MRG, BUG: Fix biased cov estimation #7369
Merged
Conversation
This file contains 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
Codecov Report
@@ Coverage Diff @@
## master #7369 +/- ##
==========================================
+ Coverage 89.98% 90.07% +0.08%
==========================================
Files 453 453
Lines 81994 81204 -790
Branches 12994 12998 +4
==========================================
- Hits 73785 73145 -640
+ Misses 5382 5231 -151
- Partials 2827 2828 +1 |
|
thx heaps @larsoner ! |
AdoNunes
pushed a commit
to AdoNunes/mne-python
that referenced
this pull request
Apr 6, 2020
* BUG: Fix biased cov estimation * FIX: Minor fix * FIX: Link * FIX: Dont require sklearn * FIX: sklearn check * MAINT: Simplify check * FIX: Flake
AdoNunes
pushed a commit
to AdoNunes/mne-python
that referenced
this pull request
Apr 6, 2020
* BUG: Fix biased cov estimation * FIX: Minor fix * FIX: Link * FIX: Dont require sklearn * FIX: sklearn check * MAINT: Simplify check * FIX: Flake
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Outdated suggestions cannot be applied.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.
We have an inconsistency where MNE-C and MNE-Python with
method='empirical'would compute an unbiased estimator of the covariance, but any other mode would compute a biased estimator (the maximum likelihood estimator). This PR makes us always compute the unbiased one.We could alternatively add a
biaskwarg tocompute_covarianceandcompute_raw_covariancewith a deprecation cycle for people. But I'm not sure it's worth it because in practice the scale factor should be quite small as it isN / (N - 1)whereNis the number of (time) samples used to compute the covariance, which is typically quite large.