8000 Larsmans sparse nmf compat for numpy 1.5 by ogrisel · Pull Request #1 · larsmans/scikit-learn · GitHub
[go: up one dir, main page]
Skip to content

Larsmans sparse nmf compat for numpy 1.5 #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 13 commits into from
Closed

Larsmans sparse nmf compat for numpy 1.5 #1

Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
178c6e2
check if y.shape[0] == X.shape[0]; this is especially troublesome for…
pprett Oct 2, 2011
8a20cb9
unified predict for sparse and dense SGD.
pprett Oct 2, 2011
1b7b031
cosmit
pprett Oct 2, 2011
3b34162
fix: use None as default value for class_weight and sample_weight for…
pprett Oct 2, 2011
93f7bd9
cosmit: pep8
pprett Oct 2, 2011
0d9c3fb
added y.shape[0] == X.shape[0] check to DiscreteNB
pprett Oct 2, 2011
b8b655f
added X.shape[0] == y.shape[0] check to ElasitcNet
pprett Oct 2, 2011
34a6bd5
Merge remote branch 'upstream/master'
pprett Oct 2, 2011
fda113f
documented changes in whats_new
pprett Oct 3, 2011
8b57f1e
Merge remote branch 'upstream/master'
pprett Oct 3, 2011
c0881f7
Fixed issue #378 on the RFE module
glouppe Oct 3, 2011
d3b31de
Merge remote-tracking branch 'larsmans/sparse-nmf' into larsmans-spar…
ogrisel Oct 4, 2011
3356b62
FIX: compat with numpy 1.5.1 and earlier
ogrisel Oct 4, 2011
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Fixed issue scikit-learn#378 on the RFE module
  • Loading branch information
@glouppe
glouppe committed Oct 3, 2011
commit c0881f746c74358bb6d03843529b513df251f68f
18 changes: 9 additions & 9 deletions sklearn/feature_selection/rfe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,10 +29,8 @@ class RFE(BaseEstimator):
----------
estimator : object
A supervised learning estimator with a `fit` method that updates a
`coef_` attribute that holds the fitted parameters. The first
dimension of the `coef_` array must be equal to the number of features
of the input dataset of the estimator. Important features must
correspond to high absolute values in the `coef_` array.
`coef_` attribute that holds the fitted parameters. Important features
must correspond to high absolute values in the `coef_` array.

For instance, this is the case for most supervised learning
algorithms such as Support Vector Classifiers and Generalized
Expand Down Expand Up @@ -121,7 +119,11 @@ def fit(self, X, y):
# Rank the remaining features
estimator = clone(self.estimator)
estimator.fit(X[:, features], y)
ranks = np.argsort(np.sum(estimator.coef_ ** 2, axis=0))

if estimator.coef_.ndim > 1:
ranks = np.argsort(np.sum(estimator.coef_ ** 2, axis=0))
else:
ranks = np.argsort(estimator.coef_ ** 2)

# Eliminate the worse features
threshold = min(step, np.sum(support_) - self.n_features_to_select)
Expand Down Expand Up @@ -191,10 +193,8 @@ class RFECV(RFE):
----------
estimator : object
A supervised learning estimator with a `fit` method that updates a
`coef_` attribute that holds the fitted parameters. The first
dimension of the `coef_` array must be equal to the number of features
of the input dataset of the estimator. Important features must
correspond to high absolute values in the `coef_` array.
`coef_` attribute that holds the fitted parameters. Important features
must correspond to high absolute values in the `coef_` array.

For instance, this is the case for most supervised learning
algorithms such as Support Vector Classifiers and Generalized
Expand Down
0