imblearn_smote_oversampling_imbalanced_dataset_quickstart.py

1from collections import Counter
2from sklearn.datasets import make_classification
3from imblearn.over_sampling import SMOTE
4
5# Create a toy imbalanced dataset
6X, y = make_classification(n_classes=2, class_sep=2,
7                           weights=[0.1, 0.9], n_informative=3, n_redundant=1, flip_y=0,
8                           n_features=20, n_clusters_per_class=1, n_samples=1000, random_state=10)
9print(f'Original dataset shape {Counter(y)}')
10
11# Apply SMOTE to balance the dataset
12sm = SMOTE(random_state=42)
13X_res, y_res = sm.fit_resample(X, y)
14
15print(f'Resampled dataset shape {Counter(y_res)}')