python3.7 or abrove install Keras, NLTK, and numpy;
Three models anf their training detail can be found in:
- Tfidf_train.py(model1): WordStemming, stopwords filter and tfidf Vectorizer(feature dimension set as 10000);
- Tfidf_train_feat3000.py(model3): WordStemming, stopwords filter and tfidf Vectorizer(feature dimension set as 3000);
- Bow_train.py(model2):BOW vectorized ( feature set as 10000);
TODO
Enter /IMDB_sentiment Analysis
TODO