Commit Graph

  • 1cc6329b18 * Add base class to do transitions Matthew Honnibal 2015-02-18 04:43:21 -0500
  • 135756ac3d * Tmp commit of NER refactoring Matthew Honnibal 2015-02-18 04:41:06 -0500
  • 49df1b7002 * Ignore .tgz files Matthew Honnibal 2015-02-17 22:10:28 -0500
  • 8715101239 * Merge changes from stash Matthew Honnibal 2015-02-17 22:08:02 -0500
  • 23c1f6fc04 * Merge changes from stash Matthew Honnibal 2015-02-17 22:07:44 -0500
  • 0ff078876a * Commit some work on ner.yx done on the plane Matthew Honnibal 2015-02-04 14:16:57 -0500
  • d81b7be6a2 * Merge train.py Matthew Honnibal 2015-02-04 14:16:14 -0500
  • 3a302ae6f2 * Merge train.py Matthew Honnibal 2015-02-17 22:02:09 -0500
  • 2e3dc3dfe2 * Merge changes in tokens.pyx Matthew Honnibal 2015-02-17 21:59:37 -0500
  • 8cc3524dc9 * Ws Matthew Honnibal 2015-02-02 16:42:04 +1100
  • 3d0570685c * Add NER transition system Matthew Honnibal 2015-02-02 16:38:52 +1100
  • 043b758cf4 * Resurrect old NER code. This version won't be the one that runs; we want to re-use the parser code. But for now this is a useful reference. Matthew Honnibal 2015-02-02 16:37:25 +1100
  • b139aa92ba * Start setting out how NER will be implemented in the data model Matthew Honnibal 2015-02-02 16:35:58 +1100
  • 0962ffc095 * Fix issue #37: missing check_flag attribute from Token class Matthew Honnibal 2015-03-26 15:06:26 +0100
  • ccbb88951b Merge 5b8e8b4f72 into 5032f2a5c7 rsomeon 2015-03-25 15:00:49 +0000
  • 5032f2a5c7 * Fix nested lists Matthew Honnibal 2015-03-25 14:38:59 +0100
  • 03636be9da * Fix table md Matthew Honnibal 2015-03-25 14:36:12 +0100
  • 2a39e87891 * Fix table md Matthew Honnibal 2015-03-25 14:35:42 +0100
  • 9937f73075 * Fix table md Matthew Honnibal 2015-03-25 14:34:36 +0100
  • 22368706ce * Add CLA stuff Matthew Honnibal 2015-03-25 14:32:50 +0100
  • 46e936adfa * Fix quickstart Matthew Honnibal 2015-03-19 00:09:39 -0400
  • d345f53dbc * Add bootstrap script to install instructions Matthew Honnibal 2015-03-16 14:14:00 -0400
  • b924a4d642 * Add bootstrap script Matthew Honnibal 2015-03-16 14:01:36 -0400
  • fc1d1456de Fix import typo in docs Viksit Gaur 2015-03-12 13:13:42 -0700
  • 5b8e8b4f72 Fix lemma_ unicode handling Someon 2015-03-11 01:26:59 +0200
  • ec25976416 Test lemma_ with unicode input Someon 2015-03-11 01:23:06 +0200
  • 2e8d0e5d45 * Upd download script Matthew Honnibal 2015-03-03 05:47:16 -0500
  • c341bfb0a2 * Inc version Matthew Honnibal 2015-03-03 05:46:14 -0500
  • a61dacb4e5 * Add tests for new subtree method Matthew Honnibal 2015-03-03 05:41:00 -0500
  • 053814ffc8 * Report LAS in train script Matthew Honnibal 2015-03-03 04:35:11 -0500
  • b07632a9ef * Upd docs, improving description of parse tree navigation Matthew Honnibal 2015-03-03 04:34:07 -0500
  • dbe26f5793 * Add children and subtree methods to Token, which are generators to assist parse-tree navigation. Matthew Honnibal 2015-03-03 04:18:41 -0500
  • 827a2337b0 * Inc version Matthew Honnibal 2015-02-27 03:56:54 -0500
  • ea90d136e8 * Fix bug in labelled parsing, that caused an 8% drop in labelled accuracy. Matthew Honnibal 2015-02-27 03:56:10 -0500
  • 5e27bd0c4c * Add en language data, for tokenizer etc Matthew Honnibal 2015-02-25 17:10:32 -0500
  • 1019939c7a * Whitespace Matthew Honnibal 2015-02-24 23:03:02 -0500
  • 74015da94b * Inc version Matthew Honnibal 2015-02-23 15:40:41 -0500
  • caf046b220 * Hastily add method to apply tags from a list of strings, instead of predicting the tags. Matthew Honnibal 2015-02-23 15:40:17 -0500
  • 6102360111 * Add -Wno-strict-prototypes, to suppress warning Matthew Honnibal 2015-02-21 20:04:37 -0500
  • 47a4371fea * Upd tokenizer with i.e. tests Matthew Honnibal 2015-02-18 06:37:04 -0500
  • ba1d3ddd7f * Move -lc++ link arg to only be used if darwin is OS. Should actually check whether GCC is compiler Matthew Honnibal 2015-02-18 06:10:23 -0500
  • 59b46e4c2f * Move libc++ argument back under check for darwin. This assumes that extensions on OSX will be built with clang, but OSX GCC builds are also possible. Need to detect compiler and disable this flag Matthew Honnibal 2015-02-18 06:03:45 -0500
  • aa475673ee * Tweak compile args for OSX Matthew Honnibal 2015-02-18 05:41:11 -0500
  • b4edd1d907 * Make new compile args conditional on darwin, as they're invalid on Linux Matthew Honnibal 2015-02-18 05:09:50 -0500
  • e885903dc6 * Add compile args to fix conda compilation on OSX, and increment version Matthew Honnibal 2015-02-18 05:01:27 -0500
  • 69d27d55b0 * Inc version, with new orphan-token bug fix Matthew Honnibal 2015-02-16 16:52:54 -0500
  • cae077b583 * Work on fixing orphaned Token objects bug Matthew Honnibal 2015-02-16 15:20:31 -0500
  • 789a6fe462 * Inc version --- 0.63 seems to have been packaged incorrectly, to not include a bug fix to tokens.pyx to transfer ownership to Token objects Matthew Honnibal 2015-02-16 11:56:14 -0500
  • 9dbc31d72c * Add test from NSchrading Matthew Honnibal 2015-02-16 11:49:31 -0500
  • 274b802830 * Fix docs bug Matthew Honnibal 2015-02-11 20:07:39 -0500
  • 773d209405 * Inc version to 0.63 Matthew Honnibal 2015-02-11 18:39:41 -0500
  • cd6367e404 * Fix cosine function in documentation Matthew Honnibal 2015-02-11 18:08:19 -0500
  • 7572e31f5e * Pass ownership of C data to Token instances if Tokens object is being garbage-collected, but Token instances are staying alive. Matthew Honnibal 2015-02-11 18:05:06 -0500
  • db3f26a51b * Remove version note Matthew Honnibal 2015-02-11 18:03:23 -0500
  • 4258b1490a * Improve API docs for Token Matthew Honnibal 2015-02-11 18:03:06 -0500
  • 64645a1c2f * Improve docstring on English Matthew Honnibal 2015-02-11 15:13:20 -0500
  • f0a9d2cb9c * Inc version Matthew Honnibal 2015-02-11 14:20:57 -0500
  • 594e50bd45 * Add option to download speech-parsing data set. Matthew Honnibal 2015-02-11 14:20:29 -0500
  • 0b7e769211 * Add POS tags to support SWBD tag set Matthew Honnibal 2015-02-11 14:08:28 -0500
  • e425de6d2b Merge branch 'develop' of ssh://github.com/honnibal/spaCy into develop Matthew Honnibal 2015-02-10 10:16:24 -0500
  • 5ff2b5c8f0 * Inc version Matthew Honnibal 2015-02-10 10:16:09 -0500
  • 312b3a45f3 * Fix issue #19: Allow parsing/pos tagging of empty strings Matthew Honnibal 2015-02-10 10:15:58 -0500
  • 363473aeed Add rokenizer test for zero length string leofidus 2015-02-07 03:01:44 +0100
  • ae36067314 Merge pull request #21 from leofidus/test_notoken honnibal 2015-02-11 00:19:38 +1100
  • 2a0615104b * Upd download script Matthew Honnibal 2015-02-09 10:22:59 -0500
  • 29bdf0d05a * Inc version Matthew Honnibal 2015-02-09 10:21:51 -0500
  • 407bb5da8b * Increment version Matthew Honnibal 2015-02-09 09:46:20 -0500
  • ee33be31dd * Fix parser training script Matthew Honnibal 2015-02-09 03:57:56 -0500
  • 5c3513583d * Clear buffered python tokens when modifying the Tokens object. Need to clean this up, and modify via a method on Tokens. Matthew Honnibal 2015-02-09 03:56:51 -0500
  • be5536d239 * Fix Issue #22: PRP and PRP$ were mapped to NOUN. Should be PRON. Matthew Honnibal 2015-02-08 18:36:18 -0500
  • 99f0a315f9 * Add test for Issue 24 Matthew Honnibal 2015-02-08 18:30:46 -0500
  • 0492cee8b4 * Fix Issue #24: Lemmas are empty when the L field is missing for special-cased tokens Matthew Honnibal 2015-02-08 18:30:30 -0500
  • 3e8c87af1a * Extend parse tree navigation tests Matthew Honnibal 2015-02-07 18:28:45 -0500
  • 933c188eb5 * Inc version Matthew Honnibal 2015-02-07 13:14:27 -0500
  • aadc57ab00 * Add tests for tokens api Matthew Honnibal 2015-02-07 13:14:07 -0500
  • d229fbd228 * Give better error on out-of-bounds array access Matthew Honnibal 2015-02-07 12:59:12 -0500
  • ab8bb047d0 * Fix negative index for __getitem__ Matthew Honnibal 2015-02-07 12:58:46 -0500
  • ef795aece8 * Upd release Matthew Honnibal 2015-02-07 12:26:34 -0500
  • 44c7eafe44 * Fix download.py Matthew Honnibal 2015-02-07 12:00:36 -0500
  • b6c8624b82 Merge branch 'master' of ssh://github.com/honnibal/spaCy Matthew Honnibal 2015-02-07 11:53:13 -0500
  • 6ca7f2eedc * Upd download script Matthew Honnibal 2015-02-07 11:32:33 -0500
  • 330b1a7a3d * Inc version Matthew Honnibal 2015-02-07 11:32:13 -0500
  • 6b68607b1f * Add some tests for the code in the index.html docstrings Matthew Honnibal 2015-02-07 08:52:13 -0500
  • a7e4f0a86c * Make corrections to example code Matthew Honnibal 2015-02-07 08:45:09 -0500
  • f0e0588833 * Fill L2 norm attribute on LexemeC struct Matthew Honnibal 2015-02-07 08:44:42 -0500
  • 75f9b7d6bf * Add L2 norm field to LexemeC struct Matthew Honnibal 2015-02-07 08:43:17 -0500
  • 51b618d646 * Add a has_repvec property to Lexeme, and a check function to check flags Matthew Honnibal 2015-02-07 08:42:44 -0500
  • 321b402739 * Store the l2 norm of the word's vector Matthew Honnibal 2015-02-07 08:42:16 -0500
  • 0ae05f77ab Add rokenizer test for zero length string leofidus 2015-02-07 03:01:44 +0100
  • c7d8644149 * Fix regression on 'prob' attr of Token. Matthew Honnibal 2015-02-03 03:32:18 +1100
  • 27986d7f5c * Fix standard conll file reading. Script needs refactoring. Matthew Honnibal 2015-02-02 23:02:48 +1100
  • c55a33d045 * Catch oracle errors Matthew Honnibal 2015-02-02 23:02:04 +1100
  • de772088e6 * Use parse tree for sbd in Tokens.sents Matthew Honnibal 2015-02-02 12:17:32 +1100
  • ba1e91189b * Fix 0.40 link in index Matthew Honnibal 2015-02-02 12:16:53 +1100
  • 56c2ef2982 * Tweak POS features for web text Matthew Honnibal 2015-02-02 11:59:36 +1100
  • d68678a93e * Add Exception class, OracleError Matthew Honnibal 2015-02-02 11:57:32 +1100
  • d7df4aeb1d Update README.md honnibal 2015-02-01 18:38:22 +1100
  • 5199af89a8 Update README.md honnibal 2015-02-01 18:37:48 +1100
  • 57a7fabc7a Update README.md honnibal 2015-02-01 18:36:52 +1100
  • bfe1bcc02d * Rename 0.4.0 to 0.40 Matthew Honnibal 2015-02-01 18:32:01 +1100