Hello!
In box 31, I used fit_tokenizer as defined above (had expected output) to fit the labels, but somehow there is an extra term with OOV.
Vocabulary of labels looks like this {’’: 1, ‘sport’: 2, ‘business’: 3, ‘politics’: 4, ‘tech’: 5, ‘entertainment’: 6}
How can I get rid of it?
Thanks!
Could you please send your notebook privately so I can see what’s going on?
Hi thanks for your reply and offer to look at the code!
I fixed it by replacing fit_tokenizer (as previous defined) by fit_on_texts for the labels and the problem seems to be solved.
Apologies for not sending my code privately- I am not sure how to send private messages yet. 