Cannot import spherecluster with scikit-learn 1.0.2: sklearn.cluster.k_means_ has been renamed

[ngiatsog]

The traceback is

Traceback (most recent call last):
File "//python3.9/site-packages/IPython/core/interactiveshell.py", line 3457, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "", line 1, in
import spherecluster
File "/snap/pycharm-community/267/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_import_hook.py", line 21, in do_import
module = self._system_import(name, *args, **kwargs)
File "//python3.9/site-packages/spherecluster/init.py", line 2, in
from .spherical_kmeans import SphericalKMeans
File "/snap/pycharm-community/267/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_import_hook.py", line 21, in do_import
module = self.system_import(name, *args, **kwargs)
File "//python3.9/site-packages/spherecluster/spherical_kmeans.py", line 7, in
from sklearn.cluster.k_means import (
File "/snap/pycharm-community/267/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_import_hook.py", line 21, in do_import
module = self.system_import(name, *args, **kwargs)
ModuleNotFoundError: No module named 'sklearn.cluster.k_means'

The problem is that 'sklearn.cluster.k_means_' has been renamed to 'sklearn.cluster._kmeans' in some intermediate scikit-learn version.

[rfayat]

Hi !

I haven't tested it with the most recent version of scikit but I solved this issue for scikit 0.24 in my pull request #34 (you can also check my fork) where the change from sklearn.cluster.k_means_ to sklearn.cluster._kmeans appeared.

From what I remember it was a bit more complicated than a simple renaming issue because some functions called by sklearn.cluster.k_means_ are now class methods of KMeans.

Hope this helps !

Romain

[Darel13712]

@rfayat Hey, I tried your version, but ran into

from sklearn.cluster import _k_means_fast as _k_means

ImportError: cannot import name '_k_means_fast' from 'sklearn.cluster'

Does it mean that interfaces changed once again or I'm just using a wrong version?

[rfayat]

Hey @Darel13712 indeed I just checked and it seems to have changed again between v0.24 and V1.0.2, my fork won't work for this newest version...

Sorry for the false hope 😞

Cheers,

RF

[yuye-aws]

Also meet with the same problem. Is there any workaround to train spherical kmeans?

[DanVicenteIhanus]

Also meet with the same problem. Is there any workaround to train spherical kmeans?

One workaround is to use Faiss library, which has a spherical parameter for their K-means implementation:
https://github.com/facebookresearch/faiss/wiki/Faiss-building-blocks:-clustering,-PCA,-quantization