TokenClassification vs SequenceClassification

Hi, can anyone simply explain that what is the difference between these two? and which one is better for Multi-Label classification tasks?

1 Like

I really need to know. anyone please?

Token classification refers to the classifications of tokens in a squence. So for example you assign classes to words in a sentece.
In sequence classification you’re classifying the whole sequence, for example assigning a class to a sentence.
They are different tasks, hence there is no “being better for multi-label classification”.

4 Likes

Thanks a lot. very simple and useful.