cluster笔记

1.ValueError: Negative values in data passed to pairwise_distances. Precomputed distance need to have non-negative values.

Traceback (most recent call last):
  File "cluster.py", line 50, in <module>
    labels = estimate_label(config, dist_mat)
  File "/home/smart4s/Documents/workplaces/mgn_cluster_backup/mgn_cluster/utils.py", line 69, in estimate_label
    labels = cluster.fit_predict(dist_mat)
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/cluster/dbscan_.py", line 383, in fit_predict
    self.fit(X, sample_weight=sample_weight)
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/cluster/dbscan_.py", line 351, in fit
    **self.get_params())
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/cluster/dbscan_.py", line 175, in dbscan
    return_distance=False)
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/neighbors/base.py", line 726, in radius_neighbors
    neigh_ind_list = sum(results, [])
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/metrics/pairwise.py", line 1448, in pairwise_distances_chunked
    n_jobs=n_jobs, **kwds)
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/metrics/pairwise.py", line 1559, in pairwise_distances
    check_non_negative(X, whom=whom)
  File "/home/smart4s/anaconda3/envs/pytorch/lib/python3.6/site-packages/sklearn/utils/validation.py", line 941, in check_non_negative
    raise ValueError("Negative values in data passed to %s" % whom)

A1:scikit-learn版本问题，原来为 0.21.3 换成0.18.1