Rnn Model Predicting Only One Class?

I am trying to use GloVe embeddings to train a rnn model based on this article. I have a labeled data: text(tweets) on one column, labels on another (hate, offensive or neither).

Solution 1:

How the distribution of your data looks like? The first suggestion is to stratify train/test split (here is the link for the documentation).

The second question is how much data do you have in comparison with the complexity of the model? Maybe, your model is so complex, that just do overfitting. You can use the command model.summary() to see the number of trainable parameters.