--- a
+++ b/Random Forest.ipynb
@@ -0,0 +1,457 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h1 align=\"center\">Machine learning-based prediction of early recurrence in glioblastoma patients: a glance towards precision medicine <br><br>[Random Forest]</h1>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2>[1] Library</h2>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# OS library\n",
+    "import os\n",
+    "import sys\n",
+    "import argparse\n",
+    "import itertools\n",
+    "import random\n",
+    "\n",
+    "# Analysis\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Sklearn\n",
+    "from boruta import BorutaPy\n",
+    "from sklearn.preprocessing import LabelEncoder\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.metrics import confusion_matrix, f1_score, recall_score, classification_report, accuracy_score, auc, roc_curve\n",
+    "from sklearn.model_selection import RandomizedSearchCV\n",
+    "\n",
+    "import scikitplot as skplt\n",
+    "from imblearn.over_sampling import RandomOverSampler, SMOTENC, SMOTE"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2>[2] Exploratory data analysis and Data Preprocessing</h2>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Load the database</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "file = os.path.join(sys.path[0], \"db.xlsx\")\n",
+    "db = pd.read_excel(file)\n",
+    "\n",
+    "print(\"N° of patients: {}\".format(len(db)))\n",
+    "print(\"N° of columns: {}\".format(db.shape[1]))\n",
+    "db.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Drop unwanted columns + create <i>'results'</i> column</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = db.drop(['Name_Surname','...'], axis = 'columns')\n",
+    "\n",
+    "print(\"Effective features to consider: {} \".format(len(df.columns)-1))\n",
+    "print(\"Creating 'result' column...\")\n",
+    "\n",
+    "# 0 = No relapse\n",
+    "df.loc[df['PFS'] > 6, 'result'] = 0\n",
+    "\n",
+    "# 1 = Early relapse (within 6 months)\n",
+    "df.loc[df['PFS'] <= 6, 'result'] = 1\n",
+    "\n",
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Check for class imbalance in the <i>'results'</i> column </h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(\"PFS Overview\")\n",
+    "print(df.result.value_counts())\n",
+    "\n",
+    "df.result.value_counts().plot(kind='pie', autopct='%1.0f%%', colors=['skyblue', 'orange'], explode=(0.05, 0.05))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Label encoding of the categorical variables </h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['sex'] =df['sex'].astype('category')\n",
+    "\n",
+    "#df['Ki67'] =df['Ki67'].astype(int)\n",
+    "df['MGMT'] =df['MGMT'].astype('category')\n",
+    "df['IDH1'] =df['IDH1'].astype('category')\n",
+    "\n",
+    "df['side'] =df['side'].astype('category')\n",
+    "df['ependima'] =df['ependima'].astype('category')\n",
+    "df['cc'] =df['cc'].astype('category')\n",
+    "df['necrotico_cistico'] =df['necrotico_cistico'].astype('category')\n",
+    "df['shift'] =df['shift'].astype('category')\n",
+    "\n",
+    "## VARIABLE TO ONE-HOT-ENCODE\n",
+    "df['localization'] =df['localization'].astype(int)\n",
+    "df['clinica_esordio'] =df['clinica_esordio'].astype(int)\n",
+    "df['immediate_p_o'] =df['immediate_p_o'].astype(int)\n",
+    "df['onco_Protocol'] =df['onco_Protocol'].astype(int)\n",
+    "\n",
+    "df['result'] =df['result'].astype(int)\n",
+    "\n",
+    "dummy_v = ['localization', 'clinica_esordio', 'onco_Protocol', 'immediate_p_o']\n",
+    "\n",
+    "df = pd.get_dummies(df, columns = dummy_v, prefix = dummy_v)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Evaluate variables' correlation with <u>'PFS'</u> columns </h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "corr = df.corr()\n",
+    "ax = sns.heatmap(\n",
+    "    corr, \n",
+    "    vmin=-1, vmax=1, center=0,\n",
+    "    cmap=sns.diverging_palette(20, 220, n=200),\n",
+    "    square=True\n",
+    ")\n",
+    "ax.set_xticklabels(\n",
+    "    ax.get_xticklabels(),\n",
+    "    rotation=60,\n",
+    "    horizontalalignment='right'\n",
+    ");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2>[3] Prediction Models</h2>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4> [-] Training and testing set splitting</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [
+    {
+     "ename": "NameError",
+     "evalue": "name 'df' is not defined",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-5-48cdcc32916c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtarget\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'result'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'result'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'PFS'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'columns'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mNameError\u001b[0m: name 'df' is not defined"
+     ]
+    }
+   ],
+   "source": [
+    "target = df['result']\n",
+    "inputs = df.drop(['result', 'PFS'], axis = 'columns')\n",
+    "x_train, x_test, y_train, y_test = train_test_split(inputs['...'],target,test_size=0.20, random_state=10)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4> [-] BORUTA Features Selection</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "x = x_train.values\n",
+    "y = y_train.values\n",
+    "y = y.ravel()\n",
+    "\n",
+    "rf_boruta = RandomForestClassifier(n_jobs=-1, class_weight={0:1, 1:3}, max_depth=3)\n",
+    "feat_selector = BorutaPy(rf_boruta, n_estimators='auto', verbose=0, random_state=42, perc='...')\n",
+    "feat_selector.fit(x, y)\n",
+    "\n",
+    "cols = inputs.columns[feat_selector.support_]\n",
+    "print(\"N° of selected features: {}\".format(len(cols)))\n",
+    "print(cols) "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4> [-] Random Grid Search Hyperparameter tuning</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# The function to measure the quality of a split\n",
+    "criterion = ['gini', 'entropy']\n",
+    "\n",
+    "# Number of trees in random forest\n",
+    "n_estimators = [int(x) for x in np.linspace(start = 20, stop = 50, num = 5)]\n",
+    "\n",
+    "# Number of features to consider at every split\n",
+    "max_features = ['auto', 'sqrt']\n",
+    "\n",
+    "# Maximum number of levels in tree\n",
+    "max_depth = [int(x) for x in np.linspace(14, 30, num = 2)]\n",
+    "max_depth.append(None)\n",
+    "\n",
+    "# Minimum number of samples required to split a node\n",
+    "min_samples_split = [ 2, 3, 4, 5, 8]\n",
+    "\n",
+    "# Minimum number of samples required at each leaf node\n",
+    "min_samples_leaf = [1, 2, 3, 4, 5, 6]\n",
+    "\n",
+    "max_leaf_nodes = [None, 2, 3, 4, 5, 6]\n",
+    "# Method of selecting samples for training each tree\n",
+    "bootstrap = [True, False]\n",
+    "\n",
+    "random_grid = {'criterion': criterion,\n",
+    "               'n_estimators': n_estimators,\n",
+    "               'max_features': max_features,\n",
+    "               'max_depth': max_depth,\n",
+    "               'min_samples_split': min_samples_split,\n",
+    "               'min_samples_leaf': min_samples_leaf,\n",
+    "               'max_leaf_nodes': max_leaf_nodes,\n",
+    "               'bootstrap':bootstrap\n",
+    "              }\n",
+    "\n",
+    "# First create the base model to tune\n",
+    "rf = RandomForestClassifier(random_state=42,\n",
+    "                            n_jobs = -1, \n",
+    "                            class_weight=class_weight)\n",
+    "\n",
+    "# Random search of parameters, using 5 fold cross validation, different combinations, and use all available cores\n",
+    "rf_random = RandomizedSearchCV(estimator = rf, \n",
+    "                               param_distributions = random_grid, \n",
+    "                               n_iter = 500, \n",
+    "                               cv = 5)\n",
+    "# Fit the random search model\n",
+    "rf_random.fit(x_train, y_train)\n",
+    "rf_random.best_params_"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] SMOTE-NC</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "smote_nc = SMOTENC(categorical_features=[3,4,10,11], k_neighbors= 3, random_state=42)\n",
+    "x_smote_train, y_smote_train = smote_nc.fit_resample(x_train, y_train)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4>[-] Random Forest Model</h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "rm_smote = RandomForestClassifier(random_state = 42,\n",
+    "                                       criterion= '...',\n",
+    "                                       n_estimators = '...',\n",
+    "                                       min_samples_split = '...',\n",
+    "                                       min_samples_leaf = '...',\n",
+    "                                       max_leaf_nodes = '...',\n",
+    "                                       max_features = '...',\n",
+    "                                       max_depth = '...',\n",
+    "                                       bootstrap = '...')\n",
+    "\n",
+    "rm_smote.fit(x_smote_train, y_smote_train)\n",
+    "print(\"Trained \\n\")\n",
+    "\n",
+    "score_rf_smote = rm_smote.score(x_test, y_test)\n",
+    "print(\"Random Forest accuracy: \", round(score_rf_smote*100,2), \"% \\n\")\n",
+    "\n",
+    "y_smote_predicted = rm_smote.predict(x_test)\n",
+    "cm_rf_smote = confusion_matrix(y_test, y_smote_predicted)\n",
+    "print(cm_rf_smote, \"\\n\")\n",
+    "\n",
+    "false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, y_smote_predicted)\n",
+    "roc_auc = auc(false_positive_rate, true_positive_rate)\n",
+    "\n",
+    "print('1. The F-1 Score of the model {} \\n '.format(round(f1_score(y_test, y_smote_predicted, average = 'macro'), 2)))\n",
+    "print('2. The Recall Score of the model {} \\n '.format(round(recall_score(y_test, y_smote_predicted, average = 'macro'), 2)))\n",
+    "print('3. Classification report \\n {} \\n'.format(classification_report(y_test, y_smote_predicted)))\n",
+    "print('3. AUC: \\n {} \\n'.format(roc_auc))\n",
+    "\n",
+    "tn, fp, fn, tp = cm_rf_smote.ravel()\n",
+    "\n",
+    "# Sensitivity, hit rate, Recall, or true positive rate\n",
+    "tpr = tp/(tp+fn)\n",
+    "print(\"Sensitivity (TPR): {}\".format(tpr))\n",
+    "\n",
+    "# Specificity or true negative rate\n",
+    "tnr = tn/(tn+fp)\n",
+    "print(\"Specificity (TNR): {}\".format(tnr))\n",
+    "\n",
+    "# Precision or positive predictive value\n",
+    "ppv = tp/(tp+fp)\n",
+    "print(\"Precision (PPV): {}\".format(ppv))\n",
+    "\n",
+    "# Negative predictive value\n",
+    "npv = tn/(tn+fn)\n",
+    "print(\"Negative Predictive Value (NPV): {}\".format(npv))\n",
+    "\n",
+    "# False positive rate\n",
+    "fpr = fp / (fp + tn)\n",
+    "print(\"False Positive Rate (FPV): {}\".format(fpr))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h4> [-] Features Importance Plot </h4>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "features = x_train.columns.values\n",
+    "\n",
+    "features[0] = 'Age'\n",
+    "features[6] = 'Tumor volume T1'\n",
+    "features[7] = 'edema volume'\n",
+    "features[8] = 'Residual tumor'\n",
+    "features[9] = 'Pre-operative KPS'\n",
+    "features[10] = 'Post-operative KPS'\n",
+    "features[11] = 'Onset neurological symptoms = 1'\n",
+    "features[12] = 'Oncological protocol = 0'\n",
+    "features[13] = 'Oncological protocol = 1'\n",
+    "features[14] = 'Oncological protocol = 2'\n",
+    "\n",
+    "indices = np.argsort(importances)\n",
+    "\n",
+    "plt.title('Random Forest Classifier Features Importance')\n",
+    "plt.barh(range(len(indices)), importances[indices], color='g', align='center')\n",
+    "plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
+    "plt.xlabel('Relative Importance')\n",
+    "\n",
+    "plt.savefig(\"RF Features importance.jpg\", dpi = 400, facecolor='w', edgecolor='w',\n",
+    "        orientation='landscape', papertype=None, format=None,\n",
+    "        transparent=False, bbox_inches='tight', pad_inches=0.3,\n",
+    "        frameon=None)\n",
+    "\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "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": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}