NB6_OptimizingRandomForest.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Objective\n",
    "\n",
    "Our main motive in this section is to optimise our best performing model; Random Forest. For fine tuning the model we will use two cross validation methods provided by sklearn:\n",
    "\n",
    "* GridSearchCV\n",
    "* RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Load necessary libraries__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import sklearn\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.preprocessing import RobustScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "\n",
    "import warnings\n",
    "# ignore warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "\n",
    "# display all columns\n",
    "pd.set_option('display.max_columns', None)\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load data\n",
    "df = pd.read_csv(\"data/data_final.csv\",index_col=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Scaling\n",
    "\n",
    "Feature Scaling is a technique to standardize the independent features present in the data in a fixed range. It is performed during the data pre-processing to handle highly varying magnitudes or values or units. If feature scaling is not done, then a machine learning algorithm tends to weigh greater values, higher and consider smaller values as the lower values, regardless of the unit of the values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#create feature (x) and target (y) variables\n",
    "x = df.drop(\"Price\", axis=1)\n",
    "Y = df[\"Price\"].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#feature scaling using robust scaler\n",
    "scaler = RobustScaler()\n",
    "df = scaler.fit_transform(df.astype(np.float64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, Y_train, Y_test = train_test_split(x,Y, test_size = .20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us create the instance for Random Forest."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#create instance\n",
    "forest = RandomForestRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fit data into model\n",
    "forest.fit(x_train, Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grid Search CV \n",
    "\n",
    "We will adjust the RF hyperparameters to fine tune our model.\n",
    "\n",
    "Most important hyperparameters of Random Forest:\n",
    "\n",
    "* n_estimators = number of trees\n",
    "* max_features = max number of features considered for splitting a node\n",
    "* max_depth = max number of levels in each decision tree\n",
    "* min_samples_split = min number of data points placed in a node before the node is split\n",
    "* min_samples_leaf = min number of data points allowed in a leaf node\n",
    "* bootstrap = method for sampling data points (with or without replacement)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = [{'n_estimators': [10, 25], 'max_features': [5, 10],'max_depth': [10, 50, None], 'bootstrap': [True, False]}]\n",
    "\n",
    "gridcv_forest = GridSearchCV(forest, param_grid, cv=10, scoring='neg_mean_squared_error')\n",
    "gridcv_forest.fit(x_train, Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Best hyperparameters as calculated by GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#find the best model of grid search\n",
    "gridcv_forest.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check accuracy and RMSE on train data\n",
    "print('************** Train Dataset **************')\n",
    "grid_best_train = gridcv_forest.best_estimator_.predict(x_train)\n",
    "errors = abs(grid_best_train - Y_train)\n",
    "mape = np.mean(100 * (errors / Y_train))\n",
    "accuracy = 100 - mape    \n",
    "print('Best model hypertuned by grid search has an accuracy of', round(accuracy, 2),'%')\n",
    "\n",
    "grid_mse = mean_squared_error(Y_train, grid_best_train)\n",
    "grid_rmse = np.sqrt(grid_mse)\n",
    "print('Best model hypertuned by grid search has a RMSE of', round(grid_rmse, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check accuracy and RMSE on test data\n",
    "print('************** Test Dataset **************')\n",
    "grid_best_test = gridcv_forest.best_estimator_.predict(x_test)\n",
    "errors = abs(grid_best_test - Y_test)\n",
    "mape = np.mean(100 * (errors / Y_test))\n",
    "accuracy = 100 - mape    \n",
    "print('Best model hypertuned by grid search has an accuracy of', round(accuracy, 2),'%')\n",
    "\n",
    "grid_mse = mean_squared_error(Y_test, grid_best_test)\n",
    "grid_rmse = np.sqrt(grid_mse)\n",
    "print('Best model hypertuned by grid search has a RMSE of', round(grid_rmse, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Randomized Search CV\n",
    "\n",
    "One of the demerits of using Grid Search is that its too costly, as many combinations are tested. Therefor we proceed with Ramdomized Search CV that uses only difined number of random combinations for each parameter at every iteration as provided by user."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create parameters\n",
    "n_estimators = [int(x) for x in np.linspace(start = 20, stop = 200, num = 5)]\n",
    "max_features = ['auto', 'sqrt']\n",
    "max_depth = [int(x) for x in np.linspace(1, 45, num = 3)]\n",
    "min_samples_split = [5, 10]\n",
    "\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_features': max_features,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use random search cv to find out best hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rcv = RandomizedSearchCV(estimator = forest, param_distributions = random_grid, n_iter = 10, \n",
    "                               cv = 10, verbose=2, n_jobs = -1, scoring='neg_mean_squared_error')\n",
    "# Fit the random search model\n",
    "rcv.fit(x_train, Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Best hyperparameters as calculated by GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# best random serach model \n",
    "rcv.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=23, max_features='sqrt', min_samples_split=10,\n",
       "                      n_estimators=110)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# estimators used\n",
    "rcv.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check accuracy and RMSE on train data\n",
    "print('************** Train Dataset **************')\n",
    "rcv_best_train = rcv.best_estimator_.predict(x_train)\n",
    "errors = abs(rcv_best_train - Y_train)\n",
    "mape = np.mean(100 * (errors / Y_train))\n",
    "accuracy = 100 - mape    \n",
    "print('Best model as calculated by randomized search has an accuracy of', round(accuracy, 2),'%')\n",
    "\n",
    "final_mse = mean_squared_error(Y_train, rcv_best_train)\n",
    "final_rmse = np.sqrt(final_mse)\n",
    "print('Best model as calculated by randomized search has a RMSE of', round(final_rmse, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check accuracy and RMSE on test data\n",
    "print('************** Test Dataset **************')\n",
    "rcv_best_test = rcv.best_estimator_.predict(x_test)\n",
    "errors = abs(rcv_best_test - Y_test)\n",
    "mape = np.mean(100 * (errors / Y_test))\n",
    "accuracy = 100 - mape    \n",
    "print('Best model as calculated by randomized search has an accuracy of', round(accuracy, 2),'%')\n",
    "\n",
    "final_mse = mean_squared_error(Y_test, rcv_best_test)\n",
    "final_rmse = np.sqrt(final_mse)\n",
    "print('Best model as calculated by randomized search has a RMSE of', round(final_rmse, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Importance\n",
    "\n",
    "Using our best parameters as provided by Random Search CV. We display which features plays major role in deciding the price of house."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "importances = rcv.best_estimator_.feature_importances_\n",
    "feature_list = list(x.columns)\n",
    "\n",
    "feature_importance= sorted(zip(importances, feature_list), reverse=True)\n",
    "\n",
    "#create two lists from the previous list of tuples\n",
    "df_imp = pd.DataFrame(feature_importance, columns=['importance', 'feature'])\n",
    "importance= list(df_imp['importance'])\n",
    "feature= list(df_imp['feature'])\n",
    "\n",
    "print(df_imp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the human mind doesn't grasp texual data we try to display this visually."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_values = list(range(len(feature_importance)))\n",
    "\n",
    "plt.figure(figsize=(15,10))\n",
    "plt.bar(x_values, importance, orientation = 'vertical')\n",
    "plt.xticks(x_values, feature, rotation='vertical')\n",
    "plt.ylabel('Importance'); \n",
    "plt.xlabel('Features'); \n",
    "plt.title('Important Features');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion :\n",
    "\n",
    "* The main feature that decides the price of a house is distance, which from the documentation we found out that it is the distance from nearest employment centers (offices)(in total 5). Next, 5 featues that plays a major role are Rooms, Longitude, Latitude, landsize and Southern Metropolitan Region, since we did onehot encode on categorical values, we were able to get this insight. Longitude and Latitude means geographical location which was expected, landsize and number of rooms. No brainer !\n",
    "\n",
    "* The type of the house seems less important than these two categories, as houseAge and Type_h are respectively 10th and 11th in this ranking. Category Type_t (or townhouse) is the less relevant feature in this cluster.\n",
    "\n",
    "* Date features (i.e. Year and the four seasons) are not that important, but in any case the year when the house was sold is slightly more important than the season."
   ]
  }
 ],
 "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": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}