sakshamj74 / news_classification

To classify a headline into three categories namely Business,technology and Sport

Heroku app link-https://classify-news.herokuapp.com/

News_classification

To classify a headline into three categories namely Business,technology and Sport news_scrapping file is used to crate our own dataset news_clssification is the main file where various models,vectorisation,encodings are performed. To join various files i have used an online text file joiner.Here is the link https://www.filesmerge.com/merge-text-files final_dataset is the final dataset after scrapping and joining.

Deployment Using Flask To Heroku

• app1.py is the main flask app file
• model.pkl is the saved trained model
• tf1.pkl cantains the saved vocubalary for the trained model

Technologies Used

• Flask
• Pyhton
• NLP
• Heroku
• scikit-learn

news_classification's Issues

Need Help in deploy an app on Heroku

Hey Sir! can you please guide me
how can i make tfidf model like 'tf1.pkl'
I want to deploy my Flask app on Heroku
but while deployment i found internal server error because of these line
nlp = spacy.load('en_core_web_sm') #use for tokeinzation and remove stopwords from dataset stopwords = list(STOP_WORDS) punct = string.punctuation def text_data_cleaning(sentence): doc = nlp(sentence) tokens = [] for token in doc: if token.lemma_ != '-PRON-': temp = token.lemma_.lower().strip() else: temp = token.lower_ tokens.append(temp) cleaned_tokens = [] for token in tokens: if token not in stopwords and token not in punct: cleaned_tokens.append(token) return cleaned_tokens tfidf = TfidfVectorizer(tokenizer=text_data_cleaning)

And after running command heroku logs it show me the error of
AttributeError: module '__main__' has no attribute 'text_data_cleaning'
please sir guide me....
Thanks in advance...!!!