bert-sentiment
bert-sentiment copied to clipboard
multi-label classification problem
how can this model be trained for a multi-label classification problem?
I am also interested in knowing how we can do multi-label classification through BERT. Does it depend on the 'out' attribute in the BERTBaseUncased Class? There is a hint in the video for this repo about changing loss functions for multi-label classification. Would like to know more about from @abhishekkrthakur .
do you have a sample dataset?
@abhishekkrthakur https://www.kaggle.com/c/jigsaw-toxic-comment-classification-challenge/data with different types of toxicity. How can we train a model for this type of problem? Thanks in advance.