Document prediction

[iamrahulrnair]

@yao8839836 Thanks for this amazing work, we are planning to do some research on this area. I can see that the model is not saved here. Our objective is to save the trained model, and if suppose a new document for example in case of movie reviews, a new review comes in, the trained model should output the corresponding label associated with it (0,1) , can you help us to approach this problem since we can't figure out within train.py how to supply new data to the model.