baohan
/
scikit-learn
mirror of https://github.com/scikit-learn/scikit-learn.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
10,497 Commits 31 Branches 142 Tags 892 MiB
Commit Graph

79 Commits

Author SHA1 Message Date
Lars Buitinck 9423e9c808 DOC various typos and other minor stuff 2012-10-11 11:26:22 +02:00
Lars Buitinck dc943722de COSMIT refactor linear classifiers 
Factored out the common code to LinearSVC, LogisticRegression, Perceptron
and SGDClassifier into a LinearClassifierMixin.
 2012-09-17 21:06:36 +02:00
Alexandre Gramfort 48687a77a4 ENH : add support for multitarget in sparse enet + simplify input checking 2012-09-02 23:16:52 +02:00
Lars Buitinck 7c8f435469 COSMIT move BaseSGD to its only place of usage 2012-08-01 15:25:36 +02:00
Andreas Mueller bab0539d37 ENH make LinearModel and LinearModelCV abstract base classes 2012-06-26 14:16:04 +02:00
Mathieu Blondel fb714df433 Fix #904. 2012-06-13 13:26:02 +09:00
Immanuel Bayer fae2f13849 move sparse_center_data to linear_model.base 2012-06-09 11:23:43 +02:00
Lars Buitinck e906d483a7 COSMIT pyflakes + pep8 linear_model/base.py 2012-05-15 15:30:52 +02:00
Olivier Hervieu d0b20f0a21 FIX inconsistent coef_.shape in LinearRegression 
Should fix issue #652.
 2012-05-15 15:30:08 +02:00
Mathieu Blondel 7b75227afb Rename "p" to "espilon". 2012-05-02 17:23:21 +09:00
flyingimmidev 5964664f85 combat for sp_linalg.lsqr 2012-03-22 23:38:10 +01:00
Alexandre Gramfort 9fe4408754 FIX : use joblib from externals 2012-03-21 21:53:43 +01:00
Andreas Mueller 2119eb1aa9 Merge pull request #685 from ibayer/master 
Make linear regression support multiple outputs
 2012-03-21 13:08:38 -07:00
Alexandre Gramfort 03443dab9d ENH: y_mean with consistent shape in _center_data 2012-03-19 12:06:19 +01:00
Alexandre Gramfort eefa816595 FIX : fix test for #708 2012-03-19 11:35:29 +01:00
flyingimmidev cef8de7675 parameter n_jobs added 2012-03-10 21:21:33 +01:00
flyingimmidev 7a218ffbbe parallelized 2012-03-08 20:29:03 +01:00
flyingimmidev e6fa5966c7 pip8 errors fixed 2012-03-07 21:55:42 +01:00
flyingimmidev 373522090c docstring updated and list append replaced with 
preallocated array
 2012-03-06 22:57:06 +01:00
flyingimmidev 5d0a63d51a bugfix: lstsq coefficients output needed to be transposed 
added: sparse linear regression supports now multiple-outcome
 2012-03-06 21:11:10 +01:00
Andreas Mueller aa3a161087 COSMIT pep8 2012-02-01 22:32:17 +01:00
Alexandre Gramfort 66d7b26d1c cosmit in doc + pep8 2012-01-31 18:09:08 +01:00
Gael varoquaux e4919e61bd Add a randomized logistic 2012-01-31 18:05:53 +01:00
Mathieu Blondel 5e3ea55706 Add Perceptron. 2012-01-29 02:34:12 +09:00
Andreas Mueller 85e986cb27 COSMIT removed unused imports 2012-01-28 15:32:37 +01:00
Mathieu Blondel 77d83b61f9 Remove CoefSelectTransformerMixin and use SelectorMixin instead. 
This fixes issue #518.
 2012-01-25 18:04:27 +09:00
Andreas Mueller 5c2a8696e3 COSMIT removed unused imports 2012-01-23 21:13:55 +01:00
Mathieu Blondel 0ae85d2034 Make sure order="C". 2012-01-19 23:08:35 +09:00
Mathieu Blondel caae5c3f64 Copy in user-land. 2012-01-18 19:41:18 +09:00
Mathieu Blondel a438087d40 warm_restart -> warm_start 2012-01-18 19:08:08 +09:00
Mathieu Blondel 98ec09b1c2 More intuitive warm-restart in SGD. 2012-01-18 18:56:41 +09:00
Lars Buitinck d5f02dae3d Merge branch 'master' into merge-linearsvcs 
Conflicts:
	examples/document_classification_20newsgroups.py
 2012-01-17 09:21:33 +01:00
Lars Buitinck a9470d12ca ENH merge dense/sparse LinearSVC, part 2: no more sparse.CoefSelectTransformer 2012-01-16 18:05:40 +01:00
Mathieu Blondel 5d18b0dec7 pep8! 2012-01-16 16:34:06 +09:00
Mathieu Blondel e50415cb2f Address @ogrisel and @larsmans 's comments. 2012-01-16 16:30:32 +09:00
Mathieu Blondel aca8445fc5 Fix bug when fit is called mutiple times. 2012-01-15 07:18:00 +09:00
Mathieu Blondel 0117b601aa More tests + fixes. 2012-01-15 07:11:58 +09:00
Mathieu Blondel 8ec3395029 Save iteration number. 2012-01-14 23:46:25 +09:00
Mathieu Blondel f9701d34b8 Add deprecation warning. 2012-01-14 03:45:46 +09:00
Mathieu Blondel 0d48f49bbf Merge dense and sparse SGD implementations. 2012-01-14 03:21:40 +09:00
Mathieu Blondel fec3557f14 Add decision_function to SGDRegressor. 2012-01-14 00:19:20 +09:00
Mathieu Blondel 5048670baf Fix more bugs + tests. 2012-01-14 00:16:11 +09:00
Mathieu Blondel 7b62b29850 Fix a few more bugs. 2012-01-14 00:02:08 +09:00
Mathieu Blondel 944ff6e81e Partial tests + fix bugs. 2012-01-13 23:35:34 +09:00
Mathieu Blondel 1e534cec90 Add partial_fit to SGDRegressor. 2012-01-13 22:02:34 +09:00
Mathieu Blondel 388e99d793 Create partial_fit and call partial_fit from fit. 2012-01-13 21:57:04 +09:00
Mathieu Blondel 68236ab14c Add decision_function to ElasticNet. 
This allows to plug ElasticNet and Lasso into the multiclass estimators.
 2012-01-10 16:22:05 +09:00
Peter Prettenhofer 4ddf69eead cosmit: more detailed doc string for why fortran style arrays 2012-01-09 09:32:27 +01:00
Peter Prettenhofer 6cadc54d1d set coef_ to fortran layout after fit - this will enhance the test time performance for predicting singe data points. 2012-01-09 09:06:21 +01:00
Lars Buitinck 5b00f4414a BUG refactor SGD classes to not store sample_weight 2012-01-08 17:41:52 +01:00