HW3-Problem2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Problem 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# import data\n",
    "import pandas as pd\n",
    "df=pd.read_csv('C:/Onedrive/OneDrive - Knights - University of Central Florida/UCF/Courses/Smart City/HW1/Bike-Sharing-Dataset/hour.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Select the required dataset\n",
    "import numpy as np\n",
    "# create design matrix X and target vector y\n",
    "X = np.array(df.iloc[:, [2,4,5,6,7,8,9,10,11,12,13]]) \t# end index is exclusive\n",
    "Y = np.array(df['cnt']) \t# another way of indexing a pandas df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  1.       1.       0.     ...,   0.2879   0.81     0.    ]\n",
      " [  1.       1.       1.     ...,   0.2727   0.8      0.    ]\n",
      " [  1.       1.       2.     ...,   0.2727   0.8      0.    ]\n",
      " ..., \n",
      " [  1.      12.      21.     ...,   0.2576   0.6      0.1642]\n",
      " [  1.      12.      22.     ...,   0.2727   0.56     0.1343]\n",
      " [  1.      12.      23.     ...,   0.2727   0.65     0.1343]]\n"
     ]
    }
   ],
   "source": [
    "print(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# split into train and test\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Scale the datasets\n",
    "from sklearn.preprocessing import StandardScaler  \n",
    "scaler = StandardScaler()  \n",
    "# Don't cheat - fit only on training data\n",
    "scaler.fit(X_train)  \n",
    "X_train = scaler.transform(X_train)  \n",
    "# apply same transformation to test data\n",
    "X_test = scaler.transform(X_test)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.858626202548\n"
     ]
    }
   ],
   "source": [
    "# ANN with lbfgs solver\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "\n",
    "clf = MLPRegressor(solver='lbfgs', alpha=1e-5, hidden_layer_sizes=(100,), max_iter=20000, random_state=1)\n",
    "y_pred = clf.fit(X_train, Y_train).predict(X_test)\n",
    "score_tst = clf.score(X_test,Y_test)\n",
    "print(score_tst)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.523351988066\n"
     ]
    }
   ],
   "source": [
    "# ANN with sgd solver\n",
    "clf = MLPRegressor(solver='sgd', alpha=1e-5, hidden_layer_sizes=(100,), max_iter=20000, random_state=1)\n",
    "y_pred = clf.fit(X_train, Y_train).predict(X_test)\n",
    "score_tst = clf.score(X_test,Y_test)\n",
    "print(score_tst)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.735598517912\n"
     ]
    }
   ],
   "source": [
    "# ANN with adam solver\n",
    "clf = MLPRegressor(solver='adam', alpha=1e-5, hidden_layer_sizes=(100,), max_iter=20000, random_state=1)\n",
    "y_pred = clf.fit(X_train, Y_train).predict(X_test)\n",
    "score_tst = clf.score(X_test,Y_test)\n",
    "print(score_tst)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Develop model based on different hidden layer\n",
    "score_list=[]\n",
    "for i in range (1,5):\n",
    "    clf = MLPRegressor(solver='lbfgs', alpha=1e-5, hidden_layer_sizes=(np.repeat(100,i)), max_iter=20000, random_state=1)\n",
    "    y_pred = clf.fit(X_train, Y_train).predict(X_test)\n",
    "    score_tst = clf.score(X_test,Y_test)\n",
    "    score_list.append(score_tst)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.85862620254837618, 0.78296701882675002, 0.77110685250856426, 0.74043152245553445]\n"
     ]
    }
   ],
   "source": [
    "print(score_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1eace41aeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy level for different kernel function\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "x=np.array([1,2,3,4])\n",
    "\n",
    "plt.plot(x,score_list, label='model a')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# b: ANN model to predict the bikesharing demand for the next hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# remove the last row of X and first obs of Y\n",
    "X1 = X[:-1] \t# end index is exclusive\n",
    "Y1 = Y[1:] \t# another way of indexing a pandas df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  1.       1.       0.     ...,   0.2879   0.81     0.    ]\n",
      " [  1.       1.       1.     ...,   0.2727   0.8      0.    ]\n",
      " [  1.       1.       2.     ...,   0.2727   0.8      0.    ]\n",
      " ..., \n",
      " [  1.      12.      20.     ...,   0.2576   0.6      0.1642]\n",
      " [  1.      12.      21.     ...,   0.2576   0.6      0.1642]\n",
      " [  1.      12.      22.     ...,   0.2727   0.56     0.1343]]\n"
     ]
    }
   ],
   "source": [
    "print(X1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split into train and test\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X1_train, X1_test, Y1_train, Y1_test = train_test_split(X1, Y1, test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Scale the datasets\n",
    "from sklearn.preprocessing import StandardScaler  \n",
    "scaler = StandardScaler()  \n",
    "# Don't cheat - fit only on training data\n",
    "scaler.fit(X1_train)  \n",
    "X1_train = scaler.transform(X1_train)  \n",
    "# apply same transformation to test data\n",
    "X1_test = scaler.transform(X1_test)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.855158714545313, 0.79831935817823396, 0.74610765072156282, 0.70569733351496766]\n"
     ]
    }
   ],
   "source": [
    "# Develop model based on different hidden layer\n",
    "b_score_list=[]\n",
    "for i in range (1,5):\n",
    "    clf = MLPRegressor(solver='lbfgs', alpha=1e-5, hidden_layer_sizes=(np.repeat(100,i)), random_state=1, max_iter=200000)\n",
    "    y_pred = clf.fit(X1_train, Y1_train).predict(X1_test)\n",
    "    score_tst = clf.score(X1_test,Y1_test)\n",
    "    b_score_list.append(score_tst)\n",
    "print(b_score_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMQHm/wMkIl2Z8pFkdwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1eacd04eac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy level for different kernel function\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "x=np.array([1,2,3,4])\n",
    "\n",
    "plt.plot(x,b_score_list, label='model b')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# c: Compare the accuracy of the ANN models developed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWv0UbspQdAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1eacd08b208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy level for different kernel function\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "x=np.array([1,2,3,4])\n",
    "\n",
    "plt.plot(x,score_list, label='model a')\n",
    "plt.plot(x,b_score_list, label='model b')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}