webeng / feature_engineering Goto Github PK

NLTK path as a environment variable

File "link.py", line 137, in
l = l.extract('https://www.wired.com/2017/05/google-just-made-email-heckuva-lot-easier-deal/')
File "link.py", line 74, in extract
keywords = k.extract([text], None, None, 'news')[0]
File "C:\Users\dinom\Desktop\feature_engineering-master\features\keywords.py", line 187, in extract
vectorizer = self.train_tfidf(tokenizer, tfidf_corpus)
File "C:\Users\dinom\Desktop\feature_engineering-master\features\keywords.py", line 71, in train_tfidf
nltk_corpus += self.get_bbc_news_corpus()
File "C:\Users\dinom\Desktop\feature_engineering-master\features\keywords.py", line 51, in get_bbc_news_corpus
onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]
WindowsError: [Error 3] The system cannot find the path specified: './data/bbc/business/.'

this is the error i am getting