a87011262a
FIX precision to float64 across the codebase
2015-10-16 11:20:29 +02:00
4366ba500d
use sklearn.utils.random
2015-09-16 15:24:28 -07:00
4af0c49f27
vocabulary of type set now coerced to list to preserve iteration ordering after serialization
2015-09-16 15:24:28 -07:00
250e509e86
ENH O(1) stop-word lookup when list provided
...
The docstring says stop_words can be a list, but it should be accessed as a set.
2015-08-08 22:55:13 +10:00
a8626b36a6
TST/COSMIT remove nose call boilerplate
2015-05-28 14:54:01 -04:00
cd2ee7e454
MAINT docstring --> comments to prevent nose from using doc in verbose mode
2015-03-21 11:16:49 +05:30
f1b8283c95
catch some warnings, be less verbose in testing.
2015-03-02 14:40:35 -05:00
73e5cf5dbf
Merge pull request #3933 from fabianp/loss_liblinear
...
MAINT Change loss names for LinearSVC and LinearSVR
2015-02-10 17:49:10 +01:00
5c0c5c672c
MAINT use absolute imports in tests
...
as per the guideline in:
http://scikit-learn.org/stable/developers/#coding-guidelines
2015-02-10 13:40:31 +01:00
dbc5d707c6
Change loss names for LinearSVC and LinearSVR()
...
the names are now consistent across methods
In LinearSVC:
'l1' -> 'hinge'
'l2' -> 'squared_hinge'
In LinearSVR:
'l1' -> 'epsilon_insensitive'
'l2' -> 'squared_epsilon_insensitive'
2015-02-09 11:09:16 +01:00
21369dd6a1
TST Removal or modification of stop_words_ should not affect transform.
...
DOC Add a line to {Count, Tfidf}Vectorizer about removal of stop_words_
DOC Add documentation of stop_words_ attr in TfidfVectorizer
2015-02-02 02:20:04 +05:30
eb660edbf3
FIX: Raise error when patch width/height is greater than image width/height
2014-11-21 10:23:47 -05:00
cd7b43ccf9
TST make catch_warnings blocks more robust
2014-10-12 18:20:18 +02:00
324e3c9419
ENH sort option for memory-efficient DictVectorizer
...
Added memory efficient UnsortedDictVectorizer
When loading really large files for SKLL, I found that temporarily
storing a list of dictionaries to pass to DictVectorizer was frequently
using up huge amounts of memory. You can now call `fit_transform` on an
iterable, and not have to waste the temporary space.
Sorting is done in-place, so it doesn't waste memory.
2014-09-22 16:44:31 +02:00
8680a6bae1
Deprecate vectorizer fixed_vocabulary attribute
2014-08-13 10:54:44 +02:00
143e5ffc37
FIX set vectorizer vocabulary outside of init
2014-08-13 10:54:44 +02:00
56057c9630
MAINT remove deprecated code
2014-07-23 16:19:40 +02:00
4bf824cd7e
TST non-regression test for CV on text pipelines
2014-07-16 16:35:41 +02:00
f208316de6
Added a comment to explain the use of a test.
2014-07-14 22:13:41 +02:00
fb62b9fa7e
Replaced assert_raises with assert_raise_message as the point is to make sure the exception message is clear.
2014-07-14 16:08:56 +02:00
d1fe163bbc
PEP8 line length fix.
2014-07-14 15:26:12 +02:00
6e2af35483
Added a test on hashing vectorizer behavior with np.nan input.
2014-07-14 15:23:21 +02:00
92add1daa1
FIX TfidfVectorizer exports idf_ attribute
2014-05-23 17:51:15 +08:00
39b859b95e
FIX TfidfVectorizer to no longer ignore binary param
...
Also changed the docs to clarify that binary=True means
binary tf, not binary output.
2014-03-24 11:18:11 +01:00
89d94ca987
TST older nosetests compat in DictVectorizer test
2014-02-23 18:38:10 +01:00
bd0cf75afa
FIX DictVectorizer handling of empty inputs
...
Fixes #2883 .
2014-02-23 18:03:50 +01:00
dbc26ce996
COSMIT remove unused imports and variables
2014-02-03 19:43:56 +11:00
c997cdee9f
COSMIT pep8 + full stop police
...
metrics/tests/test_score_objects.py made stricter by replacing . with \.
in regular expressions.
2014-02-01 15:28:53 +01:00
6ebcf4d020
COSMIT fix PEP8 errors
2014-02-02 00:30:28 +11:00
cf7cc7096c
ENH filter out zeros early in FeatureHasher
...
Fixes #2665 .
2013-12-27 15:32:27 +01:00
0bd0552bdf
ENH: Raise explicitly on non-unique vocab.
2013-12-09 23:15:54 +01:00
4c10c82dbb
ENH: address discussion
2013-11-29 13:55:39 +01:00
e1bdd99ea6
ENH: add assert_warn_message
2013-11-29 13:55:39 +01:00
d2f3a8a3da
ENH: refactor warnings 1
2013-11-29 13:55:38 +01:00
da4f2c6357
FIX regression in CountVectorizer handling of float min_df/max_df
...
Fixes #2595 .
2013-11-18 18:42:15 +01:00
baad8de295
COSMIT pyflakes feature_extraction.text tests
2013-10-02 19:10:37 +02:00
94d66a0e38
FIX max_features in CountVectorizer
...
Fixes #2443 : max_features would be selected based on document frequency
rather than term frequency.
2013-10-02 19:10:37 +02:00
c2cf21d8e7
FIX+TST non-consecutive or duplicate vocabulary indices
...
* added extra checks for custom vocab in CountVectorizer
* added test for custom vocab to check fauly vocabs
Fixes #2357 .
2013-08-15 12:52:52 +02:00
a29392eaa8
TST catch deprecation warning in feature_extraction.text
2013-07-26 10:17:50 +02:00
8c9dbc99dc
TST added tests for encoding/charset deprecation
...
Checking the old interface still works.
2013-07-26 10:01:05 +02:00
88fd230439
MAINT charset is deprecated in favor of encoding
...
closes #2107
2013-07-26 09:30:58 +02:00
baae05ec2f
Two more spots. I think that's it
2013-07-22 10:27:25 +02:00
77695ee27a
COSMIT pep8
2013-07-11 11:50:23 +02:00
d6e9598abe
Merge pull request #1880 from NicolasTr/patch_extractor_float_max_patches
...
Support float values for max_patches in PatchExtractor.transform
2013-05-12 02:20:13 -07:00
5d8a4d909d
TST test CountVectorizer.stop_words_ value
2013-05-07 14:43:04 +02:00
e163f8bfef
FIX DictVectorizer behavior on empty X and empty samples
...
More useful error message for former; return all zeros for latter.
Also changed integer type to np.intc, which matches scipy.sparse
index types more closely.
Fixes #1903 .
2013-05-04 17:09:07 +02:00
dacfd8bd5d
DOC: Replaced all BSD style licenses with "BSD 3 clause"
...
Replaced all BSD style licenses with "BSD 3 clause"
Not checked yet!
Removed duplicate "3 clause, 3 clause"
Removed trailing period if exists
Fixed some missed licences, still about 50 to do, but those can be automated
Think I got the last of them.
Apparently me and sed have different ideas of regex.
Found a few more
2013-04-30 08:34:46 +02:00
a85c89668c
Add test for PatchExtractor (float value for max_patches)
2013-04-21 15:29:47 +02:00
19591be3a5
P3K all of feature_extraction passes tests on Py2 and 3
2013-03-11 12:26:05 +01:00
09f2d90262
P3K make feature_extraction.text work
...
All tests pass with Py2 and Py3.
2013-03-10 22:20:29 +01:00
Raghav R V