adding course 6 & 7 pdf files

exos/tp2_decision_tree/main.py

@@ -3,37 +3,49 @@ from sklearn.model_selection import train_test_split, cross_val_score
 from sklearn.tree import DecisionTreeClassifier, plot_tree
 import matplotlib.pyplot as plt
 
-# Charger les données Iris
-data = pd.read_csv('resources/datas/iris.csv')
+# Specify the column names
+column_names = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'species']
 
-# Séparer les caractéristiques et les étiquettes
+# Load the Iris data without header and specify column names
+data = pd.read_csv('resources/datas/iris.csv', header=None, names=column_names)
+
+# Separate features and labels
 X = data.drop(columns=['species'])
 y = data['species']
 
-# Diviser les données en ensembles d'entraînement et de test
+# Split the data into training and testing sets
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
 
-# Initialiser les paramètres à tester
+# Initialize parameters to test
 parameters = {
     'min_samples_leaf': [1, 2, 4, 6, 8, 10],
     'max_depth': [None, 5, 10, 15, 20]
 }
 
 if __name__ == "__main__":
-    # Effectuer la validation croisée et mesurer le taux de classifications correctes
+    best_clf = None
+    best_score = 0
+
+    # Perform cross-validation and measure classification accuracy
     for min_samples_leaf in parameters['min_samples_leaf']:
         for max_depth in parameters['max_depth']:
             clf = DecisionTreeClassifier(min_samples_leaf=min_samples_leaf, max_depth=max_depth, random_state=42)
             clf.fit(X_train, y_train)
 
-            # Validation croisée
+            # Cross-validation
             train_scores = cross_val_score(clf, X_train, y_train, cv=5)
             test_scores = cross_val_score(clf, X_test, y_test, cv=5)
 
             print(f"min_samples_leaf: {min_samples_leaf}, max_depth: {max_depth}")
             print(f"Train accuracy: {train_scores.mean():.2f}, Test accuracy: {test_scores.mean():.2f}\n")
 
-            # Visualiser l'arbre de décision
+            # Save the best classifier
+            if test_scores.mean() > best_score:
+                best_score = test_scores.mean()
+                best_clf = clf
+
+    # Visualize the best decision tree
+    if best_clf:
         plt.figure(figsize=(20,10))
-            plot_tree(clf, filled=True, feature_names=X.columns, class_names=y.unique())
+        plot_tree(best_clf, filled=True, feature_names=X.columns, class_names=y.unique())
         plt.show()
\ No newline at end of file