pyclustering.xmeansでKeyErrorが出る

コード

from pyclustering.cluster.xmeans import (
    xmeans, kmeans_plusplus_initializer
)
from pyclustering.utils import draw_clusters


initial_centers = kmeans_plusplus_initializer(df, 2).initialize()

instances = xmeans(df, initial_centers, ccore=True)
instances.process()

エラー

KeyError: 0

During handling of the above exception, another exception occurred:

KeyError  Traceback (most recent call last)
<ipython-input-117-8ff3c8958745> in <module>
      6 
      7 instances = xmeans(X_trans_facility, initial_centers, ccore=True)
----> 8 instances.process()
      9 
     10 clusters = instances.get_clusters()

解決

pd.DataFrameではなくlistやnp.ndarrayを使う。

X = df.values

initial_centers = kmeans_plusplus_initializer(X, 2).initialize()

instances = xmeans(X, initial_centers, ccore=True)
instances.process()

参考

バージョン

Python==3.7.4
pyclustering==0.9.3.1
pandas==1.0.2