Final Project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np \n",
    "from sklearn import linear_model\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# This part of project is final part after data fetching, and data cleaning, shuffling the data is also done,\n",
    "# to prevent overfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "shuffled_data = pd.read_csv('Clean_shuffled_data_final.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Available  Duration  Total_Fare  no_of_days_to_flight  Airways_Air India  \\\n",
      "0          7       750        6792                    33                  0   \n",
      "1          2       130        3197                    56                  0   \n",
      "2          7       320        8173                    41                  0   \n",
      "3         19       265        4542                    29                  0   \n",
      "4          5       725        7170                    28                  0   \n",
      "\n",
      "   Airways_AirAsia  Airways_Go Air  Airways_IndiGo  Airways_Jet Airways  \\\n",
      "0                0               0               0                    1   \n",
      "1                0               0               0                    0   \n",
      "2                0               0               0                    1   \n",
      "3                0               0               1                    0   \n",
      "4                0               0               0                    1   \n",
      "\n",
      "   Airways_Spicejet      ...       Dept_Time_BSR  Dept_Time_MDSS  \\\n",
      "0                 0      ...                   0               1   \n",
      "1                 0      ...                   0               1   \n",
      "2                 0      ...                   0               0   \n",
      "3                 0      ...                   0               0   \n",
      "4                 0      ...                   0               0   \n",
      "\n",
      "   Dept_Time_SRMD  Dept_Time_SSMN  Arr_Time_BSR  Arr_Time_MDSS  Arr_Time_SRMD  \\\n",
      "0               0               0             0              0              0   \n",
      "1               0               0             0              0              0   \n",
      "2               1               0             0              1              0   \n",
      "3               1               0             0              0              1   \n",
      "4               1               0             0              1              0   \n",
      "\n",
      "   Arr_Time_SSMN  Hopping_False  Hopping_True  \n",
      "0              1              0             1  \n",
      "1              1              1             0  \n",
      "2              0              0             1  \n",
      "3              0              1             0  \n",
      "4              0              0             1  \n",
      "\n",
      "[5 rows x 21 columns]\n"
     ]
    }
   ],
   "source": [
    "# features considered, \n",
    "# Airways, \n",
    "# Arrival Time, Departure Time\n",
    "# Flight Duration\n",
    "# Days to booking and Flight\n",
    "# Hopping\n",
    "\n",
    "print shuffled_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Flight_Price = shuffled_data['Total_Fare']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del shuffled_data['Total_Fare']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "shuffled_data_train, shuffled_data_test, Flight_Price_train, Flight_Price_test = train_test_split(shuffled_data,Flight_Price, test_size=0.05, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python2.7/dist-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.linear_model import LinearRegression, LogisticRegression\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "import pandas as pd\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "models = [LinearRegression(),\n",
    "              RandomForestRegressor(n_estimators=100, max_features='sqrt'),\n",
    "              KNeighborsRegressor(n_neighbors=6),\n",
    "              SVR(kernel='linear'),\n",
    "              LogisticRegression()\n",
    "              ]\n",
    "TestModels = pd.DataFrame()\n",
    "tmp = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_name_list = list()\n",
    "r2_score_list = list()\n",
    "mean_abs_err_list = list()\n",
    "mean_sq_err_list = list()\n",
    "i = 0\n",
    "for model in models:\n",
    "    m = str(model)\n",
    "    tmp['Model'] = m[:m.index('(')]\n",
    "    model.fit(shuffled_data_train, Flight_Price_train)\n",
    "    predicted_value = model.predict(shuffled_data_test)\n",
    "    indi_r2_score = r2_score(Flight_Price_test, predicted_value )\n",
    "    r2_score_list.insert(i,indi_r2_score)\n",
    "#     model_name_list.insert(i,m)\n",
    "#     print confusion_matrix(Flight_Price_test,predicted_value)\n",
    "    mae =  mean_absolute_error(Flight_Price_test,predicted_value)\n",
    "    mse = mean_squared_error(Flight_Price_test,predicted_value)\n",
    "    mean_abs_err_list.insert(i,mae)\n",
    "    mean_sq_err_list.insert(i,mse)\n",
    "    i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model_name = ('LinearRegression','RandomForestRegressor','KNeighborsRegressor','SVR','LogisticRegression')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### The r2_score function computes R², the coefficient of determination. It provides a measure of how well future samples are likely to be predicted by the model. Best possible score is 1.0 and it can be negative (because the model can be arbitrarily worse). A constant model that always predicts the expected value of y, disregarding the input features, would get a R^2 score of 0.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a4678ec10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting R2 score values in diff Supervised Regression Algorithms\n",
    "\n",
    "y_pos = np.arange(len(model_name))\n",
    "plt.figure(figsize=(15,5))\n",
    "bar1 = plt.bar(y_pos, r2_score_list, align='center', alpha=0.7)\n",
    "\n",
    "for i,rect in enumerate(bar1):\n",
    "    height = rect.get_height()\n",
    "    plt.text(rect.get_x() + rect.get_width()/2.0, height, '%.3f' % float(r2_score_list[i]), ha='center', va='bottom')\n",
    "\n",
    "plt.xticks(y_pos, model_name)\n",
    "plt.ylabel('r2_score')\n",
    "plt.title('Comparison of R2_Score in diff Supervised ML Algorithms')\n",
    "plt.savefig('r2_score_vs_algo.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a46baed90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pos = np.arange(len(model_name))\n",
    "plt.figure(figsize=(15,5))\n",
    "bar1 = plt.bar(y_pos, mean_abs_err_list, align='center', alpha=0.7)\n",
    "\n",
    "for i,rect in enumerate(bar1):\n",
    "    height = rect.get_height()\n",
    "    plt.text(rect.get_x() + rect.get_width()/2.0, height, '%.3f' % float(mean_abs_err_list[i]), ha='center', va='bottom')\n",
    "\n",
    "plt.xticks(y_pos, model_name)\n",
    "plt.ylabel('mean abs error')\n",
    "plt.title('Comparison of Mean Absolute Error in diff Supervised ML Algorithms')\n",
    "plt.savefig('MAE_vs_algo.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a469ed550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pos = np.arange(len(model_name))\n",
    "plt.figure(figsize=(15,5))\n",
    "bar1 = plt.bar(y_pos, mean_sq_err_list, align='center', alpha=0.7)\n",
    "\n",
    "for i,rect in enumerate(bar1):\n",
    "    height = rect.get_height()\n",
    "    plt.text(rect.get_x() + rect.get_width()/2.0, height, '%.3f' % float(mean_sq_err_list[i]), ha='center', va='bottom')\n",
    "\n",
    "plt.xticks(y_pos, model_name)\n",
    "plt.ylabel('mean square error')\n",
    "plt.title('Comparison of Mean Square Error in diff Supervised ML Algorithms')\n",
    "plt.savefig('MSE_vs_algo.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Now, seeing the R2_Score, Mean Squared Error and Mean Abs Error, We come to conclusion that Random Forest with 100 Trees is best fit for this regression problem.\n",
    "\n",
    "### Now looking into RandomForests Regressor for Fit line"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "RF_model = RandomForestRegressor(n_estimators=100, max_features='sqrt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "RF_model = RF_model.fit(shuffled_data_train, Flight_Price_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a46a9ff90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "RF_model_pred_flight_price = RF_model.predict(shuffled_data_test)\n",
    "\n",
    "plt.figure(figsize=(15,5))\n",
    "plt.scatter(RF_model_pred_flight_price, Flight_Price_test)\n",
    "plt.xlabel('Prediction')\n",
    "plt.ylabel('Real Value')\n",
    "\n",
    "diagonal = np.linspace(0, np.max(Flight_Price_test), 100)\n",
    "plt.plot(diagonal, diagonal, '-r')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# This gives us the RandomForest Visualisation by printing Tree in .Dot file \n",
    "from sklearn import tree\n",
    "i_tree = 0\n",
    "for tree_in_forest in RF_model.estimators_:\n",
    "    with open('tree_' + str(i_tree) + '.dot', 'w') as my_file:\n",
    "        my_file = tree.export_graphviz(tree_in_forest, out_file = my_file)\n",
    "    i_tree = i_tree + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# This creates information about Forest trees. \n",
    "# One such can be seen at -> https://imgur.com/a/cYZQb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}