Matthew Honnibal
ecf91a2dbb
Support beam parser
2017-03-10 11:21:21 -06:00
Matthew Honnibal
ea53647362
Merge branch 'develop'
2017-03-10 02:49:39 -06:00
Matthew Honnibal
b5247c49eb
Merge branch 'develop' of https://github.com/explosion/spaCy into develop
2017-03-09 18:45:43 -06:00
Matthew Honnibal
798450136d
Set L1 penalty to 0 in tagger.
2017-03-09 18:43:47 -06:00
Matthew Honnibal
c62da02344
Use ftrl training, to learn compressed model.
2017-03-09 18:43:21 -06:00
Matthew Honnibal
f71eeef9bb
Pass path argument to end_training
2017-03-09 18:42:40 -06:00
Dan Rapp
123d3f2d38
Fix error in test case parameterization
2017-03-09 12:18:21 -07:00
Dan Rapp
b9307dfcd7
Merge branch 'master' into rappdw/tokenizer_exceptions_url_fix
2017-03-09 11:42:14 -07:00
Dan Rapp
3b1df3808d
Issue #840 - URL pattenr too broad
2017-03-09 11:39:39 -07:00
Matthew Honnibal
5b0b968d13
Merge branch 'develop' of https://github.com/explosion/spaCy into develop
2017-03-08 15:03:10 +01:00
Matthew Honnibal
0ac3d27689
Fix handling of trailing whitespace
...
Fix off-by-one error that meant trailing spaces were being dropped.
Closes #792
2017-03-08 15:01:40 +01:00
ines
c2e3e651b8
Re-add regression test for #859
2017-03-08 14:36:09 +01:00
Matthew Honnibal
0a6d7ca200
Fix spacing after token_match
...
The boolean flag indicating a space after the token was
being set incorrectly after the token_match regex was applied.
Fixes #859 .
2017-03-08 14:33:32 +01:00
Matthew Honnibal
cd33b39a04
Fix 2/3 problem for json save/load
2017-03-08 01:39:13 +01:00
Matthew Honnibal
40703988bc
Use FTRL training in parser
2017-03-08 01:38:51 +01:00
Matthew Honnibal
d108534dc2
Fix 2/3 problems for training
2017-03-08 01:37:52 +01:00
Matthew Honnibal
d03d6a13f1
Merge branch 'rominf-ud20' into develop
2017-03-07 21:48:56 +01:00
Matthew Honnibal
f7374d0b86
Merge branch 'ud20' of https://github.com/rominf/spaCy into rominf-ud20
2017-03-07 21:48:37 +01:00
Matthew Honnibal
16670d3251
Xfail the vocab pickling for now
2017-03-07 21:43:28 +01:00
Matthew Honnibal
a89c3500f6
Fixes to hacky vocab pickling
2017-03-07 20:58:55 +01:00
Matthew Honnibal
d814892805
Hackish pickle support for Vocab.
2017-03-07 20:25:12 +01:00
Matthew Honnibal
26614e028f
Add hacky support for StringCFile, to make pickling easier.
2017-03-07 20:24:37 +01:00
Matthew Honnibal
3edb8ae207
Whitespace
2017-03-07 17:16:26 +01:00
Matthew Honnibal
5de7e712b7
Add support for pickling StringStore.
2017-03-07 17:15:18 +01:00
Matthew Honnibal
4e75e74247
Update regression test for variable-length pattern problem in the matcher.
2017-03-07 16:08:32 +01:00
Matthew Honnibal
6d67213b80
Add test for 850: Matcher fails on zero-or-more.
2017-03-07 15:55:28 +01:00
Aniruddha Adhikary
696215a3fb
add tests for Bengali
2017-03-05 11:25:12 +06:00
Aniruddha Adhikary
8f3bfe9bfc
[Bengali] basic tag map, morph, lemma rules and exceptions
2017-03-04 12:36:59 +06:00
Roman Inflianskas
66e1109b53
Add support for Universal Dependencies v2.0
2017-03-03 13:17:34 +01:00
ines
8dff040032
Revert "Add regression test for #859 "
...
This reverts commit c4f16c66d1
2017-03-01 21:56:20 +01:00
ines
c4f16c66d1
Add regression test for #859
2017-03-01 16:07:27 +01:00
Aniruddha Adhikary
d91be7aed4
add punctuations for Bengali
2017-02-28 21:07:14 +06:00
Aniruddha Adhikary
5a4fc09576
add basic Bengali support
2017-02-28 07:48:37 +06:00
Matthew Honnibal
cc9b2b74e3
Merge branch 'french-tokenizer-exceptions'
2017-02-27 11:44:39 +01:00
Matthew Honnibal
bd4375a2e6
Remove comment
2017-02-27 11:44:26 +01:00
Matthew Honnibal
e7e22d8be6
Move import within get_exceptions() function, to speed import
2017-02-27 11:34:48 +01:00
Matthew Honnibal
34bcc8706d
Merge branch 'french-tokenizer-exceptions'
2017-02-27 11:21:21 +01:00
Matthew Honnibal
0aaa546435
Fix test after updating the French tokenizer stuff
2017-02-27 11:20:47 +01:00
Matthew Honnibal
26446aa728
Avoid loading all French exceptions on import
...
Move exceptions loading behind a get_tokenizer_exceptions() function
for French, instead of loading into the top-level namespace. This
cuts import times from 0.6s to 0.2s, at the expense of making the
French data a little different from the others (there's no top-level
TOKENIZER_EXCEPTIONS variable.) The current solution feels somewhat
unsatisfying.
2017-02-25 11:55:00 +01:00
ines
376c5813a7
Remove print statements from test
2017-02-24 18:26:32 +01:00
ines
7c1260e98c
Add regression test
2017-02-24 18:22:49 +01:00
ines
0e2e331b58
Convert exceptions to Python list
2017-02-24 18:22:40 +01:00
ines
51eb190ef4
Remove print statements from test
2017-02-24 17:41:12 +01:00
Matthew Honnibal
db5ada3995
Merge branch 'master' of https://github.com/explosion/spaCy
2017-02-24 14:28:12 +01:00
Matthew Honnibal
8f94897d07
Add 1 operator to matcher, and make sure open patterns are closed at end of document. Closes Issue #766
2017-02-24 14:27:02 +01:00
ines
67991b6e5f
Add more test cases to #775 regression test to cover #847
2017-02-18 14:10:44 +01:00
ines
30ce2a6793
Exclude "shed" and "Shed" from tokenizer exceptions (see #847 )
2017-02-18 14:10:44 +01:00
Ines Montani
de997c1a33
Merge pull request #842 from magnusburton/master
...
Added regular verb rules for Swedish
2017-02-17 11:18:20 +01:00
Magnus Burton
41fcfd06b8
Added regular verb rules for Swedish
2017-02-17 10:04:04 +01:00
ines
aa92d4e9b5
Fix unicode regex for Python 2 (see #834 )
2017-02-16 23:49:54 +01:00
