[Regression/Backward compatibility] ImportError: cannot import name 'MaskedArray' when building with scikit-learn release 0.23.0 · Issue #17213 · scikit-learn/scikit-learn · GitHub
You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running existing packages/programs with the latest release of 0.23.0 gives an import error on MaskedArray.
Bounding the requirements to < 0.23.0 resolves this issue, but then it reverts to the older version although would be good to be on the latest release.
Steps/Code to Reproduce
Setup
$ pip install scikit-learn>=0.20.2
or
$ pip install scikit-learn==0.23.0
ImportError while importing test module '/opt/atlassian/pipelines/agent/build/tests_package_name/module.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
tests/test_package_name.py:7: in <module>
from package_name import H2OGradientBoostingModel
package_name/module.py:5: in <module>
from skopt import gp_minimize
/usr/local/lib/python3.6/site-packages/skopt/__init__.py:54: in <module>
from .searchcv import BayesSearchCV
/usr/local/lib/python3.6/site-packages/skopt/searchcv.py:16: in <module>
from sklearn.utils.fixes import MaskedArray
E ImportError: cannot import name 'MaskedArray'
Versions
0.23.0
The text was updated successfully, but these errors were encountered:
neomatrix369
changed the title
[Error] ImportError: cannot import name 'MaskedArray' when building with scikit-learn release 0.23.0
[Regression/Backward compatibility] ImportError: cannot import name 'MaskedArray' when building with scikit-learn release 0.23.0
May 13, 2020
Uh oh!
There was an error while loading. Please reload this page.
Describe the bug
Running existing packages/programs with the latest release of 0.23.0 gives an import error on
MaskedArray
.Bounding the requirements to
< 0.23.0
resolves this issue, but then it reverts to the older version although would be good to be on the latest release.Steps/Code to Reproduce
Setup
$ pip install scikit-learn>=0.20.2 or $ pip install scikit-learn==0.23.0
Code
Expected Results
No exception or a warning with an alternative solution to the above issue. A depreciation message if the class has been removed.
Although, from https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/utils/fixes.py#L20 it appears that this is due to be removed in
0.25.0
.Actual Results
Versions
0.23.0
The text was updated successfully, but these errors were encountered: