update examples

examples/engine.py → examples/binary_engine.py

@@ -1,5 +1,4 @@
 import lightgbm as lgb
-from scipy.special import expit
 from sklearn.datasets import load_breast_cancer
 from sklearn.metrics import accuracy_score, log_loss, roc_auc_score
 from sklearn.model_selection import train_test_split
@@ -33,14 +32,15 @@
 }
 
 # Train standard LightGBM model
-bst_standard = lgb.train(
+lgb_standard = lgb.train(
     params_standard, train_data, num_boost_round=100, valid_sets=[test_data]
 )
 
 # Parameters for Imbalanced LightGBM model
 params_imbalanced = {
-    "objective": "weighted",  # focal
-    "metric": "binary_logloss",  # auc
+    "objective": "binary_focal",  # binary_weighted
+    "gamma": 2.0,  # alpha with binary_weighted
+    "metric": "auc",
     "learning_rate": 0.05,
     "num_leaves": 31,
     "feature_fraction": 0.9,
@@ -50,25 +50,24 @@
     "early_stopping_rounds": 10,
 }
 
-bst_focal = imlgb.train(
+# Train imbalanced LightGBM model
+imlgb_focal = imlgb.train(
     params_imbalanced, train_data, num_boost_round=100, valid_sets=[test_data]
 )
 
 # Predict using standard LightGBM model
-y_pred_standard = bst_standard.predict(X_test)
+y_pred_standard = lgb_standard.predict(X_test)
 y_pred_standard_binary = (y_pred_standard > 0.5).astype(int)
 
 # Predict using Imbalanced LightGBM model
-y_pred_focal_raw = bst_focal.predict(X_test)
-y_pred_focal = expit(y_pred_focal_raw)
+y_pred_focal = imlgb_focal.predict(X_test)
 y_pred_focal_binary = (y_pred_focal > 0.5).astype(int)
 
 # Evaluate models
 accuracy_standard = accuracy_score(y_test, y_pred_standard_binary)
 logloss_standard = log_loss(y_test, y_pred_standard)
 rocauc_standard = roc_auc_score(y_test, y_pred_standard)
 
-
 accuracy_focal = accuracy_score(y_test, y_pred_focal_binary)
 logloss_focal = log_loss(y_test, y_pred_focal)
 rocauc_focal = roc_auc_score(y_test, y_pred_focal)

examples/sklearn.py → examples/binary_sklearn.py

@@ -15,7 +15,10 @@
 
 # Initialize the ImbalancedLGBMClassifier using binary focal loss
 clf = imlgb.ImbalancedLGBMClassifier(
-    objective="binary_focal",
+    objective="binary_focal",  # binary_weighted
+    gamma=2.0,  # alpha with binary_weighted
+    learning_rate=0.05,
+    num_leaves=31,
 )
 
 # Train the classifier on the training data

examples/multiclass_engine.py

@@ -0,0 +1,81 @@
+import lightgbm as lgb
+import numpy as np
+from sklearn.datasets import make_classification
+from sklearn.metrics import classification_report
+from sklearn.model_selection import train_test_split
+
+import imlightgbm as imlgb
+
+# Generate dataset
+X, y = make_classification(
+    n_samples=5000,
+    n_features=10,
+    n_classes=3,
+    n_informative=5,
+    weights=[0.05, 0.15, 0.8],
+    flip_y=0,
+    random_state=42,
+)
+
+# Split the data into training and testing sets
+X_train, X_test, y_train, y_test = train_test_split(
+    X, y, test_size=0.2, random_state=42
+)
+
+# Create LightGBM datasets
+train_data = lgb.Dataset(X_train, label=y_train)
+test_data = lgb.Dataset(X_test, label=y_test)
+
+# Parameters for standard LightGBM model
+params = {
+    "objective": "multiclass",
+    "num_class": 3,
+    "metric": "multi_logloss",
+    "learning_rate": 0.05,
+    "num_leaves": 31,
+    "feature_fraction": 0.9,
+    "bagging_fraction": 0.8,
+    "bagging_freq": 5,
+    "seed": 42,
+    "early_stopping_rounds": 10,
+}
+
+# Train standard LightGBM model
+lgb_standard = lgb.train(
+    params, train_data, num_boost_round=100, valid_sets=[test_data]
+)
+
+# Predict using standard LightGBM model
+y_pred_standard = lgb_standard.predict(X_test)
+y_pred_standard_label = np.argmax(y_pred_standard, axis=1)
+
+# Parameters for Imbalanced LightGBM model
+params = {
+    "objective": "multiclass_focal",  # multiclass_weighted
+    "num_class": 3,
+    "gamma": 2.0,  # alpha with binary_weighted
+    "metric": "multi_logloss",
+    "learning_rate": 0.05,
+    "num_leaves": 31,
+    "feature_fraction": 0.9,
+    "bagging_fraction": 0.8,
+    "bagging_freq": 5,
+    "seed": 42,
+    "early_stopping_rounds": 10,
+}
+
+# Train Imbalanced LightGBM model
+imlgb_focal = imlgb.train(
+    params, train_data, num_boost_round=100, valid_sets=[test_data]
+)
+
+# Predict using Imbalanced LightGBM model
+y_pred_focal = imlgb_focal.predict(X_test)
+y_pred_focal_label = np.argmax(y_pred_focal, axis=1)
+
+
+# Evaluate models
+print("\nClassification Report for Standard:")
+print(classification_report(y_test, y_pred_standard_label))
+print("\nClassification Report for Imbalanced:")
+print(classification_report(y_test, y_pred_focal_label))