2694e50fe5
Merge pull request #1556 from larsmans/cython-cleanup
...
MRG clean up Cython and C code
2013-01-10 09:03:11 -08:00
37c1d092f8
COSMIT shut up the build by calling np.import_array in Cython modules
...
$(make in 2>&1 | wc -l) down from 706 to 517.
Took the liberty of modernizing some of the Cython code while I was at it.
2013-01-10 18:00:21 +01:00
68fd7a3646
fix: docstring for power_t in SGDClassifier was not correct (0.25 instead of 0.5)
2013-01-10 17:18:59 +01:00
11cdef7e6d
ENH common test that set_params returns self.
2013-01-09 09:27:53 +01:00
127c44b730
ENH refactoring class weights for SVM and SGD
2013-01-03 13:21:50 +01:00
b44e779b33
threshold parameter was ignored in SquaredHinge loss.
2013-01-03 07:56:41 +09:00
5a6f7b5fec
FIX stupid mistake
2012-12-27 00:28:01 +01:00
983b45ccc9
COSMIT pep8
2012-12-22 16:48:17 +01:00
966e10c252
COSMIT PEP8
2012-12-22 16:48:17 +01:00
754c2d9533
COSMIT pep8
2012-12-22 16:48:17 +01:00
270bc8e278
TST some more tests for SGDClassifier input validation
2012-12-15 22:10:48 +01:00
8eda5e1c0e
TST document and test verbosity parameter of lars_path
2012-12-15 21:18:10 +01:00
08dff76e2c
ENH input validation only in ``fit`` in LassoLarsIC, check that error is raised.
2012-12-15 21:17:51 +01:00
116b54a8c8
Fix bug in sqnorm (used by PassiveAggressive).
...
The function was using the wrong data indices T_T.
2012-11-24 23:05:52 +09:00
23ac37a4fb
API : move isotonic regression out of linear_model
2012-11-14 22:01:53 +01:00
c63b7a1f55
P3K: Fixed imports.
2012-11-14 07:22:38 +01:00
c97e281401
FIX #1354 : machine precision assertion failure in test_liblinear_random_state
2012-11-10 20:54:07 +01:00
c9cd2b7292
FIX: lars drop for good platform specific test failure
2012-11-10 20:41:21 +01:00
7cec21bbee
TEST: decrease precision in test_lars_drop_for_good
...
If the test is wrong, this will be well off. Small errors just reflect
numerical errors
2012-11-10 19:43:17 +01:00
2f0dd13e4d
DOC added comment about default for n_nonzero_coefs.
2012-11-10 15:53:58 +00:00
a351355510
ENH add checks for clustering, regressors, transformers
2012-11-10 15:53:57 +00:00
92138a4258
COSMIT pep8
2012-11-10 14:02:29 +00:00
fde726c751
DOC: n_responses -> n_targets
...
For consistency
2012-11-09 17:57:07 +01:00
970f8b2148
DOC: LinearRegression document the shape of coef_
2012-11-09 17:43:11 +01:00
9e52350533
BUG: compat with numpy 1.3
2012-11-08 11:02:54 +01:00
25707c2c54
BUG: missing import introduced by rebase
2012-11-08 10:40:10 +01:00
de4fd325af
COSMIT: better comments
...
Specifying exactly why a test is around is important
2012-11-08 10:39:22 +01:00
1f3f49755a
TST: difficult test for early stopping
2012-11-08 10:39:22 +01:00
16dbe52125
ENH lars_path: early stopping after drop for good
...
Put early stopping after dropping regressors, so that the 'drop for good'
code has a chance to kick in
2012-11-08 10:39:22 +01:00
428439f7ca
MISC: comment
2012-11-08 10:39:22 +01:00
2d9435c366
TST: add a test for lasso and lars
2012-11-08 10:39:22 +01:00
e290951871
ENH: early stopping for lars
...
When the noise in the updates of the Cholesky becomes greater than the
max residual, it is impossible to go on in the path: giving up.
2012-11-08 10:39:22 +01:00
c3b5ea2fa8
MISC lars_path: cleaner code in degenerate case
2012-11-08 10:39:22 +01:00
f739070d5f
WIP: drop for good correlated regressors
...
Attempt to drop for good correlated regressors. In Lars. Doesn't seem to
be working, as it is give a significantly different result than the
lasso (failing test)
2012-11-08 10:39:22 +01:00
caf995c967
MISC: more precise warning
2012-11-08 10:32:58 +01:00
a911ad10fb
MISC: address comments
2012-11-08 10:32:58 +01:00
156566e6a9
ENH: early stopping LARS for degenerate active set
2012-11-08 10:31:13 +01:00
e18465da6a
ENH: multi-target Lars: lists rather than arrays
...
Because the path length varies, arrays lead to wasted memory and trailing
zeros (or ones, depending on the implementation).
2012-11-08 10:24:56 +01:00
65b79cfd5e
BUG: lars corner case with path length == 1
...
Rather than 'squeeze', be specific with regards to which direction we
squeeze. This is more controlled.
2012-11-08 10:24:56 +01:00
48dd2ac338
TEST: test that alpha is decreasing in LassoLars
2012-11-08 10:24:56 +01:00
e115469160
TEST: one addition test on the length of the path
2012-11-08 10:24:56 +01:00
4f41de9f35
BUG: LassoLars path ending contained junk
...
Addresses mailing list discussion
http://sourceforge.net/mailarchive/forum.php?thread_name=20120918124544.GF27767%40phare.normalesup.org&forum_name=scikit-learn-general
2012-11-08 10:24:56 +01:00
34037306ab
Remove transform from PassiveAggressive*.
...
It really only makes sense if the estimator supports L1-penalty.
2012-11-08 15:21:19 +09:00
a3670c0fb4
Remove predict_proba from Perceptron and PassiveAggressiveClassifier.
...
Fixes #1336 .
2012-11-07 14:18:34 +09:00
326426826c
Better documentation for epsilon in SGD.
2012-11-06 16:11:34 +09:00
5706109df0
Document epsilon.
2012-11-06 16:08:11 +09:00
27be154754
Add tests for partial fit.
2012-11-06 01:07:44 +09:00
0f92da4914
Remove sample_weight and class_weight from PassiveAggressive*.
2012-11-06 00:53:50 +09:00
663a6aeb0a
Better documentation for C.
2012-11-05 23:39:55 +09:00
4661b64a36
Fix underflow detected by test_common :)
2012-11-05 22:35:53 +09:00
Lars Buitinck