Commit Graph

209 Commits

Author SHA1 Message Date
Rémy Léone 9b7176dd9d [DOC] Fix broken links 2016-03-23 12:45:34 -07:00
Florian Wilhelm 681fd2f22b DOC: Fix order in ARD references 2016-03-16 16:42:45 +01:00
Florian Wilhelm aa31fa6335 Added additional references for ARDRegression 2016-03-15 21:08:42 +01:00
Tom Dupré la Tour 06bf797c0d DOC update docstring in LogisticRegression 2016-03-15 11:11:59 +01:00
MechCoder 7e79adca9d Huber regressor
Add gradient calculation in _huber_loss_and_gradient

Add tests to check the correctness of the loss and gradient

Fix for old scipy

Add parameter sigma for robust linear regression

Add gradient formula to robust _huber_loss_and_gradient

Add fit_intercept option and fix tests

Add docs to HuberRegressor and the helper functions

Add example demonstrating ridge_regression vs huber_regression

Add sample_weight implementation

Add scaling invariant huber test

Remove exp and add bounds to fmin_l_bfgs_b

Add sparse data support

Add more tests and refactoring of code

Add narrative docs

review huber regressor

Minor additions to docs and tests

Minor fixes that deals with dealing with NaN values in targets
and old verions of SciPy and NumPy

Add HuberRegressor to robust estimator

Refactored computation of gradient and make docs render properly

Temp

Remove float64 dtype conversion

trivial optimizations and add a note about R

Remove sample_weights special_casing

address @amueller comments
2016-02-25 16:31:08 -05:00
saurabh.bansod 1487047df9 update url of ARD reference 2016-01-19 14:33:33 +05:30
TomDLT 9f136ff293 ENH add multinomial SAG solver for LogisticRegression 2015-12-04 18:36:37 +01:00
Alexandre Gramfort 93c3033c1a nitpicks 2015-10-23 11:42:54 +02:00
John Boersma a8880f1d6e Fixed typos: whitespace after '.', `` around inline code. 2015-10-23 11:42:54 +02:00
John Boersma 07a9b8dfee Corrected MultiTaskLasso objective function and moved some related definitions there
as suggested by agramfort.
2015-10-23 11:42:54 +02:00
John Boersma 0b00e2bbbb Corrected {} 2015-10-23 11:42:54 +02:00
John Boersma 29b4931935 Correct for Frobenius norm 2015-10-23 11:42:54 +02:00
John Boersma 6c455ae910 Corrected reference to "lasso" to "ElasticNet". 2015-10-23 11:42:54 +02:00
John Boersma 61f53132ce Edits to math paragraph, added reference to MultiTaskElasticNetCV. 2015-10-23 11:42:54 +02:00
John Boersma a614a5c770 Edits to intro paragraph of Multi-task Elastic net. 2015-10-23 11:42:54 +02:00
John Boersma a03333f4a9 Moved Multi-task Lasso to immediately follow Lasso 2015-10-23 11:42:54 +02:00
John Boersma e82bca5634 Removing test edit. 2015-10-23 11:42:53 +02:00
John Boersma e8a2e2705a Test edit. 2015-10-23 11:42:53 +02:00
Raghav R V 4f4c3598e6 FIX dtypes to conform to the stricter type cast rules of numpy 1.10
FIX set copy to (copy & whiten).

FIX/DOC Use float outputs for doctest
2015-10-15 18:21:51 +02:00
Andreas Mueller 6dd6f8ffe8 minor fixes to the doc build 2015-09-11 17:32:51 -04:00
TomDLT 94eb61960a ENH add sag solver in LogisticRegression and Ridge 2015-09-10 13:28:02 -04:00
mhg 6b22d97011 positive option for Lars propagated to doc. 2015-08-18 15:24:16 +09:00
Andreas Mueller 0650d5502e DOC adding backlinks to docstrings 2015-06-03 00:24:04 -04:00
Andreas Mueller b28d3bf14b Merge pull request #4464 from vortex-ape/logregcv
Modifies LogisiticRegressionCV doc and docstring
2015-04-01 16:12:17 -05:00
Jack Martin 9e7a660ff6 added import np statements 2015-03-31 10:38:47 -04:00
Jack Martin 22963a0fc0 Added import statement for Lasso example 2015-03-30 10:31:53 -04:00
Vinayak Mehta c6d420ee5f Modified docs 2015-03-30 01:38:17 +05:30
Andreas Mueller deeaf0ad35 change default to shuffle=True in SGDClassifier and friends. 2015-03-03 11:56:23 -05:00
Olivier Grisel f0fe4afd1e Merge pull request #2949 from FlorianWilhelm/theilsen
[MRG+1] TheilSen robust linear regression
2014-11-20 15:34:01 +01:00
snuderl 61375792f9 Small typo. 2014-11-20 11:43:48 +01:00
Alexandre Gramfort 6744be298a Merge pull request #3646 from s8wu/multinomial_newtoncg
[MRG+2] Multinomial newtoncg
2014-10-16 09:04:18 +02:00
Florian Wilhelm a5d2fd915a Merge branch 'master' into theilsen
Conflicts:
	doc/modules/model_evaluation.rst
2014-10-09 17:43:21 +02:00
Manoj-Kumar-S e915fcc3ab Changed default argument of precompute in ElasticNet and Lasso
Setting precompute to "auto" was found to be slower when n_samples > n_features
since the computation of the Gram matrix is computationally expensive and
outweighs the benefit of fitting the Gram for just one alpha.
2014-10-03 18:25:12 +02:00
Florian Wilhelm cf3b476aa4 FIX: TheilSen -> TheilSenRegressor in docs 2014-10-02 09:25:48 +02:00
Florian Wilhelm a8000405de COSMIT: Small changes regarding Theil-Sen
- Typo fixed in linear_model documenation
- Usage of matplotlib.pyplot instead of matplotlib.pylab
- Removed trailing backslash in import statements
- Renamed _modweiszfeld_step to _modified_weiszfeld_step
- Renamed variable fst to first_elem in _lstq
- Made get_n_jobs private in utils/__init__.py
2014-09-24 19:58:21 +02:00
Florian Wilhelm b0ab714699 Merge branch 'master' into theilsen 2014-09-24 19:30:39 +02:00
swu b742d581b7 Modified doc/modules/linear_model.rst to incorporate the newton-cg solver
option for multi_class="multinomial".
2014-09-21 12:43:11 -04:00
akshayah3 077a1ee099 Fixed a failing test in the doc 2014-09-19 16:03:42 +05:30
Florian Wilhelm 8dbec4f76c Merge branch 'master' into theilsen 2014-09-06 10:14:12 +02:00
MechCoder 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
MechCoder 43845d7115 DOC: Made the docs for LogisticRegression clearer 2014-08-21 17:37:38 +02:00
MechCoder d80ef50b11 DOC: Improved documentation and error messages 2014-08-21 17:37:38 +02:00
MechCoder 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
MechCoder 4cd177758f ENH: Merge MultinomialLR into LR with multi_class='multinomial' 2014-08-21 17:37:37 +02:00
Lars Buitinck 13ca2b2859 ENH multinomial logistic regression using L-BFGS 2014-08-21 17:37:37 +02:00
MechCoder cf4cf60b7e ENH Stochastic Coordinate Descent for ElasticNet & Lasso 2014-07-29 10:23:33 +02:00
MechCoder 2ea804e61f Update whats new! 2014-07-22 18:29:51 +02:00
MechCoder 4d516e81f4 DOC: Minor changes 2014-07-22 16:57:41 +02:00
MechCoder 5859f8b4fe Changed copy default from True to False, updated docstring for sample_weights 2014-07-22 16:57:41 +02:00
MechCoder 10a411fe66 ENH: Logistic Regression now supports newton-cg and lbfgs 2014-07-22 16:57:41 +02:00