Created
October 15, 2019 12:10
-
-
Save morkapronczay/93ae57d473a5093b4e35efaa8eae4780 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 gensim import corpora | |
# create stemmed, stopword removed corpus | |
# by language by doc (wiki page) | |
texts_bylang_byhuman = {lan: | |
{key: | |
[stemmers[lan].stem(word) | |
for word in val if not word in stopwords_bylang[lan]] | |
for key, val in texts_split[lan].items()} | |
for lan in languages} | |
# create dictionaries by language | |
dictionary_bylang_byhuman = {lan: corpora.Dictionary(texts_bylang_byhuman[lan].values()) for lan in languages} | |
# create sparsity thresholds for percentages | |
SPARSE_TRESH = [0.5, 1, 5, 10] | |
sparse_perc = {k: math.ceil(k * len(texts['en'].keys()) / 100) for k in SPARSE_TRESH} | |
# remove sparse tokens | |
filtered_dicts_sparse = {} | |
for lan in languages: | |
filtered_dicts_sparse[lan] = {} | |
for k, v in sparse_perc.items(): | |
# the method effects the dictionary itself! | |
# so copy, if you want to keep the initial | |
filtered_dicts_sparse[lan][k] = copy.deepcopy(dictionary_bylang_byhuman[lan]) | |
filtered_dicts_sparse[lan][k].filter_extremes(no_below=v, no_above=1, keep_n=len(dictionary_bylang_byhuman[lan])) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment