jinfengr

FYI: the implementations of all models are available now.

It seems the issue is fixed with the latest code push. Please add comment if the issue still comes up.

Thanks! It seems gdrive app is not able to use now.

@klintan 's solution seems work but need additional tweak. You need make following changes in basic/model.py: tf.concat(0, [word_emb_mat, self.new_emb_mat]) -> tf.concat([word_emb_mat, self.new_emb_mat], 0) Update all other concat functions as well.

The same question here, I also tried NLLLoss()