explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-11-10 19:57:17 +03:00
Code Issues Packages Projects Releases Wiki Activity
13,004 Commits 68 Branches 175 Tags 279 MiB
feature/convert-old-conll
Commit Graph

272 Commits

Author SHA1 Message Date
Matthew Honnibal
9bf22a94aa Fix tag set serialisation 2017-05-29 17:52:36 -05:00
Matthew Honnibal
920887f4e4 Specify order of vocab deserialization 2017-05-29 13:04:40 +02:00
Matthew Honnibal
6b019b0540 Update to/from bytes methods 2017-05-29 10:14:20 +02:00
Matthew Honnibal
6dad4117ad Work on serialization for models 2017-05-29 01:37:57 +02:00
Matthew Honnibal
2edd96ce47 Draft Vocab to/from disk/bytes 2017-05-28 23:34:12 +02:00
Matthew Honnibal
fe11564b8e Finish stringstore change. Also xfail vectors tests 2017-05-28 15:10:22 +02:00
Matthew Honnibal
fe4a746300 Accomodate symbols in new string scheme 2017-05-28 13:03:16 +02:00
Matthew Honnibal
a5606c3eda Work on changing StringStore to return hashes. 2017-05-28 12:36:27 +02:00
Matthew Honnibal
39293ab2ee Merge branch 'develop' of https://github.com/explosion/spaCy into develop 2017-05-28 11:46:57 +02:00
Matthew Honnibal
15f6efc127 Remove vectors from vocab 2017-05-28 11:45:32 +02:00
ines
c8543c8237 Fix formatting and docstrings and remove deprecated function 2017-05-28 00:22:40 +02:00
ines
251346b59f Fix typos and formatting 2017-05-21 14:18:46 +02:00
ines
d82ae9a585 Change "function" to "callable" in docs 2017-05-21 13:17:40 +02:00
ines
f0cc642bb9 Update docstrings and API docs for Vocab 2017-05-20 14:00:41 +02:00
Matthew Honnibal
793430aa7a Get spaCy train command working with neural network 
* Integrate models into pipeline
* Add basic serialization (maybe incorrect)
* Fix pickle on vocab
 2017-05-17 12:04:50 +02:00
Matthew Honnibal
9e167b7bb6 Strip serializer from code 2017-05-09 17:28:50 +02:00
ines
e1efd589c3 Fix json imports and use ujson 2017-04-15 12:13:34 +02:00
ines
c05ec4b89a Add compat functions and remove old workarounds 
Add ensure_path util function to handle checking instance of path
 2017-04-15 12:11:16 +02:00
ines
d24589aa72 Clean up imports, unused code, whitespace, docstrings 2017-04-15 12:05:47 +02:00
ines
561f2a3eb4 Use consistent formatting for docstrings 2017-04-15 11:59:21 +02:00
Matthew Honnibal
d013aba7b5 Merge branch 'master' of https://github.com/explosion/spaCy 2017-03-17 18:30:53 +01:00
Matthew Honnibal
854cfce7cf Make vocabs more compatible across versions 
Previously, symbols were inserted into the string-store
before strings were loaded. This meant that adding a symbol
would invalidate saved models. We now make sure that strings
are loaded faithfully, so that compatibility is maintained.
 2017-03-17 18:29:04 +01:00
Matthew Honnibal
1cc841e600 Merge branch 'master' of https://github.com/explosion/spaCy 2017-03-17 08:18:11 -05:00
Matthew Honnibal
4bfc55b532 Auto-add words to vocab when loading vectors 
When calling vocab.load_vectors_from_bin_loc, ensure that missing
entries are added to the vocab. Otherwise, loading vectors into an
empty vocab object resulted in no vectors being added.
 2017-03-17 08:15:59 -05:00
Matthew Honnibal
4382f175b3 Squelch compiler warnings 2017-03-11 12:44:43 -06:00
Matthew Honnibal
d814892805 Hackish pickle support for Vocab. 2017-03-07 20:25:12 +01:00
ines
aa92d4e9b5 Fix unicode regex for Python 2 (see #834) 2017-02-16 23:49:54 +01:00
ines
85d249d451 Revert "Revert "Merge pull request #836 from raphael0202/load_vectors (closes #834)"" 
This reverts commit ea05f78660.
 2017-02-16 23:26:25 +01:00
ines
ea05f78660 Revert "Merge pull request #836 from raphael0202/load_vectors (closes #834)" 
This reverts commit 7d8c9eee7f, reversing
changes made to f6b69babcc.
 2017-02-16 15:27:12 +01:00
Raphaël Bournhonesque
e17dc2db75 Remove useless import 2017-02-16 12:10:24 +01:00
Raphaël Bournhonesque
3fd2742649 load_vectors should accept arbitrary space characters as word tokens 
Fix bug  #834
 2017-02-16 12:08:30 +01:00
Daniel Hershcovich
99eb494a82 Fix #737: support loading word vectors with " " as a word 2017-01-12 17:00:14 +02:00
Daniel Hershcovich
8e603cc917 Avoid "True if ... else False" 2017-01-11 11:18:22 +02:00
Matthew Honnibal
cade536d1e Merge branch 'master' of ssh://github.com/explosion/spaCy 2016-12-27 21:04:10 +01:00
Matthew Honnibal
ce4539dafd Allow the vocabulary to grow to 10,000, to prevent cold-start problem. 2016-12-27 21:03:45 +01:00
Ines Montani
8978806ea6 Allow Vocab to load without serializer_freqs 2016-12-21 18:05:23 +01:00
Ines Montani
be8ed811f6 Remove trailing whitespace 2016-12-21 18:04:41 +01:00
Matthew Honnibal
6ee1df93c5 Set tag_map to None if it's not seen in the data by vocab 2016-12-18 16:51:10 +01:00
Matthew Honnibal
1e0f566d95 Fix #656, #624: Support arbitrary token attributes when adding special-case rules. 2016-11-25 12:43:24 +01:00
Matthew Honnibal
f123f92e0c Fix #617: Vocab.load() required Path. Should work with string as well. 2016-11-10 22:48:48 +01:00
Matthew Honnibal
b86f8af0c1 Fix doc strings 2016-11-01 12:25:36 +01:00
Matthew Honnibal
6036ec7c77 Fix vector norm when loading lexemes. 2016-10-23 19:40:18 +02:00
Matthew Honnibal
3e688e6d4b Fix issue #514 -- serializer fails when new entity type has been added. The fix here is quite ugly. It's best to add the entities ASAP after loading the NLP pipeline, to mitigate the brittleness. 2016-10-23 17:45:44 +02:00
Matthew Honnibal
f62088d646 Fix compile error 2016-10-23 14:50:50 +02:00
Matthew Honnibal
a0a4ada42a Fix calculation of L2-norm for Lexeme 2016-10-23 14:44:45 +02:00
Matthew Honnibal
7ab03050d4 Add resize_vectors method to Vocab 2016-10-21 01:44:50 +02:00
Matthew Honnibal
f5fe4f595b Fix json loading, for Python 3. 2016-10-20 21:23:26 +02:00
Matthew Honnibal
5ec32f5d97 Fix loading of GloVe vectors, to address Issue #541 2016-10-20 18:27:48 +02:00
Matthew Honnibal
d10c17f2a4 Fix Issue #536: oov_prob was 0 for OOV words. 2016-10-19 23:38:47 +02:00
Matthew Honnibal
2bbb050500 Fix default of serializer_freqs 2016-10-18 19:55:41 +02:00
Matthew Honnibal
2cc515b2ed Add add_flag method to Vocab, re Issue #504. 2016-10-14 12:15:38 +02:00
Matthew Honnibal
ea23b64cc8 Refactor training, with new spacy.train module. Defaults still a little awkward. 2016-10-09 12:24:24 +02:00
Matthew Honnibal
ca32a1ab01 Revert "Work on Issue #285: intern strings into document-specific pools, to address streaming data memory growth. StringStore.__getitem__ now raises KeyError when it can't find the string. Use StringStore.intern() to get the old behaviour. Still need to hunt down all uses of StringStore.__getitem__ in library and do testing, but logic looks good." 
This reverts commit 8423e8627f.
 2016-09-30 20:20:22 +02:00
Matthew Honnibal
1f1cd5013f Revert "Changes to vocab for new stringstore scheme" 
This reverts commit a51149a717.
 2016-09-30 20:10:30 +02:00
Matthew Honnibal
a51149a717 Changes to vocab for new stringstore scheme 2016-09-30 20:01:19 +02:00
Matthew Honnibal
8423e8627f Work on Issue #285: intern strings into document-specific pools, to address streaming data memory growth. StringStore.__getitem__ now raises KeyError when it can't find the string. Use StringStore.intern() to get the old behaviour. Still need to hunt down all uses of StringStore.__getitem__ in library and do testing, but logic looks good. 2016-09-30 10:14:47 +02:00
Matthew Honnibal
95aaea0d3f Refactor so that the tokenizer data is read from Python data, rather than from disk 2016-09-25 14:49:53 +02:00
Matthew Honnibal
df88690177 Fix encoding of path variable 2016-09-24 21:13:15 +02:00
Matthew Honnibal
af847e07fc Fix usage of pathlib for Python3 -- turning paths to strings. 2016-09-24 21:05:27 +02:00
Matthew Honnibal
453683aaf0 Fix spacy/vocab.pyx 2016-09-24 20:50:31 +02:00
Matthew Honnibal
fd65cf6cbb Finish refactoring data loading 2016-09-24 20:26:17 +02:00
Matthew Honnibal
83e364188c Mostly finished loading refactoring. Design is in place, but doesn't work yet. 2016-09-24 15:42:01 +02:00
Matthew Honnibal
872695759d Merge pull request #306 from wbwseeker/german_noun_chunks 
add German noun chunk functionality
 2016-04-08 00:54:24 +10:00
Henning Peters
b8f63071eb add lang registration facility 2016-03-25 18:54:45 +01:00
Wolfgang Seeker
5e2e8e951a add baseclass DocIterator for iterators over documents 
add classes for English and German noun chunks

the respective iterators are set for the document when created by the parser
as they depend on the annotation scheme of the parsing model
 2016-03-16 15:53:35 +01:00
Wolfgang Seeker
03fb498dbe introduce lang field for LexemeC to hold language id 
put noun_chunk logic into iterators.py for each language separately
 2016-03-10 13:01:34 +01:00
Matthew Honnibal
963fe5258e * Add missing __contains__ method to vocab 2016-03-08 15:49:10 +00:00
Matthew Honnibal
478aa21cb0 * Remove broken __reduce__ method on vocab 2016-03-08 15:48:21 +00:00
Henning Peters
931c07a609 initial proposal for separate vector package 2016-03-04 11:09:06 +01:00
Matthew Honnibal
a95974ad3f * Fix oov probability 2016-02-06 15:13:55 +01:00
Matthew Honnibal
dcb401f3e1 * Remove broken Vocab pickling 2016-02-06 14:08:47 +01:00
Matthew Honnibal
63e3d4e27f * Add comment on Vocab.__reduce__ 2016-01-19 20:11:25 +01:00
Henning Peters
235f094534 untangle data_path/via 2016-01-16 12:23:45 +01:00
Henning Peters
846fa49b2a distinct load() and from_package() methods 2016-01-16 10:00:57 +01:00
Henning Peters
788f734513 refactored data_dir->via, add zip_safe, add spacy.load() 2016-01-15 18:01:02 +01:00
Henning Peters
bc229790ac integrate with sputnik 2016-01-13 19:46:17 +01:00
Matthew Honnibal
eaf2ad59f1 * Fix use of mock Package object 2015-12-31 04:13:15 +01:00
Matthew Honnibal
aec130af56 Use util.Package class for io 
Previous Sputnik integration caused API change: Vocab, Tagger, etc
were loaded via a from_package classmethod, that required a
sputnik.Package instance. This forced users to first create a
sputnik.Sputnik() instance, in order to acquire a Package via
sp.pool().

Instead I've created a small file-system shim, util.Package, which
allows classes to have a .load() classmethod, that accepts either
util.Package objects, or strings. We can later gut the internals
of this and make it a proxy for Sputnik if we need more functionality
that should live in the Sputnik library.

Sputnik is now only used to download and install the data, in
spacy.en.download
 2015-12-29 18:00:48 +01:00
Matthew Honnibal
0e2498da00 * Replace from_package with load() classmethod in Vocab 2015-12-29 16:56:51 +01:00
Henning Peters
8359bd4d93 strip data/ from package, friendlier Language invocation, make data_dir backward/forward-compatible 2015-12-18 09:52:55 +01:00
Henning Peters
9027cef3bc access model via sputnik 2015-12-07 06:01:28 +01:00
Matthew Honnibal
6ed3aedf79 * Merge vocab changes 2015-11-06 00:48:08 +11:00
Matthew Honnibal
1e99fcd413 * Rename .repvec to .vector in C API 2015-11-03 23:47:59 +11:00
Matthew Honnibal
5887506f5d * Don't expect lexemes.bin in Vocab 2015-11-03 13:23:39 +11:00
Matthew Honnibal
f11030aadc * Remove out-dated TODO comment 2015-10-26 12:33:38 +11:00
Matthew Honnibal
a371a1071d * Save and load word vectors during pickling, re Issue #125 2015-10-26 12:33:04 +11:00
Matthew Honnibal
314090cc78 * Set vectors length when unpickling vocab, re Issue #125 2015-10-26 12:05:08 +11:00
Matthew Honnibal
2348a08481 * Load/dump strings with a json file, instead of the hacky strings file we were using. 2015-10-22 21:13:03 +11:00
Matthew Honnibal
7a15d1b60c * Add Python 2/3 compatibility fix for copy_reg 2015-10-13 20:04:40 +11:00
Matthew Honnibal
20fd36a0f7 * Very scrappy, likely buggy first-cut pickle implementation, to work on Issue #125: allow pickle for Apache Spark. The current implementation sends stuff to temp files, and does almost nothing to ensure all modifiable state is actually preserved. The Language() instance is a deep tree of extension objects, and if pickling during training, some of the C-data state is hard to preserve. 2015-10-13 13:44:41 +11:00
Matthew Honnibal
f8de403483 * Work on pickling Vocab instances. The current implementation is not correct, but it may serve to see whether this approach is workable. Pickling is necessary to address Issue #125 2015-10-13 13:44:41 +11:00
Matthew Honnibal
85e7944572 * Start trying to pickle Vocab 2015-10-13 13:44:41 +11:00
Matthew Honnibal
41012907a8 * Fix variable name 2015-10-13 13:44:40 +11:00
Matthew Honnibal
37b909b6b6 * Use the symbols file in vocab instead of the symbols subfiles like attrs.pxd 2015-10-13 13:44:40 +11:00
Matthew Honnibal
d70e8cac2c * Fix empty values in attributes and parts of speech, so symbols align correctly with the StringStore 2015-10-13 13:44:40 +11:00
Matthew Honnibal
a29c8ee23d * Add symbols to the vocab before reading the strings, so that they line up correctly 2015-10-13 13:44:39 +11:00
Matthew Honnibal
85ce36ab11 * Refactor symbols, so that frequency rank can be derived from the orth id of a word. 2015-10-13 13:44:39 +11:00
Matthew Honnibal
83dccf0fd7 * Use io module insteads of deprecated codecs module 2015-10-10 14:13:01 +11:00
Matthew Honnibal
3d9f41c2c9 * Add LookupError for better error reporting in Vocab 2015-10-06 10:34:59 +11:00
alvations
8caedba42a caught more codecs.open -> io.open 2015-09-30 20:20:09 +02:00
Matthew Honnibal
abf0d930af * Fix API for loading word vectors from a file. 2015-09-23 23:51:08 +10:00
Matthew Honnibal
f7283a5067 * Fix vectors bugs for OOV words 2015-09-22 02:10:25 +02:00
Matthew Honnibal
ac459278d1 * Fix vector length error reporting, and ensure vec_len is returned 2015-09-21 18:08:32 +10:00
Matthew Honnibal
ba4e563701 * Ensure vectors are same length, and return vector length in load_vectors_bz2 2015-09-21 18:03:08 +10:00
Matthew Honnibal
d6945bf880 * Add way to load vectors from bz2 file to vocab 2015-09-17 12:58:23 +10:00
Matthew Honnibal
3d87519f64 * Remove vectors argument from Vocab object 2015-09-15 14:47:14 +10:00
Matthew Honnibal
27f988b167 * Remove the vectors option to Vocab, preferring to either load vectors from disk, or set them on the Lexeme objects. 2015-09-15 14:41:48 +10:00
Matthew Honnibal
e9c59693ea * Remove assertion from vocab.pyx 2015-09-13 10:30:08 +10:00
Matthew Honnibal
e1dfaeed8a * Check serializer freqs exist before loading 2015-09-12 23:49:38 +02:00
Matthew Honnibal
a412c66c8c * Check serializer freqs exist before loading 2015-09-12 23:40:01 +02:00
Matthew Honnibal
e285ca7d6c * Load serializer freqs in vocab 2015-09-10 15:22:48 +02:00
Matthew Honnibal
094440f9f5 Merge branch 'develop' of ssh://github.com/honnibal/spaCy into develop 2015-09-10 14:51:17 +02:00
Matthew Honnibal
90da3a695d * Load lemmatizer from disk in Vocab.from_dir 2015-09-10 14:49:10 +02:00
Matthew Honnibal
f634191e27 * Fix vocab read/write 2015-09-10 14:44:38 +02:00
Matthew Honnibal
a7f4b26c8c * Tmp 2015-09-09 14:33:26 +02:00
Matthew Honnibal
d6561988cf * Fix lexemes.bin 2015-09-09 11:49:51 +02:00
Matthew Honnibal
c301bebd33 Merge branch 'master' of https://github.com/honnibal/spaCy into develop 2015-09-09 10:55:39 +02:00
Matthew Honnibal
623329b19a Merge branch 'master' of ssh://github.com/honnibal/spaCy into develop 2015-09-08 14:27:01 +02:00
Matthew Honnibal
62a01dd41d * Fix issue #92: lexemes.bin read error on 32-bit platforms. 2015-09-08 14:23:58 +02:00
Matthew Honnibal
f6ec5bf1b0 * Use empty tag map in vocab if none supplied 2015-09-06 20:19:27 +02:00
Matthew Honnibal
534e3dda3c * More work on language independent parsing 2015-08-28 03:44:54 +02:00
Matthew Honnibal
c2307fa9ee * More work on language-generic parsing 2015-08-28 02:02:33 +02:00
Matthew Honnibal
1302d35dff * Rework interfaces in vocab 2015-08-26 19:21:46 +02:00
Matthew Honnibal
6f1743692a * Work on language-independent refactoring 2015-08-23 20:49:18 +02:00
Matthew Honnibal
cad0cca4e3 * Tmp 2015-08-22 22:04:34 +02:00
Matthew Honnibal
3d43f49f69 * Revert prev change 2015-07-27 10:58:15 +02:00
Matthew Honnibal
6b586cdad4 * Change lexemes.bin format. Add a header specifying size of LexemeC and number of lexemes, and don't have the redundant orth information. 2015-07-27 08:31:51 +02:00
Matthew Honnibal
8e4c69ee8c * Add is_oov property, and fix up handling of attributes 2015-07-27 01:50:06 +02:00
Matthew Honnibal
fc268f03eb * Assert against null pointer exceptions in vocab 2015-07-27 01:00:10 +02:00
Matthew Honnibal
0f093fdb30 * Fix get_by_orth for py3 2015-07-26 19:26:41 +02:00
Matthew Honnibal
ceeda5a739 * Fix get_by_orth for py3 2015-07-26 18:39:27 +02:00
Matthew Honnibal
6bb96c122d * Host IS_ flags in attrs.pxd, and add properties for them on Token and Lexeme objects 2015-07-26 16:37:16 +02:00
Matthew Honnibal
7eb2446082 * Return empty lexeme on empty string 2015-07-26 00:18:30 +02:00
Matthew Honnibal
fd525f0675 * Pass OOV probability around 2015-07-25 23:29:51 +02:00
Matthew Honnibal
22028602a9 * Add unicode_literals declaration in vocab.pyx 2015-07-23 13:24:20 +02:00
Matthew Honnibal
a7c4d72e83 * Add serializer property to Vocab, and lazy-load it. Add get_by_orth method. 2015-07-23 01:18:19 +02:00
Matthew Honnibal
109106a949 * Replace UniStr, using unicode objects instead 2015-07-22 04:52:05 +02:00
Matthew Honnibal
1f7170e0e1 * Reinstate the fixed vocabulary --- words are only added to the lexicon in init_model, after that we create LexemeC structs with the Pool given to us. 2015-07-20 01:37:34 +02:00
Matthew Honnibal
317cbbc015 * Serialization round trip now working with decent API, but with rough spots in the organisation and requiring vocabulary to be fixed ahead of time. 2015-07-19 15:18:17 +02:00
Matthew Honnibal
82d84b0f2b * Index lexemes by orth, instead of a lexemes vector. Breaks the mechanism for deciding not to own LexemeC structs during parsing. Need to reinstate this. 2015-07-18 22:42:15 +02:00
Matthew Honnibal
c2c83120d4 * Remove codec property from Vocab 2015-07-17 16:40:11 +02:00
Matthew Honnibal
db9dfd2e23 * Major refactor of serialization. Nearly complete now. 2015-07-17 01:27:54 +02:00
Matthew Honnibal
2a5d050134 * Give codec loading back to Vocab. 2015-07-16 17:45:42 +02:00
Matthew Honnibal
b59d271510 * Move serialization functionality into Serializer class 2015-07-16 11:23:48 +02:00
Matthew Honnibal
af5cc926a4 * Add codec property to Vocab, to use the Huffman encoding 2015-07-13 13:55:14 +02:00
Matthew Honnibal
abc43b852d * Add pos_tags attr to Vocab. 2015-07-08 12:36:38 +02:00
Matthew Honnibal
c04e6ebca6 * Allow user to load different sized vectors. 2015-06-05 16:26:39 +02:00
Matthew Honnibal
adeb57cb1e * Fix long line 2015-06-01 23:07:00 +02:00
Matthew Honnibal
eba7b34f66 * Add flag to disable loading of word vectors 2015-05-25 01:02:42 +02:00
Matthew Honnibal
e73eaf2d05 * Replace some assertions with proper errors 2015-05-08 16:52:17 +02:00