hdbscan_quickstart_clustering_synthetic_blobs_with_noise_detection.py

1import hdbscan
2from sklearn.datasets import make_blobs
3
4# Generate synthetic data
5data, _ = make_blobs(n_samples=200, centers=3, n_features=2, random_state=42)
6
7# Initialize and fit the HDBSCAN clusterer
8clusterer = hdbscan.HDBSCAN(min_cluster_size=5, gen_min_span_tree=True)
9cluster_labels = clusterer.fit_predict(data)
10
11# Print the resulting labels (-1 indicates noise)
12print(cluster_labels)