k_means_kdd.py 文件源码

def fit(data, use_scikit_learn=False):
    logger.info("Starting to cluster")
    # Cluster
    n_clusters = 8
    oversampling_factor = 2
    if use_scikit_learn:
        km = sk.KMeans(n_clusters=n_clusters, random_state=0)
    else:
        km = KMeans(n_clusters=n_clusters,
                    oversampling_factor=oversampling_factor,
                    random_state=0)
    t0 = tic()
    logger.info("Starting n_clusters=%2d, oversampling_factor=%2d",
                n_clusters, oversampling_factor)
    km.fit(data)
    t1 = tic()
    logger.info("Finished in %.2f", t1 - t0)