Skip to content

Instantly share code, notes, and snippets.

@halfak
Created April 30, 2020 19:17
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save halfak/53203c62f54dd9b83a4f2abc293b8534 to your computer and use it in GitHub Desktop.
Save halfak/53203c62f54dd9b83a4f2abc293b8534 to your computer and use it in GitHub Desktop.
$ python
Python 3.5.3 (default, Sep 27 2018, 17:25:39)
[GCC 6.3.0 20170516] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from revscoring import Model
>>> model = Model.load(open("models/ptwiki.wp10.gradient_boosting.model"))
>>> importance_features = list(sorted(zip(model.estimator.feature_importances_, model.features), reverse=True))
>>> for importance, feature in importance_features:
... print(round(importance, 3), feature)
...
0.099 feature.wikitext.revision.chars
0.094 feature.wikitext.revision.content_chars
0.06 feature.(wikitext.revision.headings_by_level(2) / max(wikitext.revision.content_chars, 1))
0.053 feature.(ptwiki.revision.shortened_footnote_templates + wikitext.revision.ref_tags)
0.049 feature.((((((ptwiki.revision.image_links + ptwiki.revision.image_template) + ptwiki.revision.images_in_templates) + (ptwiki.revision.side_by_side_image_templates * 2)) + ptwiki.revision.images_in_tags) + ptwiki.infobox_images) / max(wikitext.revision.content_chars, 1))
0.049 feature.wikitext.revision.wikilinks
0.045 feature.(ptwiki.revision.category_links / max(wikitext.revision.content_chars, 1))
0.042 feature.wikitext.revision.ref_tags
0.036 feature.wikitext.revision.headings_by_level(2)
0.035 feature.(wikitext.revision.wikilinks / max(wikitext.revision.content_chars, 1))
0.035 feature.(ptwiki.revision.cite_templates + ptwiki.revision.shortened_footnote_templates)
0.035 feature.log((ptwiki.revision.paragraphs_without_refs_total_length + 1))
0.031 feature.(ptwiki.revision.non_cite_templates / max(wikitext.revision.content_chars, 1))
0.031 feature.((ptwiki.revision.shortened_footnote_templates + wikitext.revision.ref_tags) / max(wikitext.revision.content_chars, 1))
0.027 feature.(wikitext.revision.ref_tags / max(wikitext.revision.content_chars, 1))
0.027 feature.(portuguese.stemmed.revision.stems_length / max(wikitext.revision.content_chars, 1))
0.027 feature.(((((ptwiki.revision.image_links + ptwiki.revision.image_template) + ptwiki.revision.images_in_templates) + (ptwiki.revision.side_by_side_image_templates * 2)) + ptwiki.revision.images_in_tags) + ptwiki.infobox_images)
0.026 feature.max(((ptwiki.revision.shortened_footnote_templates + wikitext.revision.ref_tags) - (ptwiki.revision.cite_templates + ptwiki.revision.shortened_footnote_templates)), 0)
0.026 feature.ptwiki.revision.non_cite_templates
0.024 feature.(wikitext.revision.headings_by_level(3) / max(wikitext.revision.content_chars, 1))
0.024 feature.wikitext.revision.external_links
0.022 feature.(wikitext.revision.external_links / max(wikitext.revision.content_chars, 1))
0.022 feature.((ptwiki.revision.cite_templates + ptwiki.revision.shortened_footnote_templates) / max(wikitext.revision.content_chars, 1))
0.021 feature.(max(((ptwiki.revision.shortened_footnote_templates + wikitext.revision.ref_tags) - (ptwiki.revision.cite_templates + ptwiki.revision.shortened_footnote_templates)), 0) / max(wikitext.revision.content_chars, 1))
0.017 feature.((ptwiki.revision.cite_templates + ptwiki.revision.shortened_footnote_templates) / max((ptwiki.revision.shortened_footnote_templates + wikitext.revision.ref_tags), 1))
0.01 feature.wikitext.revision.headings_by_level(3)
0.01 feature.ptwiki.revision.category_links
0.008 feature.ptwiki.revision.infobox_templates
0.005 feature.(ptwiki.main_article_templates / max(wikitext.revision.content_chars, 1))
0.004 feature.(ptwiki.revision.cn_templates / max(wikitext.revision.content_chars, 1))
0.003 feature.ptwiki.main_article_templates
0.002 feature.(ptwiki.revision.cn_templates + 1)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment