Lesson 09 - Hyperparameters - Solutions.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://www.youtube.com/watch?v=eEGbTd7AG9s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Activation, Embedding, Flatten, LeakyReLU, Merge, BatchNormalization, Dropout\n",
    "from keras.activations import relu, sigmoid\n",
    "from keras.layers import LeakyReLU\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Description of the data:\n",
    "\n",
    "- instant: record index\n",
    "- dteday : date\n",
    "- season : season (1:springer, 2:summer, 3:fall, 4:winter)\n",
    "- yr : year (0: 2011, 1:2012)\n",
    "- mnth : month ( 1 to 12)\n",
    "- hr : hour (0 to 23)\n",
    "- holiday : weather day is holiday or not (extracted from [Web Link])\n",
    "- weekday : day of the week\n",
    "- workingday : if day is neither weekend nor holiday is 1, otherwise is 0.\n",
    "+ weathersit : \n",
    "  - 1: Clear, Few clouds, Partly cloudy, Partly cloudy\n",
    "  - 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist\n",
    "  - 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds\n",
    "  - 4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog\n",
    "- temp : Normalized temperature in Celsius. The values are derived via (t-t_min)/(t_max-t_min), t_min=-8, t_max=+39 (only in hourly scale)\n",
    "- atemp: Normalized feeling temperature in Celsius. The values are derived via (t-t_min)/(t_max-t_min), t_min=-16, t_max=+50 (only in hourly scale)\n",
    "- hum: Normalized humidity. The values are divided to 100 (max)\n",
    "- windspeed: Normalized wind speed. The values are divided to 67 (max)\n",
    "- casual: count of casual users\n",
    "- registered: count of registered users\n",
    "- cnt: count of total rental bikes including both casual and registered"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sachin/anaconda/lib/python3.5/site-packages/pandas/core/ops.py:539: PerformanceWarning: Adding/subtracting array of DateOffsets to Series not vectorized\n",
      "  \"Series not vectorized\", PerformanceWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>season</th>\n",
       "      <th>mnth</th>\n",
       "      <th>hr</th>\n",
       "      <th>holiday</th>\n",
       "      <th>weekday</th>\n",
       "      <th>workingday</th>\n",
       "      <th>weathersit</th>\n",
       "      <th>temp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>cnt</th>\n",
       "      <th>time</th>\n",
       "      <th>t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16</td>\n",
       "      <td>2011-01-01 00:00:00</td>\n",
       "      <td>1.293840e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.0</td>\n",
       "      <td>40</td>\n",
       "      <td>2011-01-01 01:00:00</td>\n",
       "      <td>1.293844e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.0</td>\n",
       "      <td>32</td>\n",
       "      <td>2011-01-01 02:00:00</td>\n",
       "      <td>1.293847e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13</td>\n",
       "      <td>2011-01-01 03:00:00</td>\n",
       "      <td>1.293851e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2011-01-01 04:00:00</td>\n",
       "      <td>1.293854e+18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   season  mnth  hr  holiday  weekday  workingday  weathersit  temp   hum  \\\n",
       "0       1     1   0        0        6           0           1  0.24  0.81   \n",
       "1       1     1   1        0        6           0           1  0.22  0.80   \n",
       "2       1     1   2        0        6           0           1  0.22  0.80   \n",
       "3       1     1   3        0        6           0           1  0.24  0.75   \n",
       "4       1     1   4        0        6           0           1  0.24  0.75   \n",
       "\n",
       "   windspeed  cnt                time             t  \n",
       "0        0.0   16 2011-01-01 00:00:00  1.293840e+18  \n",
       "1        0.0   40 2011-01-01 01:00:00  1.293844e+18  \n",
       "2        0.0   32 2011-01-01 02:00:00  1.293847e+18  \n",
       "3        0.0   13 2011-01-01 03:00:00  1.293851e+18  \n",
       "4        0.0    1 2011-01-01 04:00:00  1.293854e+18  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rides = pd.read_csv('hour.csv')\n",
    "rides['time'] = pd.to_datetime(rides['dteday'])+pd.Series([pd.DateOffset(hours = a ) for a in rides.hr.tolist()])\n",
    "rides.drop(['instant','yr','dteday','casual','registered', 'atemp'],1,inplace=True)\n",
    "rides['t'] = rides.time.values.astype(float)\n",
    "rides.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get **one hot encoded** values of the categorical variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>cnt</th>\n",
       "      <th>time</th>\n",
       "      <th>t</th>\n",
       "      <th>season_1</th>\n",
       "      <th>season_2</th>\n",
       "      <th>season_3</th>\n",
       "      <th>season_4</th>\n",
       "      <th>...</th>\n",
       "      <th>weekday_3</th>\n",
       "      <th>weekday_4</th>\n",
       "      <th>weekday_5</th>\n",
       "      <th>weekday_6</th>\n",
       "      <th>workingday_0</th>\n",
       "      <th>workingday_1</th>\n",
       "      <th>weathersit_1</th>\n",
       "      <th>weathersit_2</th>\n",
       "      <th>weathersit_3</th>\n",
       "      <th>weathersit_4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16</td>\n",
       "      <td>2011-01-01 00:00:00</td>\n",
       "      <td>1.293840e+18</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.0</td>\n",
       "      <td>40</td>\n",
       "      <td>2011-01-01 01:00:00</td>\n",
       "      <td>1.293844e+18</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.0</td>\n",
       "      <td>32</td>\n",
       "      <td>2011-01-01 02:00:00</td>\n",
       "      <td>1.293847e+18</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13</td>\n",
       "      <td>2011-01-01 03:00:00</td>\n",
       "      <td>1.293851e+18</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2011-01-01 04:00:00</td>\n",
       "      <td>1.293854e+18</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 61 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   temp   hum  windspeed  cnt                time             t  season_1  \\\n",
       "0  0.24  0.81        0.0   16 2011-01-01 00:00:00  1.293840e+18         1   \n",
       "1  0.22  0.80        0.0   40 2011-01-01 01:00:00  1.293844e+18         1   \n",
       "2  0.22  0.80        0.0   32 2011-01-01 02:00:00  1.293847e+18         1   \n",
       "3  0.24  0.75        0.0   13 2011-01-01 03:00:00  1.293851e+18         1   \n",
       "4  0.24  0.75        0.0    1 2011-01-01 04:00:00  1.293854e+18         1   \n",
       "\n",
       "   season_2  season_3  season_4      ...       weekday_3  weekday_4  \\\n",
       "0         0         0         0      ...               0          0   \n",
       "1         0         0         0      ...               0          0   \n",
       "2         0         0         0      ...               0          0   \n",
       "3         0         0         0      ...               0          0   \n",
       "4         0         0         0      ...               0          0   \n",
       "\n",
       "   weekday_5  weekday_6  workingday_0  workingday_1  weathersit_1  \\\n",
       "0          0          1             1             0             1   \n",
       "1          0          1             1             0             1   \n",
       "2          0          1             1             0             1   \n",
       "3          0          1             1             0             1   \n",
       "4          0          1             1             0             1   \n",
       "\n",
       "   weathersit_2  weathersit_3  weathersit_4  \n",
       "0             0             0             0  \n",
       "1             0             0             0  \n",
       "2             0             0             0  \n",
       "3             0             0             0  \n",
       "4             0             0             0  \n",
       "\n",
       "[5 rows x 61 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_features = ['season', 'hr', 'mnth', 'holiday', 'weekday', 'workingday', 'weathersit']\n",
    "rides = pd.get_dummies(rides, columns=cat_features)\n",
    "rides.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD8CAYAAACW/ATfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11439d4e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(rides.time[:240],rides.cnt[:240])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sachin/anaconda/lib/python3.5/site-packages/sklearn/utils/validation.py:429: DataConversionWarning: Data with input dtype int64 was converted to float64 by StandardScaler.\n",
      "  warnings.warn(msg, _DataConversionWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temp</th>\n",
       "      <th>hum</th>\n",
       "      <th>windspeed</th>\n",
       "      <th>cnt</th>\n",
       "      <th>time</th>\n",
       "      <th>t</th>\n",
       "      <th>season_1</th>\n",
       "      <th>season_2</th>\n",
       "      <th>season_3</th>\n",
       "      <th>season_4</th>\n",
       "      <th>...</th>\n",
       "      <th>weekday_3</th>\n",
       "      <th>weekday_4</th>\n",
       "      <th>weekday_5</th>\n",
       "      <th>weekday_6</th>\n",
       "      <th>workingday_0</th>\n",
       "      <th>workingday_1</th>\n",
       "      <th>weathersit_1</th>\n",
       "      <th>weathersit_2</th>\n",
       "      <th>weathersit_3</th>\n",
       "      <th>weathersit_4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.334648</td>\n",
       "      <td>0.947372</td>\n",
       "      <td>-1.553889</td>\n",
       "      <td>-0.956339</td>\n",
       "      <td>2011-01-01 00:00:00</td>\n",
       "      <td>-1.745435</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.438516</td>\n",
       "      <td>0.895539</td>\n",
       "      <td>-1.553889</td>\n",
       "      <td>-0.824022</td>\n",
       "      <td>2011-01-01 01:00:00</td>\n",
       "      <td>-1.745236</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.438516</td>\n",
       "      <td>0.895539</td>\n",
       "      <td>-1.553889</td>\n",
       "      <td>-0.868128</td>\n",
       "      <td>2011-01-01 02:00:00</td>\n",
       "      <td>-1.745038</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.334648</td>\n",
       "      <td>0.636370</td>\n",
       "      <td>-1.553889</td>\n",
       "      <td>-0.972879</td>\n",
       "      <td>2011-01-01 03:00:00</td>\n",
       "      <td>-1.744840</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.334648</td>\n",
       "      <td>0.636370</td>\n",
       "      <td>-1.553889</td>\n",
       "      <td>-1.039037</td>\n",
       "      <td>2011-01-01 04:00:00</td>\n",
       "      <td>-1.744641</td>\n",
       "      <td>1</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 61 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       temp       hum  windspeed       cnt                time         t  \\\n",
       "0 -1.334648  0.947372  -1.553889 -0.956339 2011-01-01 00:00:00 -1.745435   \n",
       "1 -1.438516  0.895539  -1.553889 -0.824022 2011-01-01 01:00:00 -1.745236   \n",
       "2 -1.438516  0.895539  -1.553889 -0.868128 2011-01-01 02:00:00 -1.745038   \n",
       "3 -1.334648  0.636370  -1.553889 -0.972879 2011-01-01 03:00:00 -1.744840   \n",
       "4 -1.334648  0.636370  -1.553889 -1.039037 2011-01-01 04:00:00 -1.744641   \n",
       "\n",
       "   season_1  season_2  season_3  season_4      ...       weekday_3  weekday_4  \\\n",
       "0         1         0         0         0      ...               0          0   \n",
       "1         1         0         0         0      ...               0          0   \n",
       "2         1         0         0         0      ...               0          0   \n",
       "3         1         0         0         0      ...               0          0   \n",
       "4         1         0         0         0      ...               0          0   \n",
       "\n",
       "   weekday_5  weekday_6  workingday_0  workingday_1  weathersit_1  \\\n",
       "0          0          1             1             0             1   \n",
       "1          0          1             1             0             1   \n",
       "2          0          1             1             0             1   \n",
       "3          0          1             1             0             1   \n",
       "4          0          1             1             0             1   \n",
       "\n",
       "   weathersit_2  weathersit_3  weathersit_4  \n",
       "0             0             0             0  \n",
       "1             0             0             0  \n",
       "2             0             0             0  \n",
       "3             0             0             0  \n",
       "4             0             0             0  \n",
       "\n",
       "[5 rows x 61 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quant_features = ['temp', 'hum', 'windspeed','t']\n",
    "x_scaler = StandardScaler()\n",
    "rides[quant_features] = x_scaler.fit_transform(rides[quant_features].values)\n",
    "\n",
    "y_scaler = StandardScaler()\n",
    "rides['cnt'] = y_scaler.fit_transform(rides['cnt'].values[:,None])\n",
    "\n",
    "rides.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initial model (1 hidden layer with Relu activations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Split data into testing and training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_data = rides[:-21*24]\n",
    "test_data = rides[-21*24:]\n",
    "\n",
    "train_x = train_data.drop(['cnt','time'],axis=1).values\n",
    "test_x = test_data.drop(['cnt','time'],axis=1).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16875, 59)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(train_x.shape[1]//2,input_dim=train_x.shape[1]))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adadelta', loss='mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.8676     \n",
      "Epoch 2/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.6291     \n",
      "Epoch 3/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.4790     \n",
      "Epoch 4/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.3694     \n",
      "Epoch 5/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.3032     \n",
      "Epoch 6/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.2589     \n",
      "Epoch 7/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.2256     \n",
      "Epoch 8/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.2027     \n",
      "Epoch 9/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1857     \n",
      "Epoch 10/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1722     \n",
      "Epoch 11/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1608     \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x11450eda0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the model\n",
    "model.fit(train_x, train_data['cnt'], batch_size=256, epochs=11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_1 (Dense)              (None, 29)                1740      \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 29)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1)                 30        \n",
      "=================================================================\n",
      "Total params: 1,770.0\n",
      "Trainable params: 1,770.0\n",
      "Non-trainable params: 0.0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAEyCAYAAAAIiE2qAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114537ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_y = model.predict(test_x, batch_size=256)\n",
    "plt.figure(figsize=(12,5))\n",
    "plt.plot(test_data.time, y_scaler.inverse_transform(test_data.cnt), label='Actual')\n",
    "plt.plot(test_data.time, y_scaler.inverse_transform(pred_y), label='Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "256/504 [==============>...............] - ETA: 0s"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.23559209562483288"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(test_x, test_data['cnt'], batch_size=256)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#! pip install graphviz\n",
    "#! pip install pydot-ng"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Uncomment the pip commands above if you want to get the image below. i.e. If the command below chucks an error uncomment the commands above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<svg height=\"264pt\" viewBox=\"0.00 0.00 176.00 264.00\" width=\"176pt\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g class=\"graph\" id=\"graph0\" transform=\"scale(1 1) rotate(0) translate(4 260)\">\n",
       "<title>G</title>\n",
       "<polygon fill=\"white\" points=\"-4,4 -4,-260 172,-260 172,4 -4,4\" stroke=\"none\"/>\n",
       "<!-- 4636838880 -->\n",
       "<g class=\"node\" id=\"node1\"><title>4636838880</title>\n",
       "<polygon fill=\"none\" points=\"0,-219.5 0,-255.5 168,-255.5 168,-219.5 0,-219.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"84\" y=\"-233.8\">dense_1_input: InputLayer</text>\n",
       "</g>\n",
       "<!-- 4636839608 -->\n",
       "<g class=\"node\" id=\"node2\"><title>4636839608</title>\n",
       "<polygon fill=\"none\" points=\"31.5,-146.5 31.5,-182.5 136.5,-182.5 136.5,-146.5 31.5,-146.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"84\" y=\"-160.8\">dense_1: Dense</text>\n",
       "</g>\n",
       "<!-- 4636838880&#45;&gt;4636839608 -->\n",
       "<g class=\"edge\" id=\"edge1\"><title>4636838880-&gt;4636839608</title>\n",
       "<path d=\"M84,-219.313C84,-211.289 84,-201.547 84,-192.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"87.5001,-192.529 84,-182.529 80.5001,-192.529 87.5001,-192.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4636839720 -->\n",
       "<g class=\"node\" id=\"node3\"><title>4636839720</title>\n",
       "<polygon fill=\"none\" points=\"8,-73.5 8,-109.5 160,-109.5 160,-73.5 8,-73.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"84\" y=\"-87.8\">activation_1: Activation</text>\n",
       "</g>\n",
       "<!-- 4636839608&#45;&gt;4636839720 -->\n",
       "<g class=\"edge\" id=\"edge2\"><title>4636839608-&gt;4636839720</title>\n",
       "<path d=\"M84,-146.313C84,-138.289 84,-128.547 84,-119.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"87.5001,-119.529 84,-109.529 80.5001,-119.529 87.5001,-119.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4720096368 -->\n",
       "<g class=\"node\" id=\"node4\"><title>4720096368</title>\n",
       "<polygon fill=\"none\" points=\"31.5,-0.5 31.5,-36.5 136.5,-36.5 136.5,-0.5 31.5,-0.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"84\" y=\"-14.8\">dense_2: Dense</text>\n",
       "</g>\n",
       "<!-- 4636839720&#45;&gt;4720096368 -->\n",
       "<g class=\"edge\" id=\"edge3\"><title>4636839720-&gt;4720096368</title>\n",
       "<path d=\"M84,-73.3129C84,-65.2895 84,-55.5475 84,-46.5691\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"87.5001,-46.5288 84,-36.5288 80.5001,-46.5289 87.5001,-46.5288\" stroke=\"black\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# from IPython.display import SVG\n",
    "# from keras.utils.vis_utils import model_to_dot\n",
    "\n",
    "# SVG(model_to_dot(model).create(prog='dot', format='svg'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_1 (Dense)              (None, 29)                1740      \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 29)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1)                 30        \n",
      "=================================================================\n",
      "Total params: 1,770.0\n",
      "Trainable params: 1,770.0\n",
      "Non-trainable params: 0.0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Two Layer hidden network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16875, 59)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.9595     \n",
      "Epoch 2/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.5774     \n",
      "Epoch 3/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.4192     \n",
      "Epoch 4/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.3131     \n",
      "Epoch 5/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.2526     \n",
      "Epoch 6/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.2160     \n",
      "Epoch 7/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1889     \n",
      "Epoch 8/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1678     \n",
      "Epoch 9/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1496     \n",
      "Epoch 10/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1345     \n",
      "Epoch 11/11\n",
      "16875/16875 [==============================] - 0s - loss: 0.1217     \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1188026a0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "h_nodes = [40, 20]\n",
    "model = Sequential()\n",
    "\n",
    "# The two hidden layers (first layer needs the dimensions of x)\n",
    "model.add(Dense(h_nodes[0],input_dim=train_x.shape[1]))\n",
    "model.add(Activation(relu))\n",
    "model.add(Dense(h_nodes[1]))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adadelta', loss='mse')\n",
    "model.fit(train_x, train_data['cnt'], batch_size=256, epochs=11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "256/504 [==============>...............] - ETA: 0s"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.25022489731273956"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(test_x, test_data['cnt'], batch_size=256)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_3 (Dense)              (None, 40)                2400      \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 40)                0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 20)                820       \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 20)                0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 21        \n",
      "=================================================================\n",
      "Total params: 3,241.0\n",
      "Trainable params: 3,241.0\n",
      "Non-trainable params: 0.0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<svg height=\"410pt\" viewBox=\"0.00 0.00 183.00 410.00\" width=\"183pt\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g class=\"graph\" id=\"graph0\" transform=\"scale(1 1) rotate(0) translate(4 406)\">\n",
       "<title>G</title>\n",
       "<polygon fill=\"white\" points=\"-4,4 -4,-406 179,-406 179,4 -4,4\" stroke=\"none\"/>\n",
       "<!-- 4814568192 -->\n",
       "<g class=\"node\" id=\"node1\"><title>4814568192</title>\n",
       "<polygon fill=\"none\" points=\"0,-365.5 0,-401.5 175,-401.5 175,-365.5 0,-365.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-379.8\">dense_34_input: InputLayer</text>\n",
       "</g>\n",
       "<!-- 4814568752 -->\n",
       "<g class=\"node\" id=\"node2\"><title>4814568752</title>\n",
       "<polygon fill=\"none\" points=\"31.5,-292.5 31.5,-328.5 143.5,-328.5 143.5,-292.5 31.5,-292.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-306.8\">dense_34: Dense</text>\n",
       "</g>\n",
       "<!-- 4814568192&#45;&gt;4814568752 -->\n",
       "<g class=\"edge\" id=\"edge1\"><title>4814568192-&gt;4814568752</title>\n",
       "<path d=\"M87.5,-365.313C87.5,-357.289 87.5,-347.547 87.5,-338.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"91.0001,-338.529 87.5,-328.529 84.0001,-338.529 91.0001,-338.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4814569088 -->\n",
       "<g class=\"node\" id=\"node3\"><title>4814569088</title>\n",
       "<polygon fill=\"none\" points=\"8,-219.5 8,-255.5 167,-255.5 167,-219.5 8,-219.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-233.8\">activation_18: Activation</text>\n",
       "</g>\n",
       "<!-- 4814568752&#45;&gt;4814569088 -->\n",
       "<g class=\"edge\" id=\"edge2\"><title>4814568752-&gt;4814569088</title>\n",
       "<path d=\"M87.5,-292.313C87.5,-284.289 87.5,-274.547 87.5,-265.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"91.0001,-265.529 87.5,-255.529 84.0001,-265.529 91.0001,-265.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4814700672 -->\n",
       "<g class=\"node\" id=\"node4\"><title>4814700672</title>\n",
       "<polygon fill=\"none\" points=\"31.5,-146.5 31.5,-182.5 143.5,-182.5 143.5,-146.5 31.5,-146.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-160.8\">dense_35: Dense</text>\n",
       "</g>\n",
       "<!-- 4814569088&#45;&gt;4814700672 -->\n",
       "<g class=\"edge\" id=\"edge3\"><title>4814569088-&gt;4814700672</title>\n",
       "<path d=\"M87.5,-219.313C87.5,-211.289 87.5,-201.547 87.5,-192.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"91.0001,-192.529 87.5,-182.529 84.0001,-192.529 91.0001,-192.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4814700728 -->\n",
       "<g class=\"node\" id=\"node5\"><title>4814700728</title>\n",
       "<polygon fill=\"none\" points=\"8,-73.5 8,-109.5 167,-109.5 167,-73.5 8,-73.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-87.8\">activation_19: Activation</text>\n",
       "</g>\n",
       "<!-- 4814700672&#45;&gt;4814700728 -->\n",
       "<g class=\"edge\" id=\"edge4\"><title>4814700672-&gt;4814700728</title>\n",
       "<path d=\"M87.5,-146.313C87.5,-138.289 87.5,-128.547 87.5,-119.569\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"91.0001,-119.529 87.5,-109.529 84.0001,-119.529 91.0001,-119.529\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4814701288 -->\n",
       "<g class=\"node\" id=\"node6\"><title>4814701288</title>\n",
       "<polygon fill=\"none\" points=\"31.5,-0.5 31.5,-36.5 143.5,-36.5 143.5,-0.5 31.5,-0.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"87.5\" y=\"-14.8\">dense_36: Dense</text>\n",
       "</g>\n",
       "<!-- 4814700728&#45;&gt;4814701288 -->\n",
       "<g class=\"edge\" id=\"edge5\"><title>4814700728-&gt;4814701288</title>\n",
       "<path d=\"M87.5,-73.3129C87.5,-65.2895 87.5,-55.5475 87.5,-46.5691\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"91.0001,-46.5288 87.5,-36.5288 84.0001,-46.5289 91.0001,-46.5288\" stroke=\"black\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SVG(model_to_dot(model).create(prog='dot', format='svg'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Best Hyper parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.wrappers.scikit_learn import KerasRegressor\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What is Cross Validation?\n",
    "https://stats.stackexchange.com/questions/1826/cross-validation-in-plain-english"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import keras.backend as K\n",
    "\n",
    "def create_model(layers, activation):\n",
    "    model = Sequential()\n",
    "    for i, nodes in enumerate(layers):\n",
    "        if i==0:\n",
    "            model.add(Dense(nodes,input_dim=train_x.shape[1]))\n",
    "            model.add(Activation(activation))\n",
    "        else:\n",
    "            model.add(Dense(nodes))\n",
    "            model.add(Activation(activation))\n",
    "    model.add(Dense(1)) # Note: no activation beyond this point\n",
    "    \n",
    "    model.compile(optimizer='adadelta', loss='mse')\n",
    "    return model\n",
    "    \n",
    "model = KerasRegressor(build_fn=create_model, verbose=0)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "layers = [[30], [40, 20], [45, 30, 15]]\n",
    "activations = [sigmoid, relu]\n",
    "param_grid = dict(layers=layers, activation=activations, batch_size = [128, 256], epochs=[30])\n",
    "grid = GridSearchCV(estimator=model, param_grid=param_grid, scoring='neg_mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_result = grid.fit(train_x, train_data['cnt'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-0.13536392705699826,\n",
       " {'activation': <function keras.activations.relu>,\n",
       "  'batch_size': 128,\n",
       "  'epochs': 30,\n",
       "  'layers': [45, 30, 15]}]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[grid_result.best_score_, grid_result.best_params_]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.354904 (0.155346) with: {'batch_size': 128, 'layers': [30], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.327239 (0.149877) with: {'batch_size': 128, 'layers': [40, 20], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.316260 (0.161740) with: {'batch_size': 128, 'layers': [45, 30, 15], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.376903 (0.187427) with: {'batch_size': 256, 'layers': [30], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.350373 (0.192372) with: {'batch_size': 256, 'layers': [40, 20], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.371596 (0.221870) with: {'batch_size': 256, 'layers': [45, 30, 15], 'activation': <function sigmoid at 0x112e911e0>, 'epochs': 30}\n",
      "-0.172043 (0.052121) with: {'batch_size': 128, 'layers': [30], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n",
      "-0.152897 (0.074760) with: {'batch_size': 128, 'layers': [40, 20], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n",
      "-0.135364 (0.037045) with: {'batch_size': 128, 'layers': [45, 30, 15], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n",
      "-0.220772 (0.073936) with: {'batch_size': 256, 'layers': [30], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n",
      "-0.171235 (0.076588) with: {'batch_size': 256, 'layers': [40, 20], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n",
      "-0.201477 (0.143043) with: {'batch_size': 256, 'layers': [45, 30, 15], 'activation': <function relu at 0x112e910d0>, 'epochs': 30}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sachin/anaconda/lib/python3.5/site-packages/sklearn/model_selection/_search.py:662: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "for params, mean_score, scores in grid_result.grid_scores_:\n",
    "    print(\"%f (%f) with: %r\" % (scores.mean(), scores.std(), params))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Finally predict the values on the actual test set with given params:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred_y = grid.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.12697137067941835"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "error = test_data['cnt'] - pred_y\n",
    "np.mean(np.square(error))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  },
  "latex_envs": {
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 0
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}