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
60ae168464
COSMIT rm unused import
2012-10-16 14:36:49 +02:00
a06c367ba2
ENH sparse matrix support in randomized logistic regression
2012-10-16 14:34:48 +02:00
8d3c75feff
Store X as an ordered array.
...
This simplifies and makes more efficient the `fit` and transform
method (re-ordering in transform is no longer needed).
2012-10-16 14:18:43 +02:00
04251e155d
Better tests + cosmetic changes.
2012-10-16 14:18:43 +02:00
342c5be9f1
ENH: Consider order in X for IsotonicRegression.
...
Before, IsotonicRegression assumed X was an increasing sequence. This
allows to shuffle both X, y (see the test) and obtain the same
result. This makes IsotonicRegression behave more like a standard
Regression (which is what I would expect).
2012-10-16 14:18:43 +02:00
72e71bb439
ENH address @agramfort's comments, fix some doctests
2012-10-11 20:56:03 +01:00
b39609af4e
ENH rename rho in SGD
2012-10-11 20:56:03 +01:00
16aabc7c3f
ENH rename rho to l1_ratio in ElasticNet and friends
2012-10-11 20:56:03 +01:00
244e8a68fa
Merge pull request #1187 from ogrisel/bugfix-logistic-ovr-probabilities
...
MRG: FIX: wrong probabilities for OvR LogisticRegression
2012-09-30 06:09:51 -07:00
ec07896195
Add minimalistic test for each solver.
2012-09-29 02:16:05 +09:00
fa93e209ff
FIX: wrong probabilities for OvR LogisticRegression
2012-09-28 19:13:46 +02:00
c0a057843c
ENH rename Scaler to StandardScaler everywhere
2012-09-23 17:31:15 +01:00
a164d57e74
TST writable coef_ and intercept_ on LogisticRegression
...
Solves issue #470 .
2012-09-20 23:04:52 +02:00
1313f25cda
COSMIT pep8
2012-09-20 14:49:14 +01:00
4183cb3b34
COSMIT use LinearClassifierMixin in RidgeClassifier
...
Had to change a test due to different tie breaking.
2012-09-17 23:22:16 +02:00
88a0e5dccf
TST improve Ridge test
2012-09-17 23:22:16 +02:00
6f13021eac
Merge pull request #1143 from larsmans/refactor-liblinear
...
Refactor liblinear bindings: rewrote prediction code in Python
2012-09-16 03:28:21 -07:00
85693d6272
Fix confusing notation in isotonic regression.
...
I replaced x to y_ to emphasize the fact that y_ is fitted
and because x had nothing to do with X used in
IsotonicRegression.fit.
2012-09-16 01:40:23 +09:00
1acd2471f6
FIX : fix LLE test (don't ask me why...)
2012-09-15 17:29:55 +02:00
d8d4af0973
ENH : finish addressing @GaelVaroquaux comments + improve coverage + add linear regression to example
2012-09-15 16:40:44 +02:00
4bc740cfb7
ENH : proper IsotonicRegression model + example + test
2012-09-15 16:40:43 +02:00
c1abbf2214
DOC - added paragraph in user documentation on the isotonic regression + an example plot.
...
Fixed a couple of things from the isotonic regression such as better type checking and renaming variables
2012-09-15 16:40:43 +02:00
2734835715
Added tests to isotonic_regression
2012-09-15 16:40:43 +02:00
5528334660
Robustify LARS. Fixes issue #487
...
Avoid division by zero when the y variable sums to zero.
2012-09-14 15:30:15 +02:00
5ecb9ac332
add predict_log_proba and test; better docstrings
2012-09-14 14:44:34 +02:00
706319655a
ENH refactor liblinear prediction code and add classes_ member
...
Liblinear bindings removed from LinearSVC and LogisticRegression
prediction code in favor of NumPy code.
* no more copying of X at predict time
* no more unchecked malloc at predict time
* -4000 lines of C code
* simpler handling of coef_ and intercept_
Also, support symbolic class names and classes_ on both estimators.
2012-09-13 00:16:42 +02:00
ac5e028b78
added test to expose the predict_proba w/ sparse matrix regression
2012-09-12 18:38:03 +02:00
185d32693f
add test for proper loss instantiation
2012-09-06 10:05:13 +02:00
b4041c33b8
added failing test for clone
2012-09-06 09:34:21 +02:00
713ba6006e
COSMIT pep8
2012-09-04 17:24:01 +01:00
50868f2aef
make smoke tests explicit; check ValueError on 2d inputs
2012-09-03 10:34:47 +02:00
0441dc58b5
Merge branch 'master' into sgd-yshape-fix
2012-09-03 10:25:39 +02:00
466479fe03
API : consistent use alpha_/alphas_ for alpha/alphas estimated by CV in linear models (issue #1041 )
2012-09-02 23:16:54 +02:00
9e3354bb6e
use assert_true
2012-09-02 23:16:54 +02:00
eac812bca3
return_path option in lars_path
2012-09-02 23:16:53 +02:00
51d02d395e
ENH : use build_dataset in multi target test
2012-09-02 23:16:53 +02:00
36cdf1c408
ENH : add support for sparse data in ElasticNetCV/LassoCV (not optimal)
2012-09-02 23:16:52 +02:00
48687a77a4
ENH : add support for multitarget in sparse enet + simplify input checking
2012-09-02 23:16:52 +02:00
2c9ede570d
ENH: multidimensional y in ElasticNet (WIP)
2012-09-02 23:16:51 +02:00
519e204a95
FIX: add test, fix memory initialization bug
2012-09-02 23:16:51 +02:00
abc9cccc0b
TEST: precomputed lasso and lars
2012-09-02 23:16:51 +02:00
fd2a0a2a69
added failing test for 2d y
2012-08-30 08:22:15 +02:00
59cf938471
Merge branch 'sgd-predict-proba'
2012-08-29 12:04:27 +02:00
053d1a5c80
Merge branch 'master' into sgd-predict-proba
2012-08-29 11:38:23 +02:00
5d73ecd7c3
Merge pull request #1009 from amueller/one_class_check
...
MRG bugfix in SVM: classifier behavior with only one class present
2012-08-29 02:30:22 -07:00
e18a32d93b
Fix #1080 .
2012-08-28 21:33:35 +09:00
42e9029a1e
TST classifier behavior with only one class present
2012-08-27 13:36:43 +02:00
2311a4cf11
COSMIT pep8
2012-08-25 21:05:27 +02:00
Mathieu Blondel