_________________________________________________________________\n",
      "dense_53 (Dense)             (None, 1)                 6         \n",
      "=================================================================\n",
      "Total params: 12,336\n",
      "Trainable params: 12,336\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Train on 54514 samples, validate on 13629 samples\n",
      "Epoch 1/25\n",
      "54514/54514 [==============================] - 7s 127us/step - loss: 1048.6264 - mse: 1048.6273 - mae: 23.2078 - val_loss: 443.9402 - val_mse: 443.9400 - val_mae: 14.5293\n",
      "Epoch 2/25\n",
      "54514/54514 [==============================] - 7s 120us/step - loss: 692.8113 - mse: 692.8117 - mae: 18.1498 - val_loss: 558.0205 - val_mse: 558.0208 - val_mae: 16.4078\n",
      "Epoch 3/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 681.4340 - mse: 681.4345 - mae: 17.9543 - val_loss: 491.5087 - val_mse: 491.5088 - val_mae: 15.4979\n",
      "Epoch 4/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 662.5500 - mse: 662.5512 - mae: 17.6828 - val_loss: 446.5633 - val_mse: 446.5631 - val_mae: 14.4217\n",
      "Epoch 5/25\n",
      "54514/54514 [==============================] - 6s 119us/step - loss: 654.0931 - mse: 654.0933 - mae: 17.5564 - val_loss: 448.9324 - val_mse: 448.9319 - val_mae: 14.5401\n",
      "Epoch 6/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 651.4502 - mse: 651.4507 - mae: 17.4668 - val_loss: 437.7996 - val_mse: 437.7994 - val_mae: 14.0860\n",
      "Epoch 7/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 641.2483 - mse: 641.2487 - mae: 17.3008 - val_loss: 438.8999 - val_mse: 438.9000 - val_mae: 14.0315\n",
      "Epoch 8/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 635.1487 - mse: 635.1483 - mae: 17.2109 - val_loss: 470.2230 - val_mse: 470.2226 - val_mae: 15.1994\n",
      "Epoch 9/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 631.2542 - mse: 631.2547 - mae: 17.1366 - val_loss: 433.9011 - val_mse: 433.9013 - val_mae: 14.0193\n",
      "Epoch 10/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 628.4504 - mse: 628.4493 - mae: 17.0786 - val_loss: 434.5802 - val_mse: 434.5802 - val_mae: 14.0267\n",
      "Epoch 11/25\n",
      "54514/54514 [==============================] - 6s 117us/step - loss: 628.5299 - mse: 628.5289 - mae: 17.0810 - val_loss: 468.5660 - val_mse: 468.5662 - val_mae: 14.8520\n",
      "Epoch 12/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 623.8756 - mse: 623.8748 - mae: 17.0170 - val_loss: 450.5926 - val_mse: 450.5927 - val_mae: 14.5257\n",
      "Epoch 13/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 622.3394 - mse: 622.3411 - mae: 17.0088 - val_loss: 457.9741 - val_mse: 457.9742 - val_mae: 14.3690\n",
      "Epoch 14/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 619.2214 - mse: 619.2220 - mae: 16.9290 - val_loss: 566.2195 - val_mse: 566.2194 - val_mae: 16.3989\n",
      "Epoch 15/25\n",
      "54514/54514 [==============================] - 6s 115us/step - loss: 618.1208 - mse: 618.1216 - mae: 16.9186 - val_loss: 483.3251 - val_mse: 483.3253 - val_mae: 15.2360\n",
      "Epoch 16/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 614.1826 - mse: 614.1835 - mae: 16.8414 - val_loss: 1230.1982 - val_mse: 1230.1981 - val_mae: 26.7752\n",
      "Epoch 17/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 615.3072 - mse: 615.3075 - mae: 16.8478 - val_loss: 443.1077 - val_mse: 443.1078 - val_mae: 14.2691\n",
      "Epoch 18/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 612.8055 - mse: 612.8056 - mae: 16.8142 - val_loss: 444.8069 - val_mse: 444.8068 - val_mae: 14.0512\n",
      "Epoch 19/25\n",
      "54514/54514 [==============================] - 6s 113us/step - loss: 610.5362 - mse: 610.5365 - mae: 16.7642 - val_loss: 439.5902 - val_mse: 439.5904 - val_mae: 14.1192\n",
      "Epoch 20/25\n",
      "54514/54514 [==============================] - 6s 113us/step - loss: 607.5678 - mse: 607.5677 - mae: 16.7374 - val_loss: 453.8201 - val_mse: 453.8198 - val_mae: 14.3795\n",
      "Epoch 21/25\n",
      "54514/54514 [==============================] - 6s 113us/step - loss: 609.2051 - mse: 609.2051 - mae: 16.7486 - val_loss: 478.1342 - val_mse: 478.1343 - val_mae: 14.8798\n",
      "Epoch 22/25\n",
      "54514/54514 [==============================] - 6s 114us/step - loss: 609.3989 - mse: 609.3979 - mae: 16.7503 - val_loss: 445.7007 - val_mse: 445.7007 - val_mae: 14.3467\n",
      "Epoch 23/25\n",
      "54514/54514 [==============================] - 6s 119us/step - loss: 609.2944 - mse: 609.2946 - mae: 16.7305 - val_loss: 454.0988 - val_mse: 454.0988 - val_mae: 14.6093\n",
      "Epoch 24/25\n",
      "54514/54514 [==============================] - 7s 137us/step - loss: 605.3606 - mse: 605.3602 - mae: 16.6759 - val_loss: 444.8856 - val_mse: 444.8857 - val_mae: 14.2720\n",
      "Epoch 25/25\n",
      "54514/54514 [==============================] - 6s 118us/step - loss: 603.2737 - mse: 603.2736 - mae: 16.6153 - val_loss: 459.1845 - val_mse: 459.1843 - val_mae: 14.7611\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7707594267645265\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "#This is the Neural Network model for the data\n",
    "model = Sequential()\n",
    "model.add(Dense(75,input_dim=94,activation='relu',)) ### kernel_initializer='normal', ))\n",
    "model.add(Dense(50, activation='relu'))\n",
    "model.add(Dense(25, activation='relu'))\n",
    "model.add(Dense(5, activation='relu'))\n",
    "model.add(Dense(1, activation='relu'))\n",
    "model.summary()\n",
    "model.compile(loss='mse', optimizer='adam', metrics=['mse','mae'])\n",
    "\n",
    "model.fit(trainx, trainy, epochs=25, batch_size=10,  verbose=1, validation_split=0.2)\n",
    "\n",
    "predy = model.predict(testx).flatten()\n",
    "plt.hist2d(testy,predy,bins=100)\n",
    "plt.show()\n",
    "\n",
    "print(r2_score(testy,predy))"
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}