baohan
/
scikit-learn
mirror of https://github.com/scikit-learn/scikit-learn.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
8,553 Commits 31 Branches 142 Tags 892 MiB
Commit Graph

44 Commits

Author SHA1 Message Date
GaelVaroquaux 10f18e6655 Revert "ENH: small speedups in coordinate descent" 
This reverts commit d434a39e76.

Apparently it creates numerical instabilities on some systems.
 2012-02-25 19:18:01 +01:00
GaelVaroquaux d434a39e76 ENH: small speedups in coordinate descent 
speedup of only ~ 5%, but for little cost
 2012-02-25 19:04:21 +01:00
Alexandre Gramfort ad8e4912cb TST: use assert_true instead of assert + remove some relative imports 2012-02-11 15:43:03 +01:00
Lars Buitinck 6fb7243728 DOC rm unused param from sparse.ElasticNet docstring 2012-01-28 20:38:22 +01:00
Andreas Mueller cb00e97dc2 COSMIT pep8 2012-01-23 10:02:01 +01:00
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 278fe52ba2 COSMIT refactor liblinear bindings 
Move knowledge of the CSR matrix format down the chain.
 2012-01-16 23:34:03 +01:00
Lars Buitinck 0b0bcfc311 ENH merge dense/sparse LinearSVC, part 4: deprecate sparse.LogisticRegression 2012-01-16 21:34:55 +01:00
Lars Buitinck a9470d12ca ENH merge dense/sparse LinearSVC, part 2: no more sparse.CoefSelectTransformer 2012-01-16 18:05:40 +01:00
Lars Buitinck 263d6c02f7 ENH merge dense/sparse LinearSVC, part 1: no more SparseBaseLibLinear 2012-01-16 17:55:47 +01:00
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 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 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 746d686a4d remove obsolete `sparse_coef_` doc string 2012-01-08 18:34:33 +01:00
Lars Buitinck 5b00f4414a BUG refactor SGD classes to not store sample_weight 2012-01-08 17:41:52 +01:00
Andreas Mueller 78f64c54d8 FIX two more underscores 2011-12-28 00:03:34 +01:00
Andreas Mueller 26a793459e COSMIT underscore fixes 2011-12-22 21:05:27 +01:00
Andreas Mueller 6c17d0d8af COSMIT make 'References' bold and minor other fixes. 2011-12-22 20:57:54 +01:00
Alexandre Gramfort 4298760a69 ENH : adding missing scale_C in docstring 2011-12-18 12:53:34 +01:00
Alexandre Gramfort db74b538e5 Merge remote-tracking branch 'origin/master' into n_samples_scaling 
Conflicts:
	sklearn/svm/sparse/base.py
	sklearn/svm/tests/test_sparse.py
 2011-12-18 12:45:58 +01:00
Alexandre Gramfort 8ec89a9fc2 ENH : C scaling of sparse models 2011-12-18 12:39:26 +01:00
Andreas Mueller 25c93f4af6 COSMIT: pep8 2011-12-15 17:24:53 +01:00
Gilles Louppe c40dbf739f DOC: Standardize the module documentation format (work in progress) 2011-11-30 10:30:13 +01:00
Andreas Mueller 78fa279dc6 COSMIT more pep8 2011-11-28 23:55:33 +01:00
Peter Prettenhofer 87cf1b32dd fix: sgd module clone issue w/ rho parameter 2011-11-15 21:06:32 +01:00
Lars Buitinck 0f1506633d COSMIT no need for csr_matrix "cast" in coord descent 2011-11-11 23:38:23 +01:00
Lars Buitinck 6cf3040f8b COSMIT rm comment in coord descent code about np.dot 
Code was changed to use safe_sparse_dot in
a8e45bd908
 2011-11-11 23:32:12 +01:00
Peter Prettenhofer a8e45bd908 use safe_sparse_dot instead of np.dot 2011-11-11 09:00:53 +01:00
Lars Buitinck e0ec08946a BUG fix SGD doctests 2011-11-07 14:08:06 +01:00
Olivier Grisel 1e4e6012ac ENH: make it possible to pass class_weight='auto' as constructor param for SGDClassifier 2011-11-01 19:00:50 +01:00
Lars Buitinck e6d5ed0c69 Revert "ENH prevent copy in sparse.LogisticRegression" 
This reverts commit b453e1e721.
 2011-10-30 21:23:49 +01:00
Lars Buitinck b453e1e721 ENH prevent copy in sparse.LogisticRegression 2011-10-30 19:57:45 +01:00
Lars Buitinck 3bde2bf97b ENH accept matrix input throughout 
Simple solution: replace np.asanyarray with np.asarray throughout.
Also, use utils.as_float_array where appropriate.
 2011-10-21 13:51:12 +02:00
Peter Prettenhofer b8b655f460 added X.shape[0] == y.shape[0] check to ElasitcNet 2011-10-02 18:41:46 +02:00
Peter Prettenhofer 8a20cb97e7 unified predict for sparse and dense SGD. 2011-10-02 18:15:33 +02:00
Lars Buitinck 02ec8e41a0 prevent some copying in sparse SGD 
Some methods would copy CSR input matrices twice.
 2011-09-26 17:17:29 +02:00
Mathieu Blondel b93cde81bd Documenting a secret feature and fixing bugs in the process. 2011-09-06 19:12:00 +09:00
Olivier Grisel 091d92ac1b batch re-cythonization with version 0.15 and new package names 2011-09-03 13:28:32 +02:00
Fabian Pedregosa ddf4b72109 Move project directory from scikits.learn to sklearn 2011-09-02 12:06:57 +02:00
Fabian Pedregosa 68f27e3790 Revert "Move project directory from scikits.learn to sklearn" 
This reverts commit fd0d3b879d.
 2011-09-02 12:03:18 +02:00
Fabian Pedregosa fd0d3b879d Move project directory from scikits.learn to sklearn 2011-09-02 11:38:24 +02:00