6744be298a
Merge pull request #3646 from s8wu/multinomial_newtoncg
...
[MRG+2] Multinomial newtoncg
2014-10-16 09:04:18 +02:00
862d82111e
Modified hessp function in _multinomial_loss_grad_hess to compute r_yhat
...
in place.
Simplied conditional statements that branches based on solver +
multi_class input combinations.
Extended tests on multinomial LogisticRegression to include newton-cg
solver case.
Updated doc string of LogisticRegression to include newton-cg solver in
multinomial case.
2014-09-20 15:11:31 -04:00
74857521ac
FIX: Make sure LogRegCV with solver=liblinear works with sparse matrices
2014-09-18 19:00:19 +02:00
af4df3345f
Adding test cases for multinomial LogisticRegression using newton-cg
...
solver.
2014-09-13 22:02:25 -04:00
891839cd8d
DOC: Explain prediction when decision_function is zero
...
LogisticRegression and liblinear's predictions differ when the
decision function is zero. Explain why and what to do in that case.
Fixes #3600 (by documenting the won't fix status).
2014-09-05 13:46:44 +02:00
a7ccbb8096
MAINT: Made the following changes
...
1. Use a single loop for both ovr and multinomial conditions
2. Better warning messages for penalty and loss conditions
3. Convergence warnings print only if verbose > 0
4. Used check_X_y instead of check_X
2014-08-30 12:26:40 +02:00
a92cd805af
FIX: PEP8 Errors and unused imports
2014-08-21 17:37:38 +02:00
d80ef50b11
DOC: Improved documentation and error messages
2014-08-21 17:37:38 +02:00
89226b69b4
COSMIT: Made the following changes
...
1. Removed true from 'true multinomial loss'
2. Replaced OvA with OvR
3. Update whatsnew.rst
2014-08-21 17:37:38 +02:00
b2e0e6e982
TST: Tests for multinomial logistic regression
2014-08-21 17:37:38 +02:00
4cd177758f
ENH: Merge MultinomialLR into LR with multi_class='multinomial'
2014-08-21 17:37:37 +02:00
d5ceb89ed5
ENH: Add support for class weights
2014-08-21 17:37:37 +02:00
13ca2b2859
ENH multinomial logistic regression using L-BFGS
2014-08-21 17:37:37 +02:00
ad5773cfca
TST: Added test to check ConvergenceWarning
2014-08-01 12:15:57 +02:00
11d396542c
FIX numerically unstable test_logistic.py
2014-07-25 00:21:42 +02:00
2374a09246
FIX: Increase testing accuracy
2014-07-23 16:04:30 +02:00
4de4849e8a
FIX: Fixes for the cross_validation failure
2014-07-23 13:24:41 +02:00
2ea804e61f
Update whats new!
2014-07-22 18:29:51 +02:00
67585f6e6f
MAINT: Improve documentation and coverage
2014-07-22 17:28:16 +02:00
c8540df986
FIX: Add DataConversionWarning
2014-07-22 16:57:41 +02:00
86dab1ea86
ENH: Added warnings for convergence, added support for l1 penalty if solver is liblinear
2014-07-22 16:57:41 +02:00
10a411fe66
ENH: Logistic Regression now supports newton-cg and lbfgs
2014-07-22 16:57:41 +02:00
7b3bfefaa0
Made the following changes
...
1. Fixed random state
2. Object dtype support
3. Remove outdated "don't test multiclass" tests
2014-07-22 16:57:41 +02:00
60f454f8ec
FIX: PEP8 and other cosmits
2014-07-22 16:57:40 +02:00
191119b74a
TST: Tests to verify OvA behavior
2014-07-22 16:57:40 +02:00
f9871cd824
ENH: Added one-vs-all fit in case of multi-class data
2014-07-22 16:57:40 +02:00
51a1831aaa
TST: Improved tests
2014-07-22 16:57:40 +02:00
7c935a3235
FIX: Doctests
2014-07-22 16:57:40 +02:00
a4d9f77805
More docs and tests for LogisticRegressionCV
2014-07-22 16:57:40 +02:00
b8c6b4c5c2
TST: Add tests to explicitly check hessian, loss and gradient for fit_intercept
2014-07-22 16:57:40 +02:00
ddb3395bff
ENH: LogisticRegressionCV can now handle sparse matrices
2014-07-22 16:57:40 +02:00
137d40bad6
Do away with intercept helper functions
2014-07-22 16:57:40 +02:00
12adea826b
COSMIT
2014-07-22 16:57:40 +02:00
8cfa8ad6cc
BUG when fit_intercept=False
2014-07-22 16:57:40 +02:00
48a862886e
Add test
2014-07-22 16:57:40 +02:00
668e0d1555
Make tests deterministic
2014-07-22 16:57:40 +02:00
4cb2d9348f
Refactor and some bug fixing
2014-07-22 16:57:39 +02:00
91ed4bb376
Some fixes for LogisticCV object
2014-07-22 16:57:39 +02:00
bb74497c67
Take into account @agramfort's comments.
...
Only cosmetic changes.
2014-07-22 16:57:39 +02:00
6e4f02c427
Implementation of logistic_regression_path.
...
This is mostly the code from Gael's logistic_cv branch. My work is only some
minor cleanup some added tests and benchmarking.
2014-07-22 16:57:39 +02:00
2afe4d5a63
TST/FIX always use "almost equal" for floats.
2013-03-01 12:14:50 +01:00
b83f1120e1
TST/FIX set random_state in logistic regression tests
2013-03-01 11:59:03 +01:00
6bf4c90c5d
ENH sparsify method for L1-reg linear models
...
Currently implemented on LinearSVC, LR and SGD.
2013-02-23 12:11:56 +01:00
754c2d9533
COSMIT pep8
2012-12-22 16:48:17 +01:00
c97e281401
FIX #1354 : machine precision assertion failure in test_liblinear_random_state
2012-11-10 20:54:07 +01:00
3c854a2fef
COSMIT pep8
2012-10-31 15:36:03 +00:00
f65efcf8eb
More use sklearn.utils.testing.
...
My model is training :)
2012-10-25 21:21:22 +09:00
e64cea0342
ENH add random seed to logistic regression
2012-10-18 13:48:10 +02:00
fa93e209ff
FIX: wrong probabilities for OvR LogisticRegression
2012-09-28 19:13:46 +02:00
a164d57e74
TST writable coef_ and intercept_ on LogisticRegression
...
Solves issue #470 .
2012-09-20 23:04:52 +02:00
Alexandre Gramfort