users = separated_ratings.userId.unique()
movies = separated_ratings.movieId.unique()
rating = separated_ratings.rating.unique()
print("Users: ", len(users), ", Movies: ", len(movies), ", Ratings: ",
len(rating))
userid2idx = {o: i for i, o in enumerate(users)}
movieid2idx = {o: i for i, o in enumerate(movies)}
movieidx2id = {i: o for i, o in enumerate(movies)}
new_ratings = separated_ratings.copy()
new_ratings.movieId = separated_ratings.movieId.apply(lambda x: movieid2idx[x])
new_ratings.userId = separated_ratings.userId.apply(lambda x: userid2idx[x])
## split data into training and validation set, 80% in training, 20% in validation
msk = np.random.rand(len(new_ratings)) < 0.8
train = new_ratings[msk]
valid = new_ratings[~msk]
print("training data: ", train.shape, ",validation data: ", valid.shape)
#neural_net.baseline(train, valid, users, movies)
#neural_net.multilayer_LSTM(train, valid, users, movies)
predicted_ratings = neural_net.bidirectional_LSTM(train, valid, users, movies,
separated_sample_submission)
#neural_net.bidirectional_LSTM_2(train, valid, users, movies)
reader_writer.write_to_file(predicted_ratings, 'sample_submission.csv',
'predicted_ratings.csv')
