nltk 3.2__zip → 3.3__zip

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (406) hide show
  1. {nltk-3.2 → nltk-3.3}/INSTALL.txt +6 -6
  2. {nltk-3.2 → nltk-3.3}/LICENSE.txt +13 -13
  3. {nltk-3.2 → nltk-3.3}/MANIFEST.in +6 -6
  4. {nltk-3.2/nltk.egg-info → nltk-3.3}/PKG-INFO +13 -5
  5. nltk-3.2/README.txt → nltk-3.3/README.md +49 -48
  6. nltk-3.3/nltk/VERSION +1 -0
  7. {nltk-3.2 → nltk-3.3}/nltk/__init__.py +185 -185
  8. {nltk-3.2 → nltk-3.3}/nltk/app/__init__.py +52 -53
  9. {nltk-3.2 → nltk-3.3}/nltk/app/chartparser_app.py +2276 -2276
  10. {nltk-3.2 → nltk-3.3}/nltk/app/chunkparser_app.py +1262 -1264
  11. {nltk-3.2 → nltk-3.3}/nltk/app/collocations_app.py +347 -350
  12. {nltk-3.2 → nltk-3.3}/nltk/app/concordance_app.py +568 -570
  13. {nltk-3.2 → nltk-3.3}/nltk/app/nemo_app.py +156 -155
  14. {nltk-3.2 → nltk-3.3}/nltk/app/rdparser_app.py +892 -894
  15. {nltk-3.2 → nltk-3.3}/nltk/app/srparser_app.py +809 -809
  16. {nltk-3.2 → nltk-3.3}/nltk/app/wordfreq_app.py +32 -32
  17. {nltk-3.2 → nltk-3.3}/nltk/app/wordnet_app.py +968 -968
  18. {nltk-3.2 → nltk-3.3}/nltk/book.py +91 -90
  19. {nltk-3.2 → nltk-3.3}/nltk/ccg/__init__.py +22 -22
  20. {nltk-3.2 → nltk-3.3}/nltk/ccg/api.py +359 -335
  21. {nltk-3.2 → nltk-3.3}/nltk/ccg/chart.py +394 -361
  22. {nltk-3.2 → nltk-3.3}/nltk/ccg/combinator.py +343 -312
  23. {nltk-3.2 → nltk-3.3}/nltk/ccg/lexicon.py +328 -290
  24. nltk-3.3/nltk/ccg/logic.py +46 -0
  25. {nltk-3.2 → nltk-3.3}/nltk/chat/__init__.py +49 -49
  26. {nltk-3.2 → nltk-3.3}/nltk/chat/eliza.py +244 -244
  27. {nltk-3.2 → nltk-3.3}/nltk/chat/iesha.py +140 -140
  28. {nltk-3.2 → nltk-3.3}/nltk/chat/rude.py +92 -92
  29. {nltk-3.2 → nltk-3.3}/nltk/chat/suntsu.py +117 -117
  30. {nltk-3.2 → nltk-3.3}/nltk/chat/util.py +122 -120
  31. {nltk-3.2 → nltk-3.3}/nltk/chat/zen.py +282 -282
  32. {nltk-3.2 → nltk-3.3}/nltk/chunk/__init__.py +190 -190
  33. {nltk-3.2 → nltk-3.3}/nltk/chunk/api.py +51 -51
  34. {nltk-3.2 → nltk-3.3}/nltk/chunk/named_entity.py +331 -331
  35. {nltk-3.2 → nltk-3.3}/nltk/chunk/regexp.py +1390 -1384
  36. {nltk-3.2 → nltk-3.3}/nltk/chunk/util.py +600 -600
  37. {nltk-3.2 → nltk-3.3}/nltk/classify/__init__.py +98 -98
  38. {nltk-3.2 → nltk-3.3}/nltk/classify/api.py +193 -193
  39. {nltk-3.2 → nltk-3.3}/nltk/classify/decisiontree.py +295 -295
  40. {nltk-3.2 → nltk-3.3}/nltk/classify/maxent.py +1510 -1492
  41. {nltk-3.2 → nltk-3.3}/nltk/classify/megam.py +179 -178
  42. {nltk-3.2 → nltk-3.3}/nltk/classify/naivebayes.py +245 -242
  43. {nltk-3.2 → nltk-3.3}/nltk/classify/positivenaivebayes.py +170 -170
  44. {nltk-3.2 → nltk-3.3}/nltk/classify/rte_classify.py +160 -191
  45. {nltk-3.2 → nltk-3.3}/nltk/classify/scikitlearn.py +153 -151
  46. {nltk-3.2 → nltk-3.3}/nltk/classify/senna.py +183 -181
  47. {nltk-3.2 → nltk-3.3}/nltk/classify/svm.py +15 -15
  48. {nltk-3.2 → nltk-3.3}/nltk/classify/tadm.py +113 -112
  49. {nltk-3.2 → nltk-3.3}/nltk/classify/textcat.py +193 -193
  50. {nltk-3.2 → nltk-3.3}/nltk/classify/util.py +324 -312
  51. {nltk-3.2 → nltk-3.3}/nltk/classify/weka.py +346 -345
  52. {nltk-3.2 → nltk-3.3}/nltk/cluster/__init__.py +86 -86
  53. {nltk-3.2 → nltk-3.3}/nltk/cluster/api.py +74 -70
  54. {nltk-3.2 → nltk-3.3}/nltk/cluster/em.py +247 -247
  55. {nltk-3.2 → nltk-3.3}/nltk/cluster/gaac.py +168 -168
  56. {nltk-3.2 → nltk-3.3}/nltk/cluster/kmeans.py +221 -221
  57. {nltk-3.2 → nltk-3.3}/nltk/cluster/util.py +304 -289
  58. nltk-3.3/nltk/collections.py +640 -0
  59. {nltk-3.2 → nltk-3.3}/nltk/collocations.py +388 -388
  60. nltk-3.3/nltk/compat.py +365 -0
  61. {nltk-3.2 → nltk-3.3}/nltk/corpus/__init__.py +341 -320
  62. {nltk-3.2 → nltk-3.3}/nltk/corpus/europarl_raw.py +44 -44
  63. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/__init__.py +147 -145
  64. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/aligned.py +115 -114
  65. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/api.py +448 -446
  66. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/bnc.py +252 -252
  67. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/bracket_parse.py +227 -227
  68. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/categorized_sents.py +179 -178
  69. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/chasen.py +140 -139
  70. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/childes.py +522 -494
  71. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/chunked.py +212 -212
  72. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/cmudict.py +97 -95
  73. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/comparative_sents.py +280 -278
  74. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/conll.py +528 -521
  75. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/crubadan.py +116 -116
  76. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/dependency.py +101 -101
  77. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/framenet.py +3059 -2093
  78. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ieer.py +112 -111
  79. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/indian.py +85 -86
  80. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ipipan.py +332 -331
  81. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/knbc.py +187 -187
  82. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/lin.py +156 -156
  83. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/mte.py +311 -310
  84. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nkjp.py +429 -428
  85. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nombank.py +421 -421
  86. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/nps_chat.py +73 -73
  87. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/opinion_lexicon.py +115 -115
  88. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/panlex_lite.py +165 -165
  89. nltk-3.3/nltk/corpus/reader/pl196x.py +292 -0
  90. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/plaintext.py +232 -233
  91. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ppattach.py +96 -95
  92. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/propbank.py +479 -481
  93. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/pros_cons.py +128 -126
  94. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/reviews.py +330 -328
  95. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/rte.py +144 -146
  96. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/semcor.py +256 -256
  97. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/senseval.py +203 -201
  98. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/sentiwordnet.py +137 -137
  99. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/sinica_treebank.py +75 -75
  100. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/string_category.py +60 -59
  101. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/switchboard.py +119 -119
  102. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/tagged.py +295 -294
  103. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/timit.py +452 -450
  104. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/toolbox.py +68 -68
  105. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/twitter.py +155 -157
  106. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/udhr.py +80 -80
  107. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/util.py +804 -800
  108. nltk-3.3/nltk/corpus/reader/verbnet.py +595 -0
  109. nltk-3.3/nltk/corpus/reader/wordlist.py +134 -0
  110. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/wordnet.py +2108 -2036
  111. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/xmldocs.py +392 -387
  112. {nltk-3.2 → nltk-3.3}/nltk/corpus/reader/ycoe.py +243 -242
  113. {nltk-3.2 → nltk-3.3}/nltk/corpus/util.py +144 -127
  114. {nltk-3.2 → nltk-3.3}/nltk/data.py +1487 -1451
  115. {nltk-3.2 → nltk-3.3}/nltk/decorators.py +219 -215
  116. {nltk-3.2 → nltk-3.3}/nltk/downloader.py +2278 -2274
  117. {nltk-3.2 → nltk-3.3}/nltk/draw/__init__.py +27 -27
  118. {nltk-3.2 → nltk-3.3}/nltk/draw/cfg.py +774 -777
  119. {nltk-3.2 → nltk-3.3}/nltk/draw/dispersion.py +58 -58
  120. {nltk-3.2 → nltk-3.3}/nltk/draw/table.py +1098 -1098
  121. {nltk-3.2 → nltk-3.3}/nltk/draw/tree.py +963 -966
  122. {nltk-3.2 → nltk-3.3}/nltk/draw/util.py +2356 -2356
  123. {nltk-3.2 → nltk-3.3}/nltk/featstruct.py +2617 -2500
  124. {nltk-3.2 → nltk-3.3}/nltk/grammar.py +1559 -1516
  125. {nltk-3.2 → nltk-3.3}/nltk/help.py +56 -56
  126. {nltk-3.2 → nltk-3.3}/nltk/inference/__init__.py +20 -20
  127. {nltk-3.2 → nltk-3.3}/nltk/inference/api.py +597 -589
  128. {nltk-3.2 → nltk-3.3}/nltk/inference/discourse.py +612 -608
  129. {nltk-3.2 → nltk-3.3}/nltk/inference/mace.py +311 -311
  130. {nltk-3.2 → nltk-3.3}/nltk/inference/nonmonotonic.py +509 -509
  131. {nltk-3.2 → nltk-3.3}/nltk/inference/prover9.py +431 -431
  132. {nltk-3.2 → nltk-3.3}/nltk/inference/resolution.py +687 -687
  133. {nltk-3.2 → nltk-3.3}/nltk/inference/tableau.py +607 -607
  134. {nltk-3.2 → nltk-3.3}/nltk/internals.py +982 -982
  135. {nltk-3.2 → nltk-3.3}/nltk/jsontags.py +63 -63
  136. {nltk-3.2 → nltk-3.3}/nltk/lazyimport.py +142 -142
  137. {nltk-3.2 → nltk-3.3}/nltk/metrics/__init__.py +30 -29
  138. {nltk-3.2 → nltk-3.3}/nltk/metrics/agreement.py +421 -401
  139. nltk-3.3/nltk/metrics/aline.py +607 -0
  140. {nltk-3.2 → nltk-3.3}/nltk/metrics/association.py +414 -410
  141. {nltk-3.2 → nltk-3.3}/nltk/metrics/confusionmatrix.py +206 -206
  142. {nltk-3.2 → nltk-3.3}/nltk/metrics/distance.py +207 -202
  143. {nltk-3.2 → nltk-3.3}/nltk/metrics/paice.py +381 -381
  144. {nltk-3.2 → nltk-3.3}/nltk/metrics/scores.py +228 -228
  145. {nltk-3.2 → nltk-3.3}/nltk/metrics/segmentation.py +231 -232
  146. {nltk-3.2 → nltk-3.3}/nltk/metrics/spearman.py +69 -69
  147. {nltk-3.2 → nltk-3.3}/nltk/misc/__init__.py +11 -11
  148. {nltk-3.2 → nltk-3.3}/nltk/misc/babelfish.py +10 -10
  149. {nltk-3.2 → nltk-3.3}/nltk/misc/chomsky.py +135 -133
  150. {nltk-3.2 → nltk-3.3}/nltk/misc/minimalset.py +83 -83
  151. {nltk-3.2 → nltk-3.3}/nltk/misc/sort.py +157 -157
  152. {nltk-3.2 → nltk-3.3}/nltk/misc/wordfinder.py +129 -129
  153. {nltk-3.2 → nltk-3.3}/nltk/parse/__init__.py +81 -80
  154. {nltk-3.2 → nltk-3.3}/nltk/parse/api.py +66 -66
  155. {nltk-3.2 → nltk-3.3}/nltk/parse/bllip.py +282 -282
  156. {nltk-3.2 → nltk-3.3}/nltk/parse/chart.py +1682 -1681
  157. nltk-3.3/nltk/parse/corenlp.py +780 -0
  158. {nltk-3.2 → nltk-3.3}/nltk/parse/dependencygraph.py +760 -758
  159. {nltk-3.2 → nltk-3.3}/nltk/parse/earleychart.py +452 -451
  160. {nltk-3.2 → nltk-3.3}/nltk/parse/evaluate.py +129 -129
  161. {nltk-3.2 → nltk-3.3}/nltk/parse/featurechart.py +580 -578
  162. {nltk-3.2 → nltk-3.3}/nltk/parse/generate.py +87 -76
  163. {nltk-3.2 → nltk-3.3}/nltk/parse/malt.py +356 -356
  164. {nltk-3.2 → nltk-3.3}/nltk/parse/nonprojectivedependencyparser.py +783 -783
  165. {nltk-3.2 → nltk-3.3}/nltk/parse/pchart.py +522 -522
  166. {nltk-3.2 → nltk-3.3}/nltk/parse/projectivedependencyparser.py +579 -578
  167. {nltk-3.2 → nltk-3.3}/nltk/parse/recursivedescent.py +655 -656
  168. {nltk-3.2 → nltk-3.3}/nltk/parse/shiftreduce.py +458 -459
  169. {nltk-3.2 → nltk-3.3}/nltk/parse/stanford.py +431 -406
  170. {nltk-3.2 → nltk-3.3}/nltk/parse/transitionparser.py +774 -772
  171. {nltk-3.2 → nltk-3.3}/nltk/parse/util.py +231 -231
  172. {nltk-3.2 → nltk-3.3}/nltk/parse/viterbi.py +401 -401
  173. {nltk-3.2 → nltk-3.3}/nltk/probability.py +2410 -2355
  174. {nltk-3.2 → nltk-3.3}/nltk/sem/__init__.py +61 -61
  175. {nltk-3.2 → nltk-3.3}/nltk/sem/boxer.py +1261 -1261
  176. {nltk-3.2 → nltk-3.3}/nltk/sem/chat80.py +782 -782
  177. {nltk-3.2 → nltk-3.3}/nltk/sem/cooper_storage.py +118 -118
  178. {nltk-3.2 → nltk-3.3}/nltk/sem/drt.py +1258 -1257
  179. {nltk-3.2 → nltk-3.3}/nltk/sem/drt_glue_demo.py +485 -489
  180. {nltk-3.2 → nltk-3.3}/nltk/sem/evaluate.py +790 -784
  181. {nltk-3.2 → nltk-3.3}/nltk/sem/glue.py +696 -694
  182. {nltk-3.2 → nltk-3.3}/nltk/sem/hole.py +386 -384
  183. {nltk-3.2 → nltk-3.3}/nltk/sem/lfg.py +210 -208
  184. {nltk-3.2 → nltk-3.3}/nltk/sem/linearlogic.py +451 -449
  185. {nltk-3.2 → nltk-3.3}/nltk/sem/logic.py +1910 -1910
  186. {nltk-3.2 → nltk-3.3}/nltk/sem/relextract.py +476 -483
  187. {nltk-3.2 → nltk-3.3}/nltk/sem/skolemize.py +101 -101
  188. {nltk-3.2 → nltk-3.3}/nltk/sem/util.py +249 -249
  189. {nltk-3.2 → nltk-3.3}/nltk/sentiment/__init__.py +14 -14
  190. {nltk-3.2 → nltk-3.3}/nltk/sentiment/sentiment_analyzer.py +229 -229
  191. {nltk-3.2 → nltk-3.3}/nltk/sentiment/util.py +762 -752
  192. {nltk-3.2 → nltk-3.3}/nltk/sentiment/vader.py +443 -445
  193. {nltk-3.2 → nltk-3.3}/nltk/stem/__init__.py +31 -33
  194. {nltk-3.2 → nltk-3.3}/nltk/stem/api.py +28 -25
  195. nltk-3.3/nltk/stem/arlstem.py +369 -0
  196. {nltk-3.2 → nltk-3.3}/nltk/stem/isri.py +345 -345
  197. {nltk-3.2 → nltk-3.3}/nltk/stem/lancaster.py +337 -310
  198. nltk-3.3/nltk/stem/porter.py +710 -0
  199. {nltk-3.2 → nltk-3.3}/nltk/stem/regexp.py +61 -61
  200. {nltk-3.2 → nltk-3.3}/nltk/stem/rslp.py +142 -142
  201. {nltk-3.2 → nltk-3.3}/nltk/stem/snowball.py +4236 -3712
  202. {nltk-3.2 → nltk-3.3}/nltk/stem/util.py +22 -12
  203. {nltk-3.2 → nltk-3.3}/nltk/stem/wordnet.py +51 -51
  204. {nltk-3.2 → nltk-3.3}/nltk/tag/__init__.py +152 -127
  205. {nltk-3.2 → nltk-3.3}/nltk/tag/api.py +86 -81
  206. {nltk-3.2 → nltk-3.3}/nltk/tag/brill.py +424 -426
  207. {nltk-3.2 → nltk-3.3}/nltk/tag/brill_trainer.py +608 -608
  208. {nltk-3.2 → nltk-3.3}/nltk/tag/crf.py +204 -200
  209. {nltk-3.2 → nltk-3.3}/nltk/tag/hmm.py +1277 -1277
  210. {nltk-3.2 → nltk-3.3}/nltk/tag/hunpos.py +132 -132
  211. {nltk-3.2 → nltk-3.3}/nltk/tag/mapping.py +101 -101
  212. {nltk-3.2 → nltk-3.3}/nltk/tag/perceptron.py +329 -314
  213. {nltk-3.2 → nltk-3.3}/nltk/tag/senna.py +147 -145
  214. {nltk-3.2 → nltk-3.3}/nltk/tag/sequential.py +740 -736
  215. {nltk-3.2 → nltk-3.3}/nltk/tag/stanford.py +219 -196
  216. {nltk-3.2 → nltk-3.3}/nltk/tag/tnt.py +595 -604
  217. {nltk-3.2 → nltk-3.3}/nltk/tag/util.py +72 -72
  218. {nltk-3.2 → nltk-3.3}/nltk/tbl/__init__.py +28 -28
  219. {nltk-3.2 → nltk-3.3}/nltk/tbl/api.py +1 -1
  220. {nltk-3.2 → nltk-3.3}/nltk/tbl/demo.py +366 -366
  221. {nltk-3.2 → nltk-3.3}/nltk/tbl/erroranalysis.py +39 -39
  222. {nltk-3.2 → nltk-3.3}/nltk/tbl/feature.py +259 -261
  223. {nltk-3.2 → nltk-3.3}/nltk/tbl/rule.py +321 -321
  224. {nltk-3.2 → nltk-3.3}/nltk/tbl/template.py +303 -308
  225. {nltk-3.2 → nltk-3.3}/nltk/test/__init__.py +18 -18
  226. {nltk-3.2 → nltk-3.3}/nltk/test/all.py +24 -23
  227. {nltk-3.2 → nltk-3.3}/nltk/test/bleu.doctest +14 -14
  228. {nltk-3.2 → nltk-3.3}/nltk/test/bnc.doctest +60 -60
  229. {nltk-3.2 → nltk-3.3}/nltk/test/ccg.doctest +376 -395
  230. nltk-3.3/nltk/test/ccg_semantics.doctest +553 -0
  231. {nltk-3.2 → nltk-3.3}/nltk/test/chat80.doctest +234 -234
  232. {nltk-3.2 → nltk-3.3}/nltk/test/childes.doctest +184 -184
  233. {nltk-3.2 → nltk-3.3}/nltk/test/childes_fixt.py +14 -13
  234. {nltk-3.2 → nltk-3.3}/nltk/test/chunk.doctest +373 -373
  235. {nltk-3.2 → nltk-3.3}/nltk/test/classify.doctest +185 -185
  236. {nltk-3.2 → nltk-3.3}/nltk/test/classify_fixt.py +11 -10
  237. nltk-3.3/nltk/test/collections.doctest +20 -0
  238. {nltk-3.2 → nltk-3.3}/nltk/test/collocations.doctest +276 -276
  239. {nltk-3.2 → nltk-3.3}/nltk/test/compat.doctest +134 -138
  240. {nltk-3.2 → nltk-3.3}/nltk/test/compat_fixt.py +9 -8
  241. nltk-3.3/nltk/test/concordance.doctest +55 -0
  242. {nltk-3.2 → nltk-3.3}/nltk/test/corpus.doctest +2200 -2212
  243. {nltk-3.2 → nltk-3.3}/nltk/test/corpus_fixt.py +3 -3
  244. {nltk-3.2 → nltk-3.3}/nltk/test/crubadan.doctest +65 -65
  245. {nltk-3.2 → nltk-3.3}/nltk/test/data.doctest +379 -379
  246. {nltk-3.2 → nltk-3.3}/nltk/test/dependency.doctest +241 -241
  247. {nltk-3.2 → nltk-3.3}/nltk/test/discourse.doctest +546 -546
  248. {nltk-3.2 → nltk-3.3}/nltk/test/discourse_fixt.py +14 -13
  249. {nltk-3.2 → nltk-3.3}/nltk/test/doctest_nose_plugin.py +159 -156
  250. {nltk-3.2 → nltk-3.3}/nltk/test/drt.doctest +517 -517
  251. {nltk-3.2 → nltk-3.3}/nltk/test/featgram.doctest +607 -607
  252. {nltk-3.2 → nltk-3.3}/nltk/test/featstruct.doctest +1229 -1230
  253. {nltk-3.2 → nltk-3.3}/nltk/test/framenet.doctest +288 -239
  254. {nltk-3.2 → nltk-3.3}/nltk/test/generate.doctest +67 -67
  255. {nltk-3.2 → nltk-3.3}/nltk/test/gensim.doctest +140 -141
  256. {nltk-3.2 → nltk-3.3}/nltk/test/gensim_fixt.py +10 -9
  257. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics.doctest +384 -386
  258. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics_malt.doctest +68 -68
  259. {nltk-3.2 → nltk-3.3}/nltk/test/gluesemantics_malt_fixt.py +12 -11
  260. {nltk-3.2 → nltk-3.3}/nltk/test/grammar.doctest +48 -48
  261. {nltk-3.2 → nltk-3.3}/nltk/test/grammartestsuites.doctest +109 -109
  262. {nltk-3.2 → nltk-3.3}/nltk/test/index.doctest +100 -100
  263. {nltk-3.2 → nltk-3.3}/nltk/test/inference.doctest +534 -534
  264. {nltk-3.2 → nltk-3.3}/nltk/test/inference_fixt.py +12 -11
  265. {nltk-3.2 → nltk-3.3}/nltk/test/internals.doctest +140 -140
  266. {nltk-3.2 → nltk-3.3}/nltk/test/japanese.doctest +48 -48
  267. {nltk-3.2 → nltk-3.3}/nltk/test/logic.doctest +1098 -1098
  268. {nltk-3.2 → nltk-3.3}/nltk/test/metrics.doctest +270 -270
  269. {nltk-3.2 → nltk-3.3}/nltk/test/misc.doctest +118 -118
  270. {nltk-3.2 → nltk-3.3}/nltk/test/nonmonotonic.doctest +286 -286
  271. {nltk-3.2 → nltk-3.3}/nltk/test/nonmonotonic_fixt.py +12 -11
  272. {nltk-3.2 → nltk-3.3}/nltk/test/paice.doctest +35 -35
  273. {nltk-3.2 → nltk-3.3}/nltk/test/parse.doctest +884 -884
  274. {nltk-3.2 → nltk-3.3}/nltk/test/portuguese_en.doctest +565 -565
  275. {nltk-3.2 → nltk-3.3}/nltk/test/portuguese_en_fixt.py +16 -16
  276. {nltk-3.2 → nltk-3.3}/nltk/test/probability.doctest +304 -304
  277. {nltk-3.2 → nltk-3.3}/nltk/test/probability_fixt.py +13 -12
  278. {nltk-3.2 → nltk-3.3}/nltk/test/propbank.doctest +176 -176
  279. {nltk-3.2 → nltk-3.3}/nltk/test/relextract.doctest +263 -263
  280. {nltk-3.2 → nltk-3.3}/nltk/test/resolution.doctest +221 -221
  281. {nltk-3.2 → nltk-3.3}/nltk/test/runtests.py +86 -64
  282. {nltk-3.2 → nltk-3.3}/nltk/test/segmentation_fixt.py +10 -10
  283. {nltk-3.2 → nltk-3.3}/nltk/test/semantics.doctest +665 -665
  284. {nltk-3.2 → nltk-3.3}/nltk/test/semantics_fixt.py +7 -7
  285. {nltk-3.2 → nltk-3.3}/nltk/test/sentiment.doctest +233 -233
  286. {nltk-3.2 → nltk-3.3}/nltk/test/sentiwordnet.doctest +41 -41
  287. {nltk-3.2 → nltk-3.3}/nltk/test/simple.doctest +85 -85
  288. {nltk-3.2 → nltk-3.3}/nltk/test/stem.doctest +78 -78
  289. {nltk-3.2 → nltk-3.3}/nltk/test/tag.doctest +33 -33
  290. {nltk-3.2 → nltk-3.3}/nltk/test/tokenize.doctest +273 -166
  291. {nltk-3.2 → nltk-3.3}/nltk/test/toolbox.doctest +307 -307
  292. {nltk-3.2 → nltk-3.3}/nltk/test/translate.doctest +242 -242
  293. {nltk-3.2 → nltk-3.3}/nltk/test/translate_fixt.py +3 -3
  294. {nltk-3.2 → nltk-3.3}/nltk/test/tree.doctest +1101 -1101
  295. {nltk-3.2 → nltk-3.3}/nltk/test/treeprettyprinter.doctest +127 -127
  296. {nltk-3.2 → nltk-3.3}/nltk/test/treetransforms.doctest +156 -156
  297. nltk-3.3/nltk/test/unit/test_2x_compat.py +63 -0
  298. nltk-3.3/nltk/test/unit/test_aline.py +52 -0
  299. nltk-3.3/nltk/test/unit/test_chunk.py +56 -0
  300. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_classify.py +55 -53
  301. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_collocations.py +88 -79
  302. nltk-3.3/nltk/test/unit/test_concordance.py +98 -0
  303. nltk-3.3/nltk/test/unit/test_corenlp.py +1000 -0
  304. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_corpora.py +200 -183
  305. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_corpus_views.py +45 -45
  306. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_hmm.py +88 -88
  307. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_json2csv_corpus.py +188 -187
  308. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_naivebayes.py +26 -26
  309. nltk-3.3/nltk/test/unit/test_rte_classify.py +86 -0
  310. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_seekable_unicode_stream_reader.py +124 -124
  311. nltk-3.3/nltk/test/unit/test_senna.py +74 -0
  312. nltk-3.3/nltk/test/unit/test_stem.py +121 -0
  313. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_tag.py +20 -20
  314. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_tgrep.py +628 -626
  315. nltk-3.3/nltk/test/unit/test_tokenize.py +168 -0
  316. {nltk-3.2 → nltk-3.3}/nltk/test/unit/test_twitter_auth.py +185 -185
  317. nltk-3.3/nltk/test/unit/test_wordnet.py +134 -0
  318. nltk-3.3/nltk/test/unit/translate/test_bleu.py +244 -0
  319. nltk-3.3/nltk/test/unit/translate/test_gdfa.py +57 -0
  320. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm1.py +75 -75
  321. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm2.py +87 -87
  322. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm3.py +105 -105
  323. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm4.py +126 -126
  324. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm5.py +166 -166
  325. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_ibm_model.py +270 -270
  326. nltk-3.3/nltk/test/unit/translate/test_nist.py +37 -0
  327. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/test_stack_decoder.py +300 -300
  328. {nltk-3.2 → nltk-3.3}/nltk/test/unit/utils.py +41 -41
  329. {nltk-3.2 → nltk-3.3}/nltk/test/util.doctest +48 -48
  330. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet.doctest +595 -581
  331. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet_fixt.py +7 -6
  332. {nltk-3.2 → nltk-3.3}/nltk/test/wordnet_lch.doctest +53 -53
  333. {nltk-3.2 → nltk-3.3}/nltk/test/wsd.doctest +68 -68
  334. {nltk-3.2 → nltk-3.3}/nltk/text.py +663 -610
  335. {nltk-3.2 → nltk-3.3}/nltk/tgrep.py +938 -937
  336. {nltk-3.2 → nltk-3.3}/nltk/tokenize/__init__.py +130 -107
  337. {nltk-3.2 → nltk-3.3}/nltk/tokenize/api.py +77 -78
  338. {nltk-3.2 → nltk-3.3}/nltk/tokenize/casual.py +343 -342
  339. {nltk-3.2 → nltk-3.3}/nltk/tokenize/mwe.py +120 -120
  340. nltk-3.3/nltk/tokenize/nist.py +173 -0
  341. {nltk-3.2 → nltk-3.3}/nltk/tokenize/punkt.py +1620 -1601
  342. {nltk-3.2 → nltk-3.3}/nltk/tokenize/regexp.py +208 -202
  343. nltk-3.3/nltk/tokenize/repp.py +151 -0
  344. {nltk-3.2 → nltk-3.3}/nltk/tokenize/sexpr.py +140 -142
  345. {nltk-3.2 → nltk-3.3}/nltk/tokenize/simple.py +138 -136
  346. {nltk-3.2 → nltk-3.3}/nltk/tokenize/stanford.py +117 -113
  347. nltk-3.3/nltk/tokenize/stanford_segmenter.py +246 -0
  348. {nltk-3.2 → nltk-3.3}/nltk/tokenize/texttiling.py +458 -460
  349. nltk-3.3/nltk/tokenize/toktok.py +155 -0
  350. nltk-3.3/nltk/tokenize/treebank.py +363 -0
  351. nltk-3.3/nltk/tokenize/util.py +261 -0
  352. {nltk-3.2 → nltk-3.3}/nltk/toolbox.py +499 -497
  353. {nltk-3.2 → nltk-3.3}/nltk/translate/__init__.py +24 -24
  354. {nltk-3.2 → nltk-3.3}/nltk/translate/api.py +331 -321
  355. {nltk-3.2 → nltk-3.3}/nltk/translate/bleu_score.py +615 -556
  356. nltk-3.3/nltk/translate/chrf_score.py +137 -0
  357. {nltk-3.2 → nltk-3.3}/nltk/translate/gale_church.py +248 -235
  358. {nltk-3.2 → nltk-3.3}/nltk/translate/gdfa.py +135 -131
  359. nltk-3.3/nltk/translate/gleu_score.py +193 -0
  360. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm1.py +250 -241
  361. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm2.py +308 -308
  362. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm3.py +337 -337
  363. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm4.py +467 -467
  364. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm5.py +639 -639
  365. {nltk-3.2 → nltk-3.3}/nltk/translate/ibm_model.py +536 -533
  366. {nltk-3.2 → nltk-3.3}/nltk/translate/metrics.py +40 -40
  367. nltk-3.3/nltk/translate/nist_score.py +178 -0
  368. {nltk-3.2 → nltk-3.3}/nltk/translate/phrase_based.py +175 -175
  369. {nltk-3.2 → nltk-3.3}/nltk/translate/ribes_score.py +325 -325
  370. {nltk-3.2 → nltk-3.3}/nltk/translate/stack_decoder.py +499 -495
  371. {nltk-3.2 → nltk-3.3}/nltk/tree.py +1605 -1603
  372. {nltk-3.2 → nltk-3.3}/nltk/treeprettyprinter.py +564 -564
  373. {nltk-3.2 → nltk-3.3}/nltk/treetransforms.py +309 -309
  374. {nltk-3.2 → nltk-3.3}/nltk/twitter/__init__.py +28 -28
  375. {nltk-3.2 → nltk-3.3}/nltk/twitter/api.py +141 -136
  376. {nltk-3.2 → nltk-3.3}/nltk/twitter/common.py +264 -258
  377. {nltk-3.2 → nltk-3.3}/nltk/twitter/twitter_demo.py +288 -288
  378. {nltk-3.2 → nltk-3.3}/nltk/twitter/twitterclient.py +530 -530
  379. {nltk-3.2 → nltk-3.3}/nltk/twitter/util.py +147 -146
  380. nltk-3.3/nltk/util.py +751 -0
  381. {nltk-3.2 → nltk-3.3}/nltk/wsd.py +53 -53
  382. {nltk-3.2 → nltk-3.3/nltk.egg-info}/PKG-INFO +13 -5
  383. {nltk-3.2 → nltk-3.3}/nltk.egg-info/SOURCES.txt +26 -3
  384. nltk-3.3/nltk.egg-info/requires.txt +31 -0
  385. {nltk-3.2 → nltk-3.3}/setup.cfg +0 -1
  386. {nltk-3.2 → nltk-3.3}/setup.py +102 -75
  387. nltk-3.2/nltk/VERSION +0 -1
  388. nltk-3.2/nltk/compat.py +0 -680
  389. nltk-3.2/nltk/corpus/reader/pl196x.py +0 -278
  390. nltk-3.2/nltk/corpus/reader/verbnet.py +0 -408
  391. nltk-3.2/nltk/corpus/reader/wordlist.py +0 -37
  392. nltk-3.2/nltk/sentiment/vader_lexicon.txt +0 -7517
  393. nltk-3.2/nltk/six.py +0 -868
  394. nltk-3.2/nltk/stem/porter.py +0 -692
  395. nltk-3.2/nltk/test/unit/test_2x_compat.py +0 -26
  396. nltk-3.2/nltk/test/unit/test_stem.py +0 -36
  397. nltk-3.2/nltk/test/unit/test_tokenize.py +0 -23
  398. nltk-3.2/nltk/test/unit/translate/test_bleu.py +0 -84
  399. nltk-3.2/nltk/tokenize/treebank.py +0 -124
  400. nltk-3.2/nltk/tokenize/util.py +0 -90
  401. nltk-3.2/nltk/util.py +0 -1364
  402. {nltk-3.2 → nltk-3.3}/nltk/test/unit/__init__.py +0 -0
  403. {nltk-3.2 → nltk-3.3}/nltk/test/unit/translate/__init__.py +0 -0
  404. {nltk-3.2 → nltk-3.3}/nltk.egg-info/dependency_links.txt +0 -0
  405. {nltk-3.2 → nltk-3.3}/nltk.egg-info/not-zip-safe +0 -0
  406. {nltk-3.2 → nltk-3.3}/nltk.egg-info/top_level.txt +0 -0
@@ -1,384 +1,386 @@
1
- # Natural Language Toolkit: Logic
2
- #
3
- # Author: Peter Wang
4
- # Updated by: Dan Garrette <dhgarrette@gmail.com>
5
- #
6
- # Copyright (C) 2001-2015 NLTK Project
7
- # URL: <http://nltk.org>
8
- # For license information, see LICENSE.TXT
9
-
10
- """
11
- An implementation of the Hole Semantics model, following Blackburn and Bos,
12
- Representation and Inference for Natural Language (CSLI, 2005).
13
-
14
- The semantic representations are built by the grammar hole.fcfg.
15
- This module contains driver code to read in sentences and parse them
16
- according to a hole semantics grammar.
17
-
18
- After parsing, the semantic representation is in the form of an underspecified
19
- representation that is not easy to read. We use a "plugging" algorithm to
20
- convert that representation into first-order logic formulas.
21
- """
22
- from __future__ import print_function, unicode_literals
23
-
24
- from functools import reduce
25
-
26
- from nltk import compat
27
- from nltk.parse import load_parser
28
-
29
- from nltk.sem.skolemize import skolemize
30
- from nltk.sem.logic import (AllExpression, AndExpression, ApplicationExpression,
31
- ExistsExpression, IffExpression, ImpExpression,
32
- LambdaExpression, NegatedExpression, OrExpression)
33
-
34
-
35
- # Note that in this code there may be multiple types of trees being referred to:
36
- #
37
- # 1. parse trees
38
- # 2. the underspecified representation
39
- # 3. first-order logic formula trees
40
- # 4. the search space when plugging (search tree)
41
- #
42
-
43
- class Constants(object):
44
- ALL = 'ALL'
45
- EXISTS = 'EXISTS'
46
- NOT = 'NOT'
47
- AND = 'AND'
48
- OR = 'OR'
49
- IMP = 'IMP'
50
- IFF = 'IFF'
51
- PRED = 'PRED'
52
- LEQ = 'LEQ'
53
- HOLE = 'HOLE'
54
- LABEL = 'LABEL'
55
-
56
- MAP = {ALL: lambda v, e: AllExpression(v.variable, e),
57
- EXISTS: lambda v, e: ExistsExpression(v.variable, e),
58
- NOT: NegatedExpression,
59
- AND: AndExpression,
60
- OR: OrExpression,
61
- IMP: ImpExpression,
62
- IFF: IffExpression,
63
- PRED: ApplicationExpression}
64
-
65
-
66
- class HoleSemantics(object):
67
- """
68
- This class holds the broken-down components of a hole semantics, i.e. it
69
- extracts the holes, labels, logic formula fragments and constraints out of
70
- a big conjunction of such as produced by the hole semantics grammar. It
71
- then provides some operations on the semantics dealing with holes, labels
72
- and finding legal ways to plug holes with labels.
73
- """
74
- def __init__(self, usr):
75
- """
76
- Constructor. `usr' is a ``sem.Expression`` representing an
77
- Underspecified Representation Structure (USR). A USR has the following
78
- special predicates:
79
- ALL(l,v,n),
80
- EXISTS(l,v,n),
81
- AND(l,n,n),
82
- OR(l,n,n),
83
- IMP(l,n,n),
84
- IFF(l,n,n),
85
- PRED(l,v,n,v[,v]*) where the brackets and star indicate zero or more repetitions,
86
- LEQ(n,n),
87
- HOLE(n),
88
- LABEL(n)
89
- where l is the label of the node described by the predicate, n is either
90
- a label or a hole, and v is a variable.
91
- """
92
- self.holes = set()
93
- self.labels = set()
94
- self.fragments = {} # mapping of label -> formula fragment
95
- self.constraints = set() # set of Constraints
96
- self._break_down(usr)
97
- self.top_most_labels = self._find_top_most_labels()
98
- self.top_hole = self._find_top_hole()
99
-
100
- def is_node(self, x):
101
- """
102
- Return true if x is a node (label or hole) in this semantic
103
- representation.
104
- """
105
- return x in (self.labels | self.holes)
106
-
107
- def _break_down(self, usr):
108
- """
109
- Extract holes, labels, formula fragments and constraints from the hole
110
- semantics underspecified representation (USR).
111
- """
112
- if isinstance(usr, AndExpression):
113
- self._break_down(usr.first)
114
- self._break_down(usr.second)
115
- elif isinstance(usr, ApplicationExpression):
116
- func, args = usr.uncurry()
117
- if func.variable.name == Constants.LEQ:
118
- self.constraints.add(Constraint(args[0], args[1]))
119
- elif func.variable.name == Constants.HOLE:
120
- self.holes.add(args[0])
121
- elif func.variable.name == Constants.LABEL:
122
- self.labels.add(args[0])
123
- else:
124
- label = args[0]
125
- assert label not in self.fragments
126
- self.fragments[label] = (func, args[1:])
127
- else:
128
- raise ValueError(usr.label())
129
-
130
- def _find_top_nodes(self, node_list):
131
- top_nodes = node_list.copy()
132
- for f in compat.itervalues(self.fragments):
133
- # the label is the first argument of the predicate
134
- args = f[1]
135
- for arg in args:
136
- if arg in node_list:
137
- top_nodes.discard(arg)
138
- return top_nodes
139
-
140
- def _find_top_most_labels(self):
141
- """
142
- Return the set of labels which are not referenced directly as part of
143
- another formula fragment. These will be the top-most labels for the
144
- subtree that they are part of.
145
- """
146
- return self._find_top_nodes(self.labels)
147
-
148
- def _find_top_hole(self):
149
- """
150
- Return the hole that will be the top of the formula tree.
151
- """
152
- top_holes = self._find_top_nodes(self.holes)
153
- assert len(top_holes) == 1 # it must be unique
154
- return top_holes.pop()
155
-
156
- def pluggings(self):
157
- """
158
- Calculate and return all the legal pluggings (mappings of labels to
159
- holes) of this semantics given the constraints.
160
- """
161
- record = []
162
- self._plug_nodes([(self.top_hole, [])], self.top_most_labels, {}, record)
163
- return record
164
-
165
- def _plug_nodes(self, queue, potential_labels, plug_acc, record):
166
- """
167
- Plug the nodes in `queue' with the labels in `potential_labels'.
168
-
169
- Each element of `queue' is a tuple of the node to plug and the list of
170
- ancestor holes from the root of the graph to that node.
171
-
172
- `potential_labels' is a set of the labels which are still available for
173
- plugging.
174
-
175
- `plug_acc' is the incomplete mapping of holes to labels made on the
176
- current branch of the search tree so far.
177
-
178
- `record' is a list of all the complete pluggings that we have found in
179
- total so far. It is the only parameter that is destructively updated.
180
- """
181
- if queue != []:
182
- (node, ancestors) = queue[0]
183
- if node in self.holes:
184
- # The node is a hole, try to plug it.
185
- self._plug_hole(node, ancestors, queue[1:], potential_labels, plug_acc, record)
186
- else:
187
- assert node in self.labels
188
- # The node is a label. Replace it in the queue by the holes and
189
- # labels in the formula fragment named by that label.
190
- args = self.fragments[node][1]
191
- head = [(a, ancestors) for a in args if self.is_node(a)]
192
- self._plug_nodes(head + queue[1:], potential_labels, plug_acc, record)
193
- else:
194
- raise Exception('queue empty')
195
-
196
- def _plug_hole(self, hole, ancestors0, queue, potential_labels0,
197
- plug_acc0, record):
198
- """
199
- Try all possible ways of plugging a single hole.
200
- See _plug_nodes for the meanings of the parameters.
201
- """
202
- # Add the current hole we're trying to plug into the list of ancestors.
203
- assert hole not in ancestors0
204
- ancestors = [hole] + ancestors0
205
-
206
- # Try each potential label in this hole in turn.
207
- for l in potential_labels0:
208
- # Is the label valid in this hole?
209
- if self._violates_constraints(l, ancestors):
210
- continue
211
-
212
- plug_acc = plug_acc0.copy()
213
- plug_acc[hole] = l
214
- potential_labels = potential_labels0.copy()
215
- potential_labels.remove(l)
216
-
217
- if len(potential_labels) == 0:
218
- # No more potential labels. That must mean all the holes have
219
- # been filled so we have found a legal plugging so remember it.
220
- #
221
- # Note that the queue might not be empty because there might
222
- # be labels on there that point to formula fragments with
223
- # no holes in them. _sanity_check_plugging will make sure
224
- # all holes are filled.
225
- self._sanity_check_plugging(plug_acc, self.top_hole, [])
226
- record.append(plug_acc)
227
- else:
228
- # Recursively try to fill in the rest of the holes in the
229
- # queue. The label we just plugged into the hole could have
230
- # holes of its own so at the end of the queue. Putting it on
231
- # the end of the queue gives us a breadth-first search, so that
232
- # all the holes at level i of the formula tree are filled
233
- # before filling level i+1.
234
- # A depth-first search would work as well since the trees must
235
- # be finite but the bookkeeping would be harder.
236
- self._plug_nodes(queue + [(l, ancestors)], potential_labels, plug_acc, record)
237
-
238
- def _violates_constraints(self, label, ancestors):
239
- """
240
- Return True if the `label' cannot be placed underneath the holes given
241
- by the set `ancestors' because it would violate the constraints imposed
242
- on it.
243
- """
244
- for c in self.constraints:
245
- if c.lhs == label:
246
- if c.rhs not in ancestors:
247
- return True
248
- return False
249
-
250
- def _sanity_check_plugging(self, plugging, node, ancestors):
251
- """
252
- Make sure that a given plugging is legal. We recursively go through
253
- each node and make sure that no constraints are violated.
254
- We also check that all holes have been filled.
255
- """
256
- if node in self.holes:
257
- ancestors = [node] + ancestors
258
- label = plugging[node]
259
- else:
260
- label = node
261
- assert label in self.labels
262
- for c in self.constraints:
263
- if c.lhs == label:
264
- assert c.rhs in ancestors
265
- args = self.fragments[label][1]
266
- for arg in args:
267
- if self.is_node(arg):
268
- self._sanity_check_plugging(plugging, arg, [label] + ancestors)
269
-
270
- def formula_tree(self, plugging):
271
- """
272
- Return the first-order logic formula tree for this underspecified
273
- representation using the plugging given.
274
- """
275
- return self._formula_tree(plugging, self.top_hole)
276
-
277
- def _formula_tree(self, plugging, node):
278
- if node in plugging:
279
- return self._formula_tree(plugging, plugging[node])
280
- elif node in self.fragments:
281
- pred, args = self.fragments[node]
282
- children = [self._formula_tree(plugging, arg) for arg in args]
283
- return reduce(Constants.MAP[pred.variable.name], children)
284
- else:
285
- return node
286
-
287
-
288
- @compat.python_2_unicode_compatible
289
- class Constraint(object):
290
- """
291
- This class represents a constraint of the form (L =< N),
292
- where L is a label and N is a node (a label or a hole).
293
- """
294
- def __init__(self, lhs, rhs):
295
- self.lhs = lhs
296
- self.rhs = rhs
297
-
298
- def __eq__(self, other):
299
- if self.__class__ == other.__class__:
300
- return self.lhs == other.lhs and self.rhs == other.rhs
301
- else:
302
- return False
303
-
304
- def __ne__(self, other):
305
- return not (self == other)
306
-
307
- def __hash__(self):
308
- return hash(repr(self))
309
-
310
- def __repr__(self):
311
- return '(%s < %s)' % (self.lhs, self.rhs)
312
-
313
-
314
- def hole_readings(sentence, grammar_filename=None, verbose=False):
315
- if not grammar_filename:
316
- grammar_filename = 'grammars/sample_grammars/hole.fcfg'
317
-
318
- if verbose:
319
- print('Reading grammar file', grammar_filename)
320
-
321
- parser = load_parser(grammar_filename)
322
-
323
- # Parse the sentence.
324
- tokens = sentence.split()
325
- trees = list(parser.parse(tokens))
326
- if verbose:
327
- print('Got %d different parses' % len(trees))
328
-
329
- all_readings = []
330
- for tree in trees:
331
- # Get the semantic feature from the top of the parse tree.
332
- sem = tree.label()['SEM'].simplify()
333
-
334
- # Print the raw semantic representation.
335
- if verbose:
336
- print('Raw: ', sem)
337
-
338
- # Skolemize away all quantifiers. All variables become unique.
339
- while isinstance(sem, LambdaExpression):
340
- sem = sem.term
341
- skolemized = skolemize(sem)
342
-
343
- if verbose:
344
- print('Skolemized:', skolemized)
345
-
346
- # Break the hole semantics representation down into its components
347
- # i.e. holes, labels, formula fragments and constraints.
348
- hole_sem = HoleSemantics(skolemized)
349
-
350
- # Maybe show the details of the semantic representation.
351
- if verbose:
352
- print('Holes: ', hole_sem.holes)
353
- print('Labels: ', hole_sem.labels)
354
- print('Constraints: ', hole_sem.constraints)
355
- print('Top hole: ', hole_sem.top_hole)
356
- print('Top labels: ', hole_sem.top_most_labels)
357
- print('Fragments:')
358
- for l, f in hole_sem.fragments.items():
359
- print('\t%s: %s' % (l, f))
360
-
361
- # Find all the possible ways to plug the formulas together.
362
- pluggings = hole_sem.pluggings()
363
-
364
- # Build FOL formula trees using the pluggings.
365
- readings = list(map(hole_sem.formula_tree, pluggings))
366
-
367
- # Print out the formulas in a textual format.
368
- if verbose:
369
- for i, r in enumerate(readings):
370
- print()
371
- print('%d. %s' % (i, r))
372
- print()
373
-
374
- all_readings.extend(readings)
375
-
376
- return all_readings
377
-
378
-
379
- if __name__ == '__main__':
380
- for r in hole_readings('a dog barks'):
381
- print(r)
382
- print()
383
- for r in hole_readings('every girl chases a dog'):
384
- print(r)
1
+ # Natural Language Toolkit: Logic
2
+ #
3
+ # Author: Peter Wang
4
+ # Updated by: Dan Garrette <dhgarrette@gmail.com>
5
+ #
6
+ # Copyright (C) 2001-2018 NLTK Project
7
+ # URL: <http://nltk.org>
8
+ # For license information, see LICENSE.TXT
9
+
10
+ """
11
+ An implementation of the Hole Semantics model, following Blackburn and Bos,
12
+ Representation and Inference for Natural Language (CSLI, 2005).
13
+
14
+ The semantic representations are built by the grammar hole.fcfg.
15
+ This module contains driver code to read in sentences and parse them
16
+ according to a hole semantics grammar.
17
+
18
+ After parsing, the semantic representation is in the form of an underspecified
19
+ representation that is not easy to read. We use a "plugging" algorithm to
20
+ convert that representation into first-order logic formulas.
21
+ """
22
+ from __future__ import print_function, unicode_literals
23
+
24
+ from functools import reduce
25
+
26
+ from six import itervalues
27
+
28
+ from nltk import compat
29
+ from nltk.parse import load_parser
30
+
31
+ from nltk.sem.skolemize import skolemize
32
+ from nltk.sem.logic import (AllExpression, AndExpression, ApplicationExpression,
33
+ ExistsExpression, IffExpression, ImpExpression,
34
+ LambdaExpression, NegatedExpression, OrExpression)
35
+
36
+
37
+ # Note that in this code there may be multiple types of trees being referred to:
38
+ #
39
+ # 1. parse trees
40
+ # 2. the underspecified representation
41
+ # 3. first-order logic formula trees
42
+ # 4. the search space when plugging (search tree)
43
+ #
44
+
45
+ class Constants(object):
46
+ ALL = 'ALL'
47
+ EXISTS = 'EXISTS'
48
+ NOT = 'NOT'
49
+ AND = 'AND'
50
+ OR = 'OR'
51
+ IMP = 'IMP'
52
+ IFF = 'IFF'
53
+ PRED = 'PRED'
54
+ LEQ = 'LEQ'
55
+ HOLE = 'HOLE'
56
+ LABEL = 'LABEL'
57
+
58
+ MAP = {ALL: lambda v, e: AllExpression(v.variable, e),
59
+ EXISTS: lambda v, e: ExistsExpression(v.variable, e),
60
+ NOT: NegatedExpression,
61
+ AND: AndExpression,
62
+ OR: OrExpression,
63
+ IMP: ImpExpression,
64
+ IFF: IffExpression,
65
+ PRED: ApplicationExpression}
66
+
67
+
68
+ class HoleSemantics(object):
69
+ """
70
+ This class holds the broken-down components of a hole semantics, i.e. it
71
+ extracts the holes, labels, logic formula fragments and constraints out of
72
+ a big conjunction of such as produced by the hole semantics grammar. It
73
+ then provides some operations on the semantics dealing with holes, labels
74
+ and finding legal ways to plug holes with labels.
75
+ """
76
+ def __init__(self, usr):
77
+ """
78
+ Constructor. `usr' is a ``sem.Expression`` representing an
79
+ Underspecified Representation Structure (USR). A USR has the following
80
+ special predicates:
81
+ ALL(l,v,n),
82
+ EXISTS(l,v,n),
83
+ AND(l,n,n),
84
+ OR(l,n,n),
85
+ IMP(l,n,n),
86
+ IFF(l,n,n),
87
+ PRED(l,v,n,v[,v]*) where the brackets and star indicate zero or more repetitions,
88
+ LEQ(n,n),
89
+ HOLE(n),
90
+ LABEL(n)
91
+ where l is the label of the node described by the predicate, n is either
92
+ a label or a hole, and v is a variable.
93
+ """
94
+ self.holes = set()
95
+ self.labels = set()
96
+ self.fragments = {} # mapping of label -> formula fragment
97
+ self.constraints = set() # set of Constraints
98
+ self._break_down(usr)
99
+ self.top_most_labels = self._find_top_most_labels()
100
+ self.top_hole = self._find_top_hole()
101
+
102
+ def is_node(self, x):
103
+ """
104
+ Return true if x is a node (label or hole) in this semantic
105
+ representation.
106
+ """
107
+ return x in (self.labels | self.holes)
108
+
109
+ def _break_down(self, usr):
110
+ """
111
+ Extract holes, labels, formula fragments and constraints from the hole
112
+ semantics underspecified representation (USR).
113
+ """
114
+ if isinstance(usr, AndExpression):
115
+ self._break_down(usr.first)
116
+ self._break_down(usr.second)
117
+ elif isinstance(usr, ApplicationExpression):
118
+ func, args = usr.uncurry()
119
+ if func.variable.name == Constants.LEQ:
120
+ self.constraints.add(Constraint(args[0], args[1]))
121
+ elif func.variable.name == Constants.HOLE:
122
+ self.holes.add(args[0])
123
+ elif func.variable.name == Constants.LABEL:
124
+ self.labels.add(args[0])
125
+ else:
126
+ label = args[0]
127
+ assert label not in self.fragments
128
+ self.fragments[label] = (func, args[1:])
129
+ else:
130
+ raise ValueError(usr.label())
131
+
132
+ def _find_top_nodes(self, node_list):
133
+ top_nodes = node_list.copy()
134
+ for f in itervalues(self.fragments):
135
+ # the label is the first argument of the predicate
136
+ args = f[1]
137
+ for arg in args:
138
+ if arg in node_list:
139
+ top_nodes.discard(arg)
140
+ return top_nodes
141
+
142
+ def _find_top_most_labels(self):
143
+ """
144
+ Return the set of labels which are not referenced directly as part of
145
+ another formula fragment. These will be the top-most labels for the
146
+ subtree that they are part of.
147
+ """
148
+ return self._find_top_nodes(self.labels)
149
+
150
+ def _find_top_hole(self):
151
+ """
152
+ Return the hole that will be the top of the formula tree.
153
+ """
154
+ top_holes = self._find_top_nodes(self.holes)
155
+ assert len(top_holes) == 1 # it must be unique
156
+ return top_holes.pop()
157
+
158
+ def pluggings(self):
159
+ """
160
+ Calculate and return all the legal pluggings (mappings of labels to
161
+ holes) of this semantics given the constraints.
162
+ """
163
+ record = []
164
+ self._plug_nodes([(self.top_hole, [])], self.top_most_labels, {}, record)
165
+ return record
166
+
167
+ def _plug_nodes(self, queue, potential_labels, plug_acc, record):
168
+ """
169
+ Plug the nodes in `queue' with the labels in `potential_labels'.
170
+
171
+ Each element of `queue' is a tuple of the node to plug and the list of
172
+ ancestor holes from the root of the graph to that node.
173
+
174
+ `potential_labels' is a set of the labels which are still available for
175
+ plugging.
176
+
177
+ `plug_acc' is the incomplete mapping of holes to labels made on the
178
+ current branch of the search tree so far.
179
+
180
+ `record' is a list of all the complete pluggings that we have found in
181
+ total so far. It is the only parameter that is destructively updated.
182
+ """
183
+ if queue != []:
184
+ (node, ancestors) = queue[0]
185
+ if node in self.holes:
186
+ # The node is a hole, try to plug it.
187
+ self._plug_hole(node, ancestors, queue[1:], potential_labels, plug_acc, record)
188
+ else:
189
+ assert node in self.labels
190
+ # The node is a label. Replace it in the queue by the holes and
191
+ # labels in the formula fragment named by that label.
192
+ args = self.fragments[node][1]
193
+ head = [(a, ancestors) for a in args if self.is_node(a)]
194
+ self._plug_nodes(head + queue[1:], potential_labels, plug_acc, record)
195
+ else:
196
+ raise Exception('queue empty')
197
+
198
+ def _plug_hole(self, hole, ancestors0, queue, potential_labels0,
199
+ plug_acc0, record):
200
+ """
201
+ Try all possible ways of plugging a single hole.
202
+ See _plug_nodes for the meanings of the parameters.
203
+ """
204
+ # Add the current hole we're trying to plug into the list of ancestors.
205
+ assert hole not in ancestors0
206
+ ancestors = [hole] + ancestors0
207
+
208
+ # Try each potential label in this hole in turn.
209
+ for l in potential_labels0:
210
+ # Is the label valid in this hole?
211
+ if self._violates_constraints(l, ancestors):
212
+ continue
213
+
214
+ plug_acc = plug_acc0.copy()
215
+ plug_acc[hole] = l
216
+ potential_labels = potential_labels0.copy()
217
+ potential_labels.remove(l)
218
+
219
+ if len(potential_labels) == 0:
220
+ # No more potential labels. That must mean all the holes have
221
+ # been filled so we have found a legal plugging so remember it.
222
+ #
223
+ # Note that the queue might not be empty because there might
224
+ # be labels on there that point to formula fragments with
225
+ # no holes in them. _sanity_check_plugging will make sure
226
+ # all holes are filled.
227
+ self._sanity_check_plugging(plug_acc, self.top_hole, [])
228
+ record.append(plug_acc)
229
+ else:
230
+ # Recursively try to fill in the rest of the holes in the
231
+ # queue. The label we just plugged into the hole could have
232
+ # holes of its own so at the end of the queue. Putting it on
233
+ # the end of the queue gives us a breadth-first search, so that
234
+ # all the holes at level i of the formula tree are filled
235
+ # before filling level i+1.
236
+ # A depth-first search would work as well since the trees must
237
+ # be finite but the bookkeeping would be harder.
238
+ self._plug_nodes(queue + [(l, ancestors)], potential_labels, plug_acc, record)
239
+
240
+ def _violates_constraints(self, label, ancestors):
241
+ """
242
+ Return True if the `label' cannot be placed underneath the holes given
243
+ by the set `ancestors' because it would violate the constraints imposed
244
+ on it.
245
+ """
246
+ for c in self.constraints:
247
+ if c.lhs == label:
248
+ if c.rhs not in ancestors:
249
+ return True
250
+ return False
251
+
252
+ def _sanity_check_plugging(self, plugging, node, ancestors):
253
+ """
254
+ Make sure that a given plugging is legal. We recursively go through
255
+ each node and make sure that no constraints are violated.
256
+ We also check that all holes have been filled.
257
+ """
258
+ if node in self.holes:
259
+ ancestors = [node] + ancestors
260
+ label = plugging[node]
261
+ else:
262
+ label = node
263
+ assert label in self.labels
264
+ for c in self.constraints:
265
+ if c.lhs == label:
266
+ assert c.rhs in ancestors
267
+ args = self.fragments[label][1]
268
+ for arg in args:
269
+ if self.is_node(arg):
270
+ self._sanity_check_plugging(plugging, arg, [label] + ancestors)
271
+
272
+ def formula_tree(self, plugging):
273
+ """
274
+ Return the first-order logic formula tree for this underspecified
275
+ representation using the plugging given.
276
+ """
277
+ return self._formula_tree(plugging, self.top_hole)
278
+
279
+ def _formula_tree(self, plugging, node):
280
+ if node in plugging:
281
+ return self._formula_tree(plugging, plugging[node])
282
+ elif node in self.fragments:
283
+ pred, args = self.fragments[node]
284
+ children = [self._formula_tree(plugging, arg) for arg in args]
285
+ return reduce(Constants.MAP[pred.variable.name], children)
286
+ else:
287
+ return node
288
+
289
+
290
+ @compat.python_2_unicode_compatible
291
+ class Constraint(object):
292
+ """
293
+ This class represents a constraint of the form (L =< N),
294
+ where L is a label and N is a node (a label or a hole).
295
+ """
296
+ def __init__(self, lhs, rhs):
297
+ self.lhs = lhs
298
+ self.rhs = rhs
299
+
300
+ def __eq__(self, other):
301
+ if self.__class__ == other.__class__:
302
+ return self.lhs == other.lhs and self.rhs == other.rhs
303
+ else:
304
+ return False
305
+
306
+ def __ne__(self, other):
307
+ return not (self == other)
308
+
309
+ def __hash__(self):
310
+ return hash(repr(self))
311
+
312
+ def __repr__(self):
313
+ return '(%s < %s)' % (self.lhs, self.rhs)
314
+
315
+
316
+ def hole_readings(sentence, grammar_filename=None, verbose=False):
317
+ if not grammar_filename:
318
+ grammar_filename = 'grammars/sample_grammars/hole.fcfg'
319
+
320
+ if verbose:
321
+ print('Reading grammar file', grammar_filename)
322
+
323
+ parser = load_parser(grammar_filename)
324
+
325
+ # Parse the sentence.
326
+ tokens = sentence.split()
327
+ trees = list(parser.parse(tokens))
328
+ if verbose:
329
+ print('Got %d different parses' % len(trees))
330
+
331
+ all_readings = []
332
+ for tree in trees:
333
+ # Get the semantic feature from the top of the parse tree.
334
+ sem = tree.label()['SEM'].simplify()
335
+
336
+ # Print the raw semantic representation.
337
+ if verbose:
338
+ print('Raw: ', sem)
339
+
340
+ # Skolemize away all quantifiers. All variables become unique.
341
+ while isinstance(sem, LambdaExpression):
342
+ sem = sem.term
343
+ skolemized = skolemize(sem)
344
+
345
+ if verbose:
346
+ print('Skolemized:', skolemized)
347
+
348
+ # Break the hole semantics representation down into its components
349
+ # i.e. holes, labels, formula fragments and constraints.
350
+ hole_sem = HoleSemantics(skolemized)
351
+
352
+ # Maybe show the details of the semantic representation.
353
+ if verbose:
354
+ print('Holes: ', hole_sem.holes)
355
+ print('Labels: ', hole_sem.labels)
356
+ print('Constraints: ', hole_sem.constraints)
357
+ print('Top hole: ', hole_sem.top_hole)
358
+ print('Top labels: ', hole_sem.top_most_labels)
359
+ print('Fragments:')
360
+ for l, f in hole_sem.fragments.items():
361
+ print('\t%s: %s' % (l, f))
362
+
363
+ # Find all the possible ways to plug the formulas together.
364
+ pluggings = hole_sem.pluggings()
365
+
366
+ # Build FOL formula trees using the pluggings.
367
+ readings = list(map(hole_sem.formula_tree, pluggings))
368
+
369
+ # Print out the formulas in a textual format.
370
+ if verbose:
371
+ for i, r in enumerate(readings):
372
+ print()
373
+ print('%d. %s' % (i, r))
374
+ print()
375
+
376
+ all_readings.extend(readings)
377
+
378
+ return all_readings
379
+
380
+
381
+ if __name__ == '__main__':
382
+ for r in hole_readings('a dog barks'):
383
+ print(r)
384
+ print()
385
+ for r in hole_readings('every girl chases a dog'):
386
+ print(r)