Advertisement
Guest User

Untitled

a guest
Apr 18th, 2019
90
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 7.71 KB | None | 0 0
  1. $ python
  2. Python 3.5.1+ (default, Mar 30 2016, 22:46:26)
  3. [GCC 5.3.1 20160330] on linux
  4. Type "help", "copyright", "credits" or "license" for more information.
  5. >>> from revscoring import Model
  6. >>> m = Model.load(open("models/enwiki.damaging.gradient_boosting.model"))
  7. >>> m.estimator.feature_importances_
  8. array([8.39962459e-03, 7.93533540e-03, 3.04496444e-08, 2.25381150e-02,
  9. 2.08058610e-02, 2.26880141e-02, 1.87132900e-02, 1.60180859e-02,
  10. 2.23545834e-02, 2.14488512e-02, 1.90494208e-02, 2.54534679e-02,
  11. 2.66413539e-02, 2.75077614e-02, 2.24572831e-02, 2.85643380e-02,
  12. 1.40673462e-02, 1.08283797e-02, 5.28226208e-03, 1.98192473e-02,
  13. 1.20970667e-02, 7.50752762e-03, 9.40235191e-03, 9.02839723e-03,
  14. 3.37030110e-03, 1.13464298e-02, 1.21129339e-02, 6.38404279e-03,
  15. 3.09859252e-03, 2.05198411e-03, 1.40617645e-03, 5.02068509e-03,
  16. 2.84700947e-03, 2.23312854e-03, 2.60885299e-02, 1.92700904e-02,
  17. 1.58282365e-02, 1.34496524e-02, 4.13376140e-03, 7.59400902e-03,
  18. 1.05878865e-02, 6.09718173e-03, 1.14254854e-02, 5.77975715e-03,
  19. 3.86463005e-03, 3.94155522e-03, 5.69761327e-03, 2.79282947e-06,
  20. 5.06161491e-03, 3.82128496e-05, 2.79054289e-02, 2.26680674e-05,
  21. 3.60489149e-02, 1.23666918e-01, 5.83742795e-03, 1.04978762e-02,
  22. 2.40607425e-03, 2.32938135e-03, 2.94554717e-04, 4.04082964e-03,
  23. 3.80689748e-03, 2.61396452e-04, 3.73664007e-03, 3.22918560e-03,
  24. 3.67060822e-04, 7.11120470e-03, 6.88971235e-03, 2.62036017e-04,
  25. 1.51509897e-02, 1.34228730e-02, 9.61967865e-03, 2.93981526e-02,
  26. 2.76017846e-02, 1.56533996e-02, 9.17425514e-03, 6.41188488e-03,
  27. 4.61004255e-03, 2.15756091e-02, 1.37733335e-02, 7.55350051e-03])
  28. >>> feature_importances = list(zip(m.features, m.estimator.feature_importances_))
  29. >>> feature_importances.sort(key=lambda i: i[1], reverse=True)
  30. >>> for feature, importance in feature_importances:
  31. ... print(feature, importance)
  32. ...
  33. feature.log((temporal.revision.user.seconds_since_registration + 1)) 0.12366691844592674
  34. feature.revision.user.is_anon 0.036048914867770496
  35. feature.english.dictionary.revision.diff.dict_word_prop_delta_sum 0.02939815258260414
  36. feature.revision.parent.markups_per_token 0.028564338001569347
  37. feature.revision.user.is_patroller 0.027905428860413462
  38. feature.english.dictionary.revision.diff.dict_word_prop_delta_increase 0.027601784631011272
  39. feature.revision.parent.words_per_token 0.027507761420741125
  40. feature.revision.parent.chars_per_word 0.026641353940771998
  41. feature.revision.diff.chars_change 0.02608852986429265
  42. feature.log((wikitext.revision.parent.ref_tags + 1)) 0.02545346787161345
  43. feature.log((len(<datasource.wikitext.revision.parent.words>) + 1)) 0.022688014137860642
  44. feature.log((wikitext.revision.parent.chars + 1)) 0.022538115006001207
  45. feature.revision.parent.uppercase_words_per_word 0.022457283134036873
  46. feature.log((wikitext.revision.parent.wikilinks + 1)) 0.022354583376726862
  47. feature.english.dictionary.revision.diff.non_dict_word_prop_delta_sum 0.021575609104639532
  48. feature.log((wikitext.revision.parent.external_links + 1)) 0.021448851189920102
  49. feature.log((len(<datasource.tokenized(datasource.revision.parent.text)>) + 1)) 0.020805860972874828
  50. feature.wikitext.revision.diff.markup_prop_delta_sum 0.01981924734354541
  51. feature.revision.diff.tokens_change 0.019270090409458637
  52. feature.log((wikitext.revision.parent.templates + 1)) 0.01904942084841798
  53. feature.log((len(<datasource.wikitext.revision.parent.uppercase_words>) + 1)) 0.01871328996256505
  54. feature.log((wikitext.revision.parent.headings + 1)) 0.016018085890865324
  55. feature.revision.diff.words_change 0.01582823654911597
  56. feature.english.dictionary.revision.diff.dict_word_prop_delta_decrease 0.015653399551568116
  57. feature.english.dictionary.revision.diff.dict_word_delta_sum 0.015150989716091658
  58. feature.wikitext.revision.diff.markup_delta_sum 0.014067346247203705
  59. feature.english.dictionary.revision.diff.non_dict_word_prop_delta_increase 0.013773333528693383
  60. feature.revision.diff.markups_change 0.013449652435971635
  61. feature.english.dictionary.revision.diff.dict_word_delta_increase 0.01342287297765271
  62. feature.wikitext.revision.diff.number_prop_delta_increase 0.01211293387957829
  63. feature.wikitext.revision.diff.markup_prop_delta_increase 0.012097066726975006
  64. feature.revision.diff.tags_change 0.011425485365658749
  65. feature.wikitext.revision.diff.number_prop_delta_sum 0.01134642977250017
  66. feature.wikitext.revision.diff.markup_delta_increase 0.010828379723584245
  67. feature.revision.diff.wikilinks_change 0.010587886494373096
  68. feature.revision.comment.has_link 0.010497876181199798
  69. feature.english.dictionary.revision.diff.dict_word_delta_decrease 0.009619678652952076
  70. feature.wikitext.revision.diff.number_delta_sum 0.0094023519140284
  71. feature.english.dictionary.revision.diff.non_dict_word_delta_sum 0.0091742551371313
  72. feature.wikitext.revision.diff.number_delta_increase 0.009028397234705816
  73. feature.revision.page.is_articleish 0.00839962458724949
  74. feature.revision.page.is_mainspace 0.007935335395640522
  75. feature.revision.diff.external_links_change 0.0075940090208491155
  76. feature.english.dictionary.revision.diff.non_dict_word_prop_delta_decrease 0.007553500511543981
  77. feature.wikitext.revision.diff.markup_prop_delta_decrease 0.007507527624647461
  78. feature.english.informals.revision.diff.match_prop_delta_sum 0.007111204704111318
  79. feature.english.informals.revision.diff.match_prop_delta_increase 0.006889712352520465
  80. feature.english.dictionary.revision.diff.non_dict_word_delta_increase 0.00641188487936276
  81. feature.wikitext.revision.diff.number_prop_delta_decrease 0.0063840427928057355
  82. feature.revision.diff.templates_change 0.006097181732340459
  83. feature.revision.comment.suggests_section_edit 0.00583742794779075
  84. feature.revision.diff.ref_tags_change 0.005779757152431128
  85. feature.revision.user.is_bot 0.005697613274778659
  86. feature.wikitext.revision.diff.markup_delta_decrease 0.005282262078290739
  87. feature.revision.user.is_admin 0.005061614910269018
  88. feature.wikitext.revision.diff.uppercase_word_prop_delta_sum 0.005020685087434651
  89. feature.english.dictionary.revision.diff.non_dict_word_delta_decrease 0.004610042548827057
  90. feature.revision.diff.headings_change 0.004133761404837777
  91. feature.english.badwords.revision.diff.match_prop_delta_sum 0.004040829642063472
  92. feature.revision.diff.longest_new_repeated_char 0.003941555219983169
  93. feature.revision.diff.longest_new_token 0.003864630046476644
  94. feature.english.badwords.revision.diff.match_prop_delta_increase 0.0038068974752894892
  95. feature.english.informals.revision.diff.match_delta_sum 0.0037366400654502505
  96. feature.wikitext.revision.diff.number_delta_decrease 0.003370301104268293
  97. feature.english.informals.revision.diff.match_delta_increase 0.0032291856024817024
  98. feature.wikitext.revision.diff.uppercase_word_delta_sum 0.0030985925217799654
  99. feature.wikitext.revision.diff.uppercase_word_prop_delta_increase 0.0028470094671780564
  100. feature.english.badwords.revision.diff.match_delta_sum 0.0024060742467007412
  101. feature.english.badwords.revision.diff.match_delta_increase 0.0023293813475623493
  102. feature.wikitext.revision.diff.uppercase_word_prop_delta_decrease 0.002233128541417208
  103. feature.wikitext.revision.diff.uppercase_word_delta_increase 0.002051984108632766
  104. feature.wikitext.revision.diff.uppercase_word_delta_decrease 0.001406176448753365
  105. feature.english.informals.revision.diff.match_delta_decrease 0.0003670608217441542
  106. feature.english.badwords.revision.diff.match_delta_decrease 0.00029455471697025686
  107. feature.english.informals.revision.diff.match_prop_delta_decrease 0.0002620360172541197
  108. feature.english.badwords.revision.diff.match_prop_delta_decrease 0.0002613964515332821
  109. feature.revision.user.is_trusted 3.82128496286288e-05
  110. feature.revision.user.is_curator 2.2668067373879832e-05
  111. feature.revision.user.has_advanced_rights 2.7928294714833776e-06
  112. feature.revision.page.is_draftspace 3.044964439759688e-08
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement