I try to create a neural network with a long-term short-term memory layer to classify the toxic comments for the Kaggle competition . In the original database we have in train: ...
['id', 'comment_text', 'toxic', 'severe_toxic', 'obscene', 'threat', 'insult', 'identity_hate']
['0000997932d777bf', "Explanation\nWhy the edits made under my username Hardcore Metallica Fan were reverted? They weren't vandalisms, just closure on some GAs after I voted at New York Dolls FAC. And please don't remove the template from the talk page since I'm retired now.89.205.38.27", '0', '0', '0', '0', '0', '0']
...
However, we can use other databases for the competition and I wanted to know how we can use them. For example, I want to use the Youtube database . But how can I use it to transform in the same way as the original database? ?