Milestone-3-RandomForests_cp.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Milestone - Random Forets\n",
    "\n",
    "This notebook corresponds to the third stage of the Machine Learning final project, as part of the Copernicus Master in Digital Earth, in the Data Science track at University of South Brittany, Vannes, France by Candela Sol PELLIZA & Rajeswari PARASA.\n",
    "\n",
    "We demonstrate the use of Random Forests for the regression probelm of predicting housing prices in Ames, Iowa. In this workflow, we conduct hyperparameter tuning using OOB score, grid search and cross validation; evaluate feature importance; perform feature selection; and then evaluate the model on the test set. We then compare the results of the model with and without feature selection, and also with the linear regressor previously modelled."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Data importing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.inspection import permutation_importance\n",
    "from sklearn.model_selection import GridSearchCV, KFold, cross_val_score\n",
    "from sklearn.metrics import mean_squared_log_error, mean_squared_error, r2_score, make_scorer\n",
    "import seaborn as sns\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first step consists on importing the csv fieles for train and test dataset. We will skip the train and test split, as they were already divided in the previous preprocessing steps, and used the test and train files we exported in Milestone 2 Notebook.\n",
    "\n",
    "The csv datasets files can be obtained by running the mentioned Notebook or can be directly downloaded from the [GitHub Repository](https://github.com/rajesvariparasa/ml-processing-pipeline-for-predicting-houseprices/tree/main/OpenData)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import train and test datasets in csv format\n",
    "data_train = pd.read_csv('OpenData/ames_train_clean.csv')\n",
    "data_test = pd.read_csv('OpenData/ames_test_clean.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a second step, we will connvert the train and test datasets into NumPy arrays, extracting also the variables' names as an independent vector."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating train and test dataframes\n",
    "X_train = data_train.drop(columns = ['PID', 'SoldPrice'])\n",
    "X_test = data_test.drop(columns = ['PID', 'SoldPrice'])\n",
    "X_names = data_train.drop(columns = ['PID', 'SoldPrice']).columns.tolist()\n",
    "\n",
    "Y_train = data_train['SoldPrice']\n",
    "Y_test = data_test['SoldPrice']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After that, as done in the previous notebook, we will define our own score metrics, which correspond to the accuracy metrics we defined at the start of the project (`RMSE`, `RMSLE` and `R2`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function for calculating RMSE\n",
    "def rmse(y_test, y_pred):\n",
    "    \"\"\"\n",
    "    This function calculates the RMSE between the predicted and the actual values\n",
    "    -------------------------------------------\n",
    "    Parameters:\n",
    "    y_test (array): Array of actual values\n",
    "    y_pred (array): Array of predicted values\n",
    "\n",
    "    Returns:\n",
    "    float: RMSE value\n",
    "    \"\"\"\n",
    "    \n",
    "    return mean_squared_error(y_test, y_pred, squared=False)\n",
    "# Define a custom scorer for RMSE\n",
    "rmse_scorer = make_scorer(rmse, greater_is_better=False)\n",
    "\n",
    "def rmsle(y_test, y_pred):\n",
    "    \"\"\"\n",
    "    This function calculates the RMSLE between the predicted and the actual values\n",
    "    -------------------------------------------\n",
    "    Parameters:\n",
    "    y_test (array): Array of actual values\n",
    "    y_pred (array): Array of predicted values\n",
    "    \n",
    "    Returns:\n",
    "    float: RMSLE value\n",
    "    \"\"\"\n",
    "    return mean_squared_log_error(y_test, y_pred, squared=False)\n",
    "\n",
    "# Define a custom scorer for RMSLE\n",
    "rmsle_scorer = make_scorer(rmsle, greater_is_better=False)\n",
    "\n",
    "# Define a custom scorer for R2\n",
    "r2_scorer = make_scorer(r2_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Hyperparameter tuning\n",
    "In this notebook, we will apply a random forest model to predict the properties' sold price, as we did previously with the linear regression models. For this, as we have a regression problem, we will use the scikit function `RandomForestRegressor()`\n",
    "\n",
    "The random forest model requires tuning two main parameters:\n",
    "\n",
    "- **`n_estimators`:** number of trees in the forest\n",
    "- **`max_features`:** number of features to take into account when looking for the best split\n",
    "\n",
    "The tuning will be done in two steps: \n",
    "\n",
    "First we will tune the `n_estimators` parameter, with `max_features=None.`\n",
    "Once we have found the `n_estimators` optimal parameter, we will tune `max_features` using `gridsearchCV` to find the value that gives the best performance. \n",
    "Finally we will check again the `n_estimators` parameter, with the newly defined `max_features`, to see if an adjustment is necessary."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.1. Tuning `n_estimators`\n",
    "We will first tune the `n_estimators` paramater. For this step, `max_features` is set to None, which will consider the total number of variables. \n",
    "\n",
    "We will calculate the OOB error for random forest models with different numbers of trees, from a minimum of 100 to a maximum of 1500 in steps of 50. Finally, the OOB error will be plotted for each n_estimators tested value for visually inspecting at which value the error stabilizes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\cande\\anaconda3\\envs\\scientific\\lib\\site-packages\\sklearn\\ensemble\\_forest.py:578: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable OOB estimates.\n",
      "  warn(\n"
     ]
    }
   ],
   "source": [
    "# Create an empty list for OBB error and i values iteration\n",
    "error_rate1 = []\n",
    "i_list1 = []\n",
    "\n",
    "# Define a loop to iterate over the different values of n_estimators\n",
    "for i in range(10, 300, 10):\n",
    "    \n",
    "    # warm_start = true: reuse the solution of the previous iteration, adding more estimators (otherwise a whole new tree would be created)\n",
    "    # random_state = 33: allows reproducibility, giving the same results each time the code block is ran\n",
    "    # oob_score = true: sets oob score as the metric for generalisation score\n",
    "   \n",
    "    rf = RandomForestRegressor(warm_start=True, n_estimators=i, max_features=None, random_state=33, oob_score=True)\n",
    "    rf.fit(X_train, Y_train)\n",
    "\n",
    "    # oob_score attribute returns the accuracy, but we are interested on the error itself (then: 1-oob_score)\n",
    "    oob_error = 1 - rf.oob_score_\n",
    "\n",
    "    # Append OOB error rate and i values to the lists\n",
    "    error_rate1.append(oob_error)\n",
    "    i_list1.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot n_esitmators vs OOB error\n",
    "\n",
    "plt.plot(i_list1, error_rate1, label=\"OOB error rate\")\n",
    "plt.ylim(0.1, 1.1*max(error_rate1)) \n",
    "plt.xlabel(\"n_estimators\")\n",
    "plt.ylabel(\"OOB error rate\")\n",
    "plt.title(\"OOB error rate vs n_estimators\")\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the obtained graphic, we can conclude that the OOB error gets stabilized when `n_estimators` reaches 200. Consequently, we will use this value for the model and for calibrating the next hyperparameter - `max_features`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.2 Tuning max_features\n",
    "Once we have tuned the `n_estimators` parameter, we will use that for tuning the `max_features` parameter. \n",
    "\n",
    "We will tune this parameter using Cross Validation technique, with 5 k-folds, which means that 20% of the data will be used for testing purpose at each iteration, which is a generalized standard value for this task.\n",
    "\n",
    "We will tune the hyperparameter based on the acuracy and std scores of our main accuracy metric defined at the beggining of the project, `RMSLE`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  params  mean_test_RMSLE  std_test_RMSLE\n",
      "0  {'max_features': 0.1}            -0.13            0.03\n",
      "1  {'max_features': 0.2}            -0.13            0.03\n",
      "2  {'max_features': 0.3}            -0.13            0.03\n",
      "3  {'max_features': 0.4}            -0.13            0.03\n",
      "4  {'max_features': 0.5}            -0.13            0.03\n",
      "5  {'max_features': 0.6}            -0.14            0.03\n",
      "6  {'max_features': 0.7}            -0.14            0.03\n",
      "7  {'max_features': 0.8}            -0.14            0.03\n",
      "8  {'max_features': 0.9}            -0.14            0.03\n"
     ]
    }
   ],
   "source": [
    "# Define scoring metrics\n",
    "scoring = {'RMSLE': rmsle_scorer}\n",
    "\n",
    "# Create a dict of possible max_features values\n",
    "cv_vals = {'max_features': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]}\n",
    "\n",
    "# Define the model\n",
    "# n_estimators=200 based on the previous parameter tuning step\n",
    "# random_state = 33: allows reproducibility, giving the same reuslts each time the notebook is ran\n",
    "# n_jobs =-1 to use all the cores of the computer\n",
    "rf = RandomForestRegressor(n_estimators=200, random_state=33, n_jobs=-1)\n",
    "\n",
    "# cv=5 for 5 k-folds\n",
    "# scoring gets the defined scoring metrics\n",
    "# refit=Flase allows calculaitng more than one metric\n",
    "gridsearch_cv1 = GridSearchCV(estimator=rf, param_grid=cv_vals, cv=5, scoring=scoring, refit= False)\n",
    "gridsearch_cv1.fit(X_train, Y_train)\n",
    "\n",
    "results_cv1=pd.DataFrame(gridsearch_cv1.cv_results_)[[\"params\", \"mean_test_RMSLE\", \"std_test_RMSLE\"]]\n",
    "print(round(results_cv1, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the results obtained with CrossValidation, we will tune the max_features parameter to 0.4. This is because 0.4 is one of the values that show a better trade-off between RMSLE score and standard deviation, but takes a considerable proportion of features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.3. Re-testing n_estimators value\n",
    "As the value of `max_features` changed, after tuning, from the one that was originally considered when tuning `n_estimators`, we will test again if `n_estimators`=`200` is a good value for a random forest model with `max_features`=`0.4`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\cande\\anaconda3\\envs\\scientific\\lib\\site-packages\\sklearn\\ensemble\\_forest.py:578: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable OOB estimates.\n",
      "  warn(\n"
     ]
    }
   ],
   "source": [
    "# Create an empty list for OBB error and i values iteration\n",
    "error_rate = []\n",
    "i_list = []\n",
    "\n",
    "# Define a loop to iterate over the different values of n_estimators\n",
    "for i in range(10, 300, 10):\n",
    "    # We use the same configuration as in the tuning, but now max_features= 0.4\n",
    "    # n_jobs =-1 to use all the cores of the computer\n",
    "    rf = RandomForestRegressor(warm_start=True, n_estimators=i, max_features=0.4, random_state=33, oob_score=True, n_jobs=-1)\n",
    "    rf.fit(X_train, Y_train)\n",
    "\n",
    "    # oob_score attribute returns the accuracy, but we are interested on the error itself (then: 1-oob_score)\n",
    "    oob_error = 1 - rf.oob_score_\n",
    "\n",
    "    # Append OOB error rate and i values to the lists\n",
    "    error_rate.append(oob_error)\n",
    "    i_list.append(i)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot n_esitmators vs OOB error\n",
    "plt.plot(i_list, error_rate, label=\"OOB error rate\")\n",
    "plt.ylim(0.1, 1.1*max(error_rate)) \n",
    "plt.xlabel(\"n_estimators\")\n",
    "plt.ylabel(\"OOB error rate\")\n",
    "plt.title(\"OOB error rate vs n_estimators\")\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that, even after changing the max_features parameter value, n_estimators=200 shows to continue being a good stabilization value for OOB error."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.4. Applying the final Random Forest model\n",
    "Now we will train the final model with the tuned hyperparameters and compute the OOB error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestRegressor(max_features=0.4, n_estimators=200, oob_score=True,\n",
       "                      random_state=33)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestRegressor</label><div class=\"sk-toggleable__content\"><pre>RandomForestRegressor(max_features=0.4, n_estimators=200, oob_score=True,\n",
       "                      random_state=33)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestRegressor(max_features=0.4, n_estimators=200, oob_score=True,\n",
       "                      random_state=33)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf1 = RandomForestRegressor(n_estimators = 200, max_features= 0.4, oob_score=True, random_state=33)\n",
    "# Step 2: build the random forest on the train set by indicating the input data and the target variable \n",
    "rf1.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "OOB error: 0.12\n"
     ]
    }
   ],
   "source": [
    "# OOB error\n",
    "print(f'OOB error: {round(1-rf1.oob_score_, 2)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can observe that the random forest model, trained with the tuned parameters, gives an OOB error equal to 0.12. This is a good approximation on how the model would perform with unseen data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Measure the importance of variables\n",
    "Once the final model is trained, we will analyze the importance of the different variables. For this we will use the Mean Decrease in Accuracy metric (MDA), which calculates the difference between the error of the forest on Tm and the error of the forestpermuted sample e T jm. We decide to use this metric instead of Gini Impurity, other popular metric, because MDA has been demonstrated to be more robust when dealing with variables of different scales and when there are correlations between predictive variables (Nicodemus, 2011 & Han et al. 2016). These two characteristics are present in our dataset as it was analyzed in the EDA.\n",
    "\n",
    "The n_repeats parameter stablishes the number of times that a feature is randomly shuffled and permuted. In this case, it is set to 10, given that the default value is 5, and multiplying that by 2 seems to be a good trade off for obtaining a better model without increasing too much the computational cost."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = permutation_importance(rf1, X_train, Y_train, n_repeats=10, random_state=33, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will plot now the the results of the permutation importance, showing only the 15 more important features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot of the MDA for the 15 features with the highest average MDA\n",
    "sorted_idx = result.importances_mean.argsort()\n",
    "invert_sorted_idx = sorted_idx[::-1][:15]  # keep only the 15 features with the highest average MDA\n",
    "fig, ax = plt.subplots(figsize=(12, 6))\n",
    "\n",
    "# Use patch_artist=True to modify the box filling color\n",
    "bp = ax.boxplot(result.importances[invert_sorted_idx].T, vert=False, labels=X_train.columns[invert_sorted_idx], patch_artist=True)\n",
    "\n",
    "# Set the box filling color for all boxes\n",
    "for box in bp['boxes']:\n",
    "    box.set_facecolor('lightsalmon')\n",
    "\n",
    "# Set the median line color for all medians\n",
    "for median in bp['medians']:\n",
    "    median.set_color('orangered')\n",
    "\n",
    "# Add grid lines\n",
    "ax.grid(axis='x', linestyle='--', alpha=0.7)\n",
    "\n",
    "# Increase the size of the x-axis interval\n",
    "ax.set_xlim(0, 0.23)\n",
    "\n",
    "ax.set_title(\"MDA computed on the test set for the 15 features with the highest average MDA\")\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The variable importance analysis shows Overall Quality and Building Area as the two main variables, with a big distance from the following one. This is congruent with waht we have seen in the linear regression model analysis, where the same variables showed the highest correlation with the target variable 'Sold Price'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Variable selection\n",
    "In this section we will select the variables that will be used for training the final model. For this, we use Boruta feature selection strategy. The Boruta algorithm iteratively removes the features of less relevance to the prediction, based on a statistical test [(Kursa & Rudnicki, 2010)](https://www.jstatsoft.org/article/view/v036i11). The [Boruta python package](https://github.com/scikit-learn-contrib/boruta_py) provides an implementation of this algorithm.\n",
    "\n",
    "We pick Boruta because of three reasons: **One**, it overcomes the *minimal-optimal* problem. Minima-optimal problem occurs when a feature selection algorithm selects a minimal set of features that provide best possible prediction results [(Nilsson, Pe˜na, Bj¨orkegren, and Tegn´er 2007)](https://www.researchgate.net/publication/220320569_Consistent_Feature_Selection_for_Pattern_Recognition_in_Polynomial_Time). But Boruta allows for a selection of *all relevant* features [(Kursa & Rudnicki, 2010)](https://www.jstatsoft.org/article/view/v036i11). This is useful when exploring a dataset has a large set of features. **Two**, we do not want to specify the number of variables to be selected, in case we miss out on important ones due to such a limit, this is a limitation with Recurisve Feature Elimination (RFE) because the user needs to specify this limit [(Samo, 2022)](https://blog.kxy.ai/effective-feature-selection/index.html#limits). And **finally**, Boruta requires far fewer iterations compared to RFE, which can be an advantage when working with large datasets *(Ibid)*. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.a. Setting up the Boruta algorithm\n",
    "We will first install the boruta package and import it, and make necessary datatype corrections to the boruta.py file to make the implementation compatible with the latest numpy version. This is because the boruta package has not been updated for the latest version of numpy ([issue 120](https://github.com/scikit-learn-contrib/boruta_py/issues/120), [issue 122](https://github.com/scikit-learn-contrib/boruta_py/issues/122)).\n",
    "\n",
    "We then initialize the boruta algorithm with the RandomForestRegressor model, and fit it to the training data.\n",
    "\n",
    "Instructions before using boruta:\n",
    "1. Install the package using `!pip install boruta`\n",
    "2. Make the following changes to the boruta.py file in case the following code blocks throw an error. Change all the occurances of these datatypes:\n",
    "    - `np.int` to `np.int32`, \n",
    "    - `np.float` to `np.float64` and \n",
    "    - `np.bool` to `np.bool_` in the boruta.py file. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  !pip install Boruta\n",
    "from boruta import BorutaPy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Boruta algorithm is created as a wrapper around a random forest estimator. The algorithm suggests that we can pass the `n_estimators` and `max_features` values that are suitable to the dataset [(Kursa & Rudnicki, 2010)](https://www.jstatsoft.org/article/view/v036i11) to initialize a random forest estimator (or leave model defaults, if unknown). We calibrated these parameter values in the previous section, so we will use them in constructing our random forest model to be used as an estimator for the Boruta algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>BorutaPy(estimator=RandomForestRegressor(max_features=0.4, n_estimators=200,\n",
       "                                         n_jobs=-1, oob_score=True,\n",
       "                                         random_state=RandomState(MT19937) at 0x20E9F094140),\n",
       "         n_estimators=200, random_state=RandomState(MT19937) at 0x20E9F094140)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">BorutaPy</label><div class=\"sk-toggleable__content\"><pre>BorutaPy(estimator=RandomForestRegressor(max_features=0.4, n_estimators=200,\n",
       "                                         n_jobs=-1, oob_score=True,\n",
       "                                         random_state=RandomState(MT19937) at 0x20E9F094140),\n",
       "         n_estimators=200, random_state=RandomState(MT19937) at 0x20E9F094140)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: RandomForestRegressor</label><div class=\"sk-toggleable__content\"><pre>RandomForestRegressor(max_features=0.4, n_estimators=200, n_jobs=-1,\n",
       "                      oob_score=True,\n",
       "                      random_state=RandomState(MT19937) at 0x20E9F094140)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestRegressor</label><div class=\"sk-toggleable__content\"><pre>RandomForestRegressor(max_features=0.4, n_estimators=200, n_jobs=-1,\n",
       "                      oob_score=True,\n",
       "                      random_state=RandomState(MT19937) at 0x20E9F094140)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "BorutaPy(estimator=RandomForestRegressor(max_features=0.4, n_estimators=200,\n",
       "                                         n_jobs=-1, oob_score=True,\n",
       "                                         random_state=RandomState(MT19937) at 0x20E9F094140),\n",
       "         n_estimators=200, random_state=RandomState(MT19937) at 0x20E9F094140)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We use n_jobs =-1 to use all the cores of the computer, we set random_state=33 for reproducibility\n",
    "\n",
    "rf = RandomForestRegressor(n_estimators=200,n_jobs=-1, max_features=0.4,oob_score=True,random_state=33)\n",
    "rf.fit(X_train, Y_train)\n",
    "\n",
    "# define Boruta feature selection method\n",
    "boruta = BorutaPy(rf, n_estimators=200, random_state=33)\n",
    "\n",
    "# find all relevant features\n",
    "boruta.fit(X_train.values, Y_train.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.b. Selecting the relevant features\n",
    "We will now inspect all the features identified as relevant as important by Boruta. For this we use the attributes of the Boruta object, namely, `.support_`, `.ranking_` and `.support_weak_`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True, False, False, False, False, False,  True,\n",
       "       False,  True,  True,  True,  True, False,  True, False, False,\n",
       "       False,  True, False, False, False,  True, False, False,  True,\n",
       "        True, False, False,  True, False, False, False,  True,  True,\n",
       "       False, False, False,  True, False,  True,  True, False, False,\n",
       "       False, False,  True, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# To check the selected features\n",
    "boruta.support_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  1,   1,   1, 167,  56,  36, 169,  34,   1,   9,   1,   1,   1,\n",
       "         1,  39,   1,  97,  22,  12,   1,  78,  59,   2,   1,  18, 108,\n",
       "         1,   1,  11,  54,   1,  17,  11,  81,   1,   1,  46,   5,   3,\n",
       "         1,  22,   1,   1,  85,  77, 108,   5,   1,  71, 121,   7, 130,\n",
       "       169,  88,  73,   3,  20,  58, 155,  92,  14, 159, 166,  91, 135,\n",
       "        76,  93,  49, 170, 148,  96, 158,  61,  35,  13,  27,  51,  19,\n",
       "        74,  54, 115, 122,  64, 121,  84, 158, 105,  72,  29,   8, 114,\n",
       "       164, 164, 119,  64,  89,  24,  37, 127, 154, 111, 103,  60,  15,\n",
       "        95,  44, 135, 100,  65,  88, 101, 132, 149, 175, 137, 181, 106,\n",
       "       123, 103,  31, 143,  75,  90,  98,  86,  40, 146,  25, 140, 109,\n",
       "       162,  34, 180,  16,  84, 127,  47, 179, 145, 151,  52, 130, 127,\n",
       "        41,  23,  57, 103, 141,  32,  79, 153, 156, 176,  69,  28, 124,\n",
       "        71, 165, 100, 145,  30,  69,  94, 112,  54, 118,  62, 136,  27,\n",
       "         6, 127, 179, 173, 182, 133, 139, 116, 160,  39, 143,  82, 171,\n",
       "        66, 117,  80, 173, 114, 131, 110, 162, 151, 139,  48, 147, 177,\n",
       "        42,  67, 174, 152,  43,  46,  50])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# To check the ranking of the features\n",
    "boruta.ranking_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['BldgArea', 'LotFrontage', 'LotArea', 'OverallQual', 'YearBuilt', 'YearRemodAdd', 'MasVnrArea', 'ExterQual', 'BsmtQual', 'BsmtFinSF1', 'TotalBsmtSF', 'X1FloorSF', 'X2FloorSF', 'FullBath', 'KitchenQual', 'TotRmsAbvGrd', 'GarageYrBlt', 'GarageCars', 'GarageArea', 'OpenPorchSF']\n",
      "20\n"
     ]
    }
   ],
   "source": [
    "# Collect the selected features in a list\n",
    "boruta_subset = [f for f,s in zip(X_train.columns, boruta.support_) if s]\n",
    "print(boruta_subset)\n",
    "print(len(boruta_subset))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['BsmtUnfSF']\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "# we can use .support_weak_ to check for weakly relevant features \n",
    "# it is up to us whether we want to keep them or not\n",
    "weakly_relevant = boruta.support_weak_ \n",
    "boruta_weak_subset = [f for f,s in zip(X_train.columns, boruta.support_weak_) if s]\n",
    "print(boruta_weak_subset)\n",
    "print(len(boruta_weak_subset))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Boruta has selected the following 20 variables as relevant for the prediction:\n",
    "\n",
    "`BldgArea`, `LotFrontage`, `LotArea`, `OverallQual`, `YearBuilt`, `YearRemodAdd`, `MasVnrArea`, `ExterQual`, `BsmtQual`, `BsmtFinSF1`, `TotalBsmtSF`, `X1FloorSF`, `X2FloorSF`, `FullBath`, `KitchenQual`, `GarageCars`, `GarageArea`, `OpenPorchSF`\n",
    "\n",
    "We can use these variables to train the final model. It also selected onw weakly relevant features, `TotRmsAbvGrd` which we will not use for training the final model. However, we will revisit these features *if* required, after evaluating the model performance with the selected features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['BldgArea' 'BsmtFinSF1' 'BsmtQual' 'ExterQual' 'GarageArea' 'GarageCars'\n",
      " 'KitchenQual' 'LotArea' 'LotFrontage' 'MasVnrArea' 'OverallQual'\n",
      " 'TotalBsmtSF' 'X1FloorSF' 'X2FloorSF' 'YearRemodAdd']\n",
      "15\n"
     ]
    }
   ],
   "source": [
    "# this codeblock is just to check how much the results match with the MDA - we dont use this intersection for training\n",
    "intersection = np.intersect1d(boruta_subset,X_train.columns[invert_sorted_idx])\n",
    "print(intersection)\n",
    "print(len(intersection))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We observe that all the top 15 features observed in MDA feature importance are also selected by Boruta. This is a good sign that the variables selected by Boruta are indeed relevant for the prediction. We do not want to take an intersection/subset of the variables selected by Boruta and the top 15 variables from MDA because we want to keep 'all relevant' features as discussed above. Hence we only take the Boruta selected variables for training the final model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Evaluate test performance\n",
    "\n",
    "In this section we will evaluate the performance of the final model on the test dataset. We will first use the entire feature set to train the model and then use only the boruta subset. We will use the same metrics as before, RMSLE and R2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# function to plot actual and predicted values along with x=y line\n",
    "def plot_Yt_Yp(y_test, y_pred, set_limit=None):\n",
    "    \"\"\"\n",
    "    Plots the actual vs predicted values\n",
    "    ------------------------------------\n",
    "    Parameters:\n",
    "    y_test: actual values\n",
    "    y_pred: predicted values\n",
    "\n",
    "    Returns:\n",
    "    None\n",
    "    \"\"\"\n",
    "    plt.scatter(y_test, y_pred, alpha=0.3, label= 'Property Sold')\n",
    "    plt.xlabel('Actual Sold Price')\n",
    "    plt.ylabel('Predicted Sold Price')\n",
    "    plt.title('Actual vs Predicted Sold Price')\n",
    "    # Set axis limits if set_limit parameter is provided\n",
    "    if set_limit:\n",
    "        plt.axis(set_limit)\n",
    "    #plot x=y line\n",
    "    plt.plot(np.linspace(np.min(y_test), np.max(y_test), 1000), np.linspace(np.min(y_test), np.max(y_test) , 1000), color='red', label= 'x=y line')\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 4.a. Model trained using all features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE on test set:  0.13\n",
      "R2 on test set:  0.89\n",
      "RMSE on test set:  30534.56\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rf1 = RandomForestRegressor(n_estimators = 200, n_jobs=-1, max_features= 0.4, oob_score=True, random_state=33)\n",
    "rf1.fit(X_train, Y_train)\n",
    "y_pred1 = rf1.predict(X_test)\n",
    "print('RMSLE on test set: ', round(rmsle(Y_test, y_pred1),2))\n",
    "print('R2 on test set: ', round(rf1.score(X_test, Y_test),2))\n",
    "print('RMSE on test set: ', round(rmse(Y_test, y_pred1),2))\n",
    "plot_Yt_Yp(Y_test, y_pred1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We observe from this graph that the plot between actual vs predicted prices (full feature set) reasonably follows the x=y line. A keener observation is that it performs better for properties sold in the lower price ranges than at higher price ranges. This is because, divergence from the x=y line increases as the Sold price increases. The RMSLE is 0.13, the RMSE is 30489.41 and the R2 is 0.89. This is a good performance for the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 4.b. Model trained using Boruta selected features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE on test set:  0.13\n",
      "R2 on test set:  0.89\n",
      "RMSE on test set:  29740.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Using the subset of variables selected by Boruta\n",
    "X_train_new = X_train[boruta_subset]\n",
    "X_test_new = X_test[boruta_subset]\n",
    "\n",
    "rf2 = RandomForestRegressor(n_estimators = 200, n_jobs=-1, max_features= 0.4, oob_score=True, random_state=33)\n",
    "rf2.fit(X_train_new, Y_train)\n",
    "y_pred2 = rf2.predict(X_test_new)\n",
    "print('RMSLE on test set: ', round(rmsle(Y_test, y_pred2),2))\n",
    "print('R2 on test set: ', round(rf2.score(X_test_new, Y_test),2))\n",
    "print('RMSE on test set: ', round(rmse(Y_test, y_pred2),2))\n",
    "plot_Yt_Yp(Y_test, y_pred2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We observe from this graph that the plot between actual vs predicted prices (Boruta subset) shows a similar behavior as the previous plot (model trained on the full set), except for a few outliers. Interestingly, the R2 and RMSLE are also same as the model trained with the full feature set. This is a (good) sign that the Boruta selected features are indeed relevant for the prediction. Moreover, when analyzing RMSE, we see that this metric shows even a smaller error for the Boruta selected features model than when using all the features.\n",
    "\n",
    "It is safe to say that the model trained with the Boruta selected features performs as well (and even better) than the model trained with the full feature set. This is a good result because we can now use the Boruta selected features to train the model, which will reduce the computational cost of the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Conclusions\n",
    "\n",
    "1. **Configuration of the best model**: The best model is a random forest model with `n_estimators` = 200 trees and `max_features`= 0.5. This model has an OOB error of 0.13, which is a good approximation of the error that the model would have with unseen data. The model can be trained with the subset of variables selected by Boruta algorithm, which are 20 variables. These are: `BldgArea`, `LotFrontage`, `LotArea`, `OverallQual`, `YearBuilt`, `YearRemodAdd`, `MasVnrArea`, `ExterQual`, `BsmtQual`, `BsmtFinSF1`, `TotalBsmtSF`, `X1FloorSF`, `X2FloorSF`, `FullBath`, `KitchenQual`, `TotRmsAbvGrd`, `GarageYrBlt`, `GarageCars`, `GarageArea`, `OpenPorchSF`\n",
    "\n",
    "2. **Comparison between models trained on full feature set and Boruta subset**: We observe that the values of our chosen evaluation metrics are the same for both the full feature set and the model trained with Boruta selected subset **(R2=0.89, RMSLE=0.13)**. However, when checking RMSE, we see that the model eprforms even better for the Boruta subset model **(RMSE=29936 in Boruta subset vs RMSE=30489 in original model)**. We can conclude that the model trained with the subset of variables is really similar, and even a bit better than the one trained with the full set of variables. This is a very good result, as it means that we can use a model with less variables, which is easier to interpret, without losing performance.\n",
    "\n",
    "3. **Comparison with linear regression from previous milestone**: In our previous milestone we explored linear regression models - with and without regularization. In case of without regularization, we obtained R2 of ~`-inf` and RMSLE couldn't be calculated. In case of ElasticNet regularization, we obtained an **R2 of 0.76** and an **RMSLE of 0.16**. When analyzing RMSE error, we can see that our prediction are, in average, almost 13000 dollars closer to the real value **(RMSE=29936 in Boruta subset RF vs RMSE=43809 in ElasticNet)**). We can conclude that our random forest model **(R2=0.89, RMSLE=0.14)** performs better than the ElasticNet model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### References\n",
    "Kristin K. Nicodemus, Letter to the Editor: On the stability and ranking of predictors from random forest variable importance measures, Briefings in Bioinformatics, Volume 12, Issue 4, July 2011, Pages 369–373, https://doi.org/10.1093/bib/bbr016\n",
    "\n",
    "Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer. URL: https://link.springer.com/book/10.1007/978-0-387-84858-7\n",
    "\n",
    "Hong Han, Xiaoling Guo and Hua Yu, \"Variable selection using Mean Decrease Accuracy and Mean Decrease Gini based on Random Forest,\" 2016 7th IEEE International Conference on Software Engineering and Service Science (ICSESS), Beijing, 2016, pp. 219-224, doi: 10.1109/ICSESS.2016.7883053.\n",
    "\n",
    "Kursa, M. B., & Rudnicki, W. R. (2010). Feature Selection with the Boruta Package. Journal of Statistical Software, 36(11), 1 - 13. doi:http://dx.doi.org/10.18637/jss.v036.i11\n",
    "\n",
    "Samo, K. (2022, January 3). Effective Feature Selection. Retrieved from https://blog.kxy.ai/effective-feature-selection/index.html#limits\n",
    "\n",
    "Nilsson, R., Pe˜na, J. M., Bj¨orkegren, J., and Tegn´er, J. (2007). Consistent feature selection for pattern recognition in polynomial time. Journal of Machine Learning Research, 8:589–612.\n",
    "\n",
    "**Acknowledgments:** Two AI Generative tools (ChatGPT and GitHub Copilot) were used as helping tools for code improvement and debugging. However, no content was directly taken from these sources.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "scientific",
   "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}