Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.
Sign up[MRG+1] Ensure coef_ is an ndarray when fitting LassoLars #8160
Conversation
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
jnothman
Jan 6, 2017
Member
Great, it looks like you've got most of the way to a regression test at #1615. Please add it?
|
Great, it looks like you've got most of the way to a regression test at #1615. Please add it? |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
Tests added. |
jnothman
changed the title from
[MRG] Ensure coef_ is an ndarray when fitting LassoLars
to
[MRG+1] Ensure coef_ is an ndarray when fitting LassoLars
Jan 7, 2017
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
(I've confirmed the test fails in master) |
sklearn/linear_model/least_angle.py
| @@ -689,6 +689,8 @@ def fit(self, X, y, Xy=None): | ||
| a[0] for a in (self.alphas_, self.active_, self.coef_path_, | ||
| self.coef_)] | ||
| self.n_iter_ = self.n_iter_[0] | ||
| else: | ||
| self.coef_ = np.array(self.coef_) |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
agramfort
Jan 7, 2017
Member
it would be more efficient to preallocate self.coef_ to an empty array rather than first having a list than copying the data to an array
agramfort
Jan 7, 2017
Member
it would be more efficient to preallocate self.coef_ to an empty array rather than first having a list than copying the data to an array
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
perimosocordiae
Jan 10, 2017
Contributor
True. Would you like me to add that optimization to this PR?
perimosocordiae
Jan 10, 2017
Contributor
True. Would you like me to add that optimization to this PR?
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
jnothman
approved these changes
Jan 17, 2017
a changelog entry would be great, otherwise let's merge.
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
LGTM see why CIs complain but diff looks clean |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
tguillemot
Jan 17, 2017
Contributor
@agramfort Error on circleci are not related. We have the same problem on several PR (ex : #8135).
|
@agramfort Error on circleci are not related. We have the same problem on several PR (ex : #8135). |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
@perimosocordiae Ci problems seems solved now, can you rebase ? |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
Nah. Merging. |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
@jnothman Thanks :) |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
@perimosocordiae Thanks for your work |
added a commit
to sergeyf/scikit-learn
that referenced
this pull request
Feb 28, 2017
added a commit
to Sundrique/scikit-learn
that referenced
this pull request
Jun 14, 2017
added a commit
to NelleV/scikit-learn
that referenced
this pull request
Aug 11, 2017
added a commit
to paulha/scikit-learn
that referenced
this pull request
Aug 19, 2017
added a commit
to maskani-moh/scikit-learn
that referenced
this pull request
Nov 15, 2017
perimosocordiae commentedJan 5, 2017
This PR fixes #1615, an issue related to fitting a
LassoLarsmodel with the specific combination offit_path=True,fit_intercept=False, and >1 targets.The bug was masked in the
fit_intercept=Truecase because of this line,in which the coefficient list is promoted to an ndarray by division with another ndarray.