search

mim-solutions / bert_for_longer_texts

BERT classification model for processing texts longer than 512 tokens. Text is first divided into smaller chunks and after feeding them to BERT, intermediate results are pooled. The implementation allows fine-tuning.
Other
129 stars 30 forks source link

multilabel classification #50

Open edwarddjs opened 5 days ago

edwarddjs commented 5 days ago

I believe in the readme you refer to multilabel classification instead of multi class however reviewing the code it appears that the output layer in linear and therefore multi class classification is happening

mwachnicki commented 3 days ago

What is happening is definitely multi class classification. However, it is indeed described as multilabel classification in the readme. Thank you for pointing this out!

mwachnicki commented 3 days ago

For now, I'm adding enhancement label to correct this.

edwarddjs commented 3 days ago

No problem glade I could help.