arthur/ArtStudies
1
0
Fork 0
mirror of https://github.com/ArthurDanjou/ArtStudies.git synced 2026-01-14 15:54:13 +01:00
Code Issues Packages Projects Releases Wiki Activity

Ajout de la séparation des données en ensembles d'apprentissage et de test, et implémentation de la recherche de grille pour les hyperparamètres du modèle Random Forest.

Browse Source
This commit is contained in:
Arthur Danjou
2025-10-13 20:00:44 +02:00
parent a0b0a9f8bd
commit ec5e23e3d4
1 changed files with 57 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

65
M2/Machine Learning/TP_3/2025_TP_3_M2_ISF.ipynb
View File

@@ -3598,7 +3598,14 @@
"id": "d9342ad6",
"metadata": {},
"outputs": [],
"source": []
"source": [
"X = data_model_preprocessed\n",
"Y = data_model[\"CM\"]\n",
"\n",
"X_train, X_test, y_train, y_test = train_test_split(\n",
" X, Y, test_size=0.2, random_state=42\n",
")"
]
},
{
"cell_type": "markdown",
@@ -3618,7 +3625,7 @@
},
{
"cell_type": "code",
"execution_count": 79,
"execution_count": null,
"id": "6d58dbc2",
"metadata": {},
"outputs": [],
@@ -3636,9 +3643,9 @@
"RMSE_scores = []\n",
"\n",
"# Hyperparamètres à tester\n",
"n_estimators_values = [] #Complétez ici par les paramètres à tester\n",
"max_depth_values = [] #Complétez ici par les paramètres à tester\n",
"min_samples_split_values = [] #Complétez ici par les paramètres à tester\n",
"n_estimators_values = []\n",
"max_depth_values = []\n",
"min_samples_split_values = []\n",
"\n",
"# Liste pour sauveagrder les meilleurs résultats\n",
"best_score = np.inf\n",
@@ -3651,17 +3658,59 @@
},
{
"cell_type": "code",
"execution_count": 80,
"execution_count": 87,
"id": "47da5172",
"metadata": {},
"outputs": [],
"source": [
"#Complétez ici avec votre code"
"#Complétez ici avec votre code\n",
"for n_estimators in n_estimators_values:\n",
" for max_depth in max_depth_values:\n",
" for min_samples_split in min_samples_split_values:\n",
" rf_regressor = RandomForestRegressor(\n",
" n_estimators=n_estimators,\n",
" max_depth=max_depth,\n",
" min_samples_split=min_samples_split,\n",
" random_state=42\n",
" )\n",
" MAE_scores = []\n",
" MSE_scores = []\n",
" RMSE_scores = []\n",
"\n",
" for train_index, val_index in kf.split(X_train):\n",
" X_train_fold, X_val_fold = X_train.iloc[train_index], X_train.iloc[val_index]\n",
" y_train_fold, y_val_fold = y_train.iloc[train_index], y_train.iloc[val_index]\n",
"\n",
" rf_regressor.fit(X_train_fold, y_train_fold)\n",
" y_pred_fold = rf_regressor.predict(X_val_fold)\n",
"\n",
" mae = metrics.mean_absolute_error(y_val_fold, y_pred_fold)\n",
" mse = metrics.mean_squared_error(y_val_fold, y_pred_fold)\n",
" rmse = metrics.root_mean_squared_error(y_val_fold, y_pred_fold)\n",
"\n",
" MAE_scores.append(mae)\n",
" MSE_scores.append(mse)\n",
" RMSE_scores.append(rmse)\n",
"\n",
" avg_mae = np.mean(MAE_scores)\n",
" avg_mse = np.mean(MSE_scores)\n",
" avg_rmse = np.mean(RMSE_scores)\n",
"\n",
" if avg_rmse < best_score:\n",
" best_score = avg_rmse\n",
" best_params = {\n",
" 'n_estimators': n_estimators,\n",
" 'max_depth': max_depth,\n",
" 'min_samples_split': min_samples_split\n",
" }\n",
" MAE_best_score = MAE_scores\n",
" MSE_best_score = MSE_scores\n",
" RMSE_best_score = RMSE_scores"
]
},
{
"cell_type": "code",
"execution_count": 81,
"execution_count": 88,
"id": "d4936c46",
"metadata": {},
"outputs": [