Skip to content

Instantly share code, notes, and snippets.

What would you like to do?
利用 sklearn 计算文本 tf-idf
import shutil
import tempfile
from datetime import datetime, timedelta
import jieba
import pandas as pd
from django.conf import settings
from import BaseCommand
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from apps.articles.models import Article, get_stop_words
idf_path = getattr(settings, "IDF_PATH", None)
def jieba_token(value):
return jieba.cut(value, cut_all=False)
class Command(BaseCommand):
help = "gen article tfidf file"
docs = list()
def add_arguments(self, parser):
parser.add_argument("-d", "--delta", type=int, help="cal", default=30)
def handle(self, *args, **options):
days = options["delta"]
since = - timedelta(days=days)
for row in Article.objects(published_at__gte=since):
count_vect = CountVectorizer(tokenizer=jieba_token, min_df=1, stop_words=get_stop_words())
tfidf_transformer = TfidfTransformer(smooth_idf=True, use_idf=True)
x_counts = count_vect.fit_transform(
df = pd.DataFrame(tfidf_transformer.idf_, index=count_vect.get_feature_names(), columns=["idf_weights"])
fp = tempfile.NamedTemporaryFile()
for row in df.sort_values("idf_weights", ascending=False).itertuples():
_ = "%s %.9f\n" % (row[0], row[1])
shutil.copy(, dst=idf_path)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.