MRG: Fix bleeding-edge library bugs #5811
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
larsoner
commented
Dec 20, 2018
mne/inverse_sparse/mxne_optim.py
Outdated
| clf = MultiTaskLasso(alpha=alpha / len(M), tol=tol / sum_squared(M), | ||
| normalize=False, fit_intercept=False, max_iter=maxit, | ||
| warm_start=True) | ||
| clf.coef_ = init | ||
| if init is not None: | ||
| clf.coef_ = init.T |
There was a problem hiding this comment.
FYI these +2 / -4 are the ones that fixed the mxne_optim bug
larsoner
commented
Dec 20, 2018
| nopos_chanlocs[ind][dt[fld][0]] = vals[fld] | ||
| # In theory this should work and be simpler, but there is an obscure | ||
| # SciPy writing bug that pops up sometimes: | ||
| # nopos_chanlocs = np.array(chanlocs[['labels', 'Z']]) |
There was a problem hiding this comment.
And these are the lines that fixed the obscure savemat bug
|
Turns out that the "fix" for sklearn |
agramfort
reviewed
Dec 21, 2018
|
|
||
| import numpy as np | ||
| from numpy.testing import (assert_array_almost_equal, assert_array_equal, | ||
| assert_allclose, assert_equal) | ||
| import pytest | ||
| import matplotlib.pyplot as plt |
There was a problem hiding this comment.
It was moved up from below in test_find_layout
There was a problem hiding this comment.
(Eventually there is lout.plot and plt.close)
|
@larsoner we had this conversation in the past that matplotlib should not be a hard dep of mne: 14ab6b6#diff-d859f370c264e1ee2bcb22eefd2c5c2d does this PR change this? |
|
No, just a dependency to run tests as usual |
|
LGTM ok for you @massich ? |
massich
reviewed
Dec 21, 2018
|
|
||
|
|
||
| @pytest.fixture(scope='package') | ||
| def matplotlib_config(): |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Since SciPy 1.2.0 just came out and NumPy 1.16.0 is in RC status, I updated all my libs locally and hunted down errors. This PR thus (in three separate commits):
sklearnvar (we shouldn't ever setclf.coefs_ = None, sklearn can't handle this anymore). I also fixed PEP8 errors while in the file.scipy.io.savematbug when subselecting fields from a structured array intest_eeglab.py. I also fixed PEP8 errors while in the file.matplotlibtests. I noticed mine had been horribly slow for some time. Turns out it's because I addedinteractive=Trueto mymatplotlibrc. We now use a package-widepytestfixture to setmatplotlibparams. This allows us to stop nesting ourpltimports in tests, and no longer needmatploltlib.useat the top of every test file that has viz tests. This makes our code more DRY and lowers maintenance and new-contributor burden! (I am really starting to appreciate all thatpytestbrings to the table.)