Warnings with Scikit-Learn >= 0.20 #44
Comments
jbschiratti
added a commit
to jbschiratti/mne-features
that referenced
this issue
Sep 20, 2018
`_compute_pow_freq_bands_feat_names` in univariate.py + changes in feature_extraction.py. * Minor changes in tests and examples to get rid of several warnings (issue mne-tools#44).
|
Addressed in #42. |
jbschiratti
changed the title
agramfort
pushed a commit
that referenced
this issue
Sep 28, 2018
* Attempt at improving `get_feature_names` (so that `extract_features` returns a dataframe with a meaningful multiindex). * * Specific feature names for `compute_pow_freq_bands` : added `_compute_pow_freq_bands_feat_names` in univariate.py + changes in feature_extraction.py. * Minor changes in tests and examples to get rid of several warnings (issue #44). * Fixes for Travis. * Added fit method to FeatureFunctionExtractor (does not extract features ; only used to get feature names when this is possible). * Added test for feature names with `compute_pow_freq_bands`. * Allow `freq_bands` parameter (in `compute_pow_freq_bands` and `compute_energy_freq_bands`) to be a dict with band names as keys. Added feature names for `compute_enregy_freq_bands` + updated tests.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
With Scikit-Learn >= 0.20 the examples give rise to the following warning:
LogisticRegression:FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning. FutureWarning)The text was updated successfully, but these errors were encountered: