Last active
December 20, 2016 07:31
-
-
Save halfak/a19f4d23df26151be2a57a42a6451caf to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from revscoring.features import wikitext | |
char_based = [ | |
wikitext.revision.chars, | |
wikitext.revision.whitespace_chars, | |
wikitext.revision.markup_chars, | |
wikitext.revision.cjk_chars, | |
wikitext.revision.entity_chars, | |
wikitext.revision.url_chars, | |
wikitext.revision.word_chars, | |
wikitext.revision.uppercase_word_chars, | |
wikitext.revision.punctuation_chars, | |
wikitext.revision.break_chars, | |
wikitext.revision.longest_repeated_char | |
] | |
token_based = [ | |
wikitext.revision.tokens, | |
wikitext.revision.numbers, | |
wikitext.revision.whitespaces, | |
wikitext.revision.markups, | |
wikitext.revision.cjks, | |
wikitext.revision.entities, | |
wikitext.revision.urls, | |
wikitext.revision.words, | |
wikitext.revision.uppercase_words, | |
wikitext.revision.punctuations, | |
wikitext.revision.breaks, | |
wikitext.revision.longest_token, | |
wikitext.revision.longest_word | |
] | |
parse_based = [ | |
wikitext.revision.content_chars, | |
wikitext.revision.headings, | |
wikitext.revision.external_links, | |
wikitext.revision.wikilinks, | |
wikitext.revision.tags, | |
wikitext.revision.ref_tags, | |
wikitext.revision.templates | |
] | |
char_normalized = [ | |
wikitext.revision.whitespace_chars / wikitext.revision.chars, | |
wikitext.revision.markup_chars / wikitext.revision.chars, | |
wikitext.revision.cjk_chars / wikitext.revision.chars, | |
wikitext.revision.entity_chars / wikitext.revision.chars, | |
wikitext.revision.url_chars / wikitext.revision.chars, | |
wikitext.revision.word_chars / wikitext.revision.chars, | |
wikitext.revision.uppercase_word_chars / wikitext.revision.chars, | |
wikitext.revision.punctuation_chars / wikitext.revision.chars, | |
wikitext.revision.break_chars / wikitext.revision.chars, | |
wikitext.revision.longest_repeated_char / wikitext.revision.chars, | |
] | |
token_normalized = [ | |
wikitext.revision.numbers / wikitext.revision.tokens, | |
wikitext.revision.whitespaces / wikitext.revision.tokens, | |
wikitext.revision.markups / wikitext.revision.tokens, | |
wikitext.revision.cjks / wikitext.revision.tokens, | |
wikitext.revision.entities / wikitext.revision.tokens, | |
wikitext.revision.urls / wikitext.revision.tokens, | |
wikitext.revision.words / wikitext.revision.tokens, | |
wikitext.revision.uppercase_words / wikitext.revision.tokens, | |
wikitext.revision.punctuations / wikitext.revision.tokens, | |
wikitext.revision.breaks / wikitext.revision.tokens, | |
wikitext.revision.longest_token / wikitext.revision.tokens, | |
wikitext.revision.longest_word / wikitext.revision.tokens | |
] | |
draft_quality = (char_based + token_based + parse_based + | |
char_normalized + token_normalized) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
>>> from pprint import pprint | |
>>> | |
>>> from revscoring.datasources import revision_oriented as ro | |
>>> from revscoring.dependencies import solve | |
>>> | |
>>> | |
>>> text = """ | |
... {{Infobox|foo}} | |
... I am an article | |
... | |
... == Header! == | |
... * list | |
... * items | |
... * [[stuff|and a link]]<ref>some stuff</ref> | |
... """ | |
>>> pprint(list(zip(draft_quality, solve(draft_quality, cache={ro.revision.text: text})))) | |
[(<feature.wikitext.revision.chars>, 107), | |
(<feature.wikitext.revision.whitespace_chars>, 17), | |
(<feature.wikitext.revision.markup_chars>, 12), | |
(<feature.wikitext.revision.cjk_chars>, 0), | |
(<feature.wikitext.revision.entity_chars>, 0), | |
(<feature.wikitext.revision.url_chars>, 0), | |
(<feature.wikitext.revision.word_chars>, 59), | |
(<feature.wikitext.revision.uppercase_word_chars>, 0), | |
(<feature.wikitext.revision.punctuation_chars>, 1), | |
(<feature.wikitext.revision.break_chars>, 2), | |
(<feature.wikitext.revision.longest_repeated_char>, 2), | |
(<feature.len(<datasource.tokenized(datasource.revision.text)>)>, 47), | |
(<feature.len(<datasource.wikitext.revision.numbers>)>, 0), | |
(<feature.len(<datasource.wikitext.revision.whitespaces>)>, 17), | |
(<feature.len(<datasource.wikitext.revision.markups>)>, 6), | |
(<feature.len(<datasource.wikitext.revision.cjks>)>, 0), | |
(<feature.len(<datasource.wikitext.revision.entities>)>, 0), | |
(<feature.len(<datasource.wikitext.revision.urls>)>, 0), | |
(<feature.len(<datasource.wikitext.revision.words>)>, 15), | |
(<feature.len(<datasource.wikitext.revision.uppercase_words>)>, 0), | |
(<feature.len(<datasource.wikitext.revision.punctuations>)>, 1), | |
(<feature.len(<datasource.wikitext.revision.breaks>)>, 1), | |
(<feature.max(<datasource.map(<built-in function len>, <datasource.tokenized(datasource.revision.text)>)>)>, 7), | |
(<feature.max(<datasource.map(<built-in function len>, <datasource.wikitext.revision.words>)>)>, 7), | |
(<feature.wikitext.revision.content_chars>, 61), | |
(<feature.wikitext.revision.headings>, 1), | |
(<feature.wikitext.revision.external_links>, 0), | |
(<feature.wikitext.revision.wikilinks>, 1), | |
(<feature.wikitext.revision.tags>, 4), | |
(<feature.wikitext.revision.ref_tags>, 1), | |
(<feature.wikitext.revision.templates>, 1), | |
(<feature.(wikitext.revision.whitespace_chars / wikitext.revision.chars)>, 0.1588785046728972), | |
(<feature.(wikitext.revision.markup_chars / wikitext.revision.chars)>, 0.11214953271028037), | |
(<feature.(wikitext.revision.cjk_chars / wikitext.revision.chars)>, 0.0), | |
(<feature.(wikitext.revision.entity_chars / wikitext.revision.chars)>, 0.0), | |
(<feature.(wikitext.revision.url_chars / wikitext.revision.chars)>, 0.0), | |
(<feature.(wikitext.revision.word_chars / wikitext.revision.chars)>, 0.5514018691588785), | |
(<feature.(wikitext.revision.uppercase_word_chars / wikitext.revision.chars)>, 0.0), | |
(<feature.(wikitext.revision.punctuation_chars / wikitext.revision.chars)>, 0.009345794392523364), | |
(<feature.(wikitext.revision.break_chars / wikitext.revision.chars)>, 0.018691588785046728), | |
(<feature.(wikitext.revision.longest_repeated_char / wikitext.revision.chars)>, 0.018691588785046728), | |
(<feature.(len(<datasource.wikitext.revision.numbers>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.0), | |
(<feature.(len(<datasource.wikitext.revision.whitespaces>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.3617021276595745), | |
(<feature.(len(<datasource.wikitext.revision.markups>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.1276595744680851), | |
(<feature.(len(<datasource.wikitext.revision.cjks>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.0), | |
(<feature.(len(<datasource.wikitext.revision.entities>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.0), | |
(<feature.(len(<datasource.wikitext.revision.urls>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.0), | |
(<feature.(len(<datasource.wikitext.revision.words>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.3191489361702128), | |
(<feature.(len(<datasource.wikitext.revision.uppercase_words>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.0), | |
(<feature.(len(<datasource.wikitext.revision.punctuations>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.02127659574468085), | |
(<feature.(len(<datasource.wikitext.revision.breaks>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.02127659574468085), | |
(<feature.(max(<datasource.map(<built-in function len>, <datasource.tokenized(datasource.revision.text)>)>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.14893617021276595), | |
(<feature.(max(<datasource.map(<built-in function len>, <datasource.wikitext.revision.words>)>) / len(<datasource.tokenized(datasource.revision.text)>))>, 0.14893617021276595)] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment