dbscan.py 文件源码

 复制代码def main():
    # Load the dataset
    X, y = datasets.make_moons(n_samples=300, noise=0.08, shuffle=False)

    # Cluster the data using DBSCAN
    clf = DBSCAN(eps=0.17, min_samples=5)
    y_pred = clf.predict(X)

    # Project the data onto the 2 primary principal components
    p = Plot()
    p.plot_in_2d(X, y_pred, title="DBSCAN")
    p.plot_in_2d(X, y, title="Actual Clustering")