search

georgian-io / Multimodal-Toolkit

Multimodal model for text and tabular data with HuggingFace transformers as building block for text data
https://multimodal-toolkit.readthedocs.io
Apache License 2.0
591 stars 85 forks source link

Support for multilabel classificaiton #8

Open neillbyrne opened 3 years ago

neillbyrne commented 3 years ago

Hi, thanks for assembling this project. I was wondering if there were any plans to incorporate multi-label classification. The labels API on the forward pass would be ideally changed from a single dim tensor (batch_size) of correct class indexes to a 2 dim tensor of (batch_size, n_labels) with binary indicators to remain consistent. Thanks again

senaerdogan commented 2 years ago

Hello, is there any news on the multilabel classification problem? Also, @neillbyrne have you managed to solve your problem? I have a multilabel text classification problem but I could not manage to use this project :(

jrivd commented 1 year ago

Hi there. I find this toolkit awesome. Did anyone get to try multilabel here anyhow? Thanks!

camilleduchesne commented 1 year ago

Hi there! I'm still experiencing issues with multilabel classification. Even with the binary classification example you presented in your google collab, at prediction time, the model breaks. Does anyone know how to solve this? Thanks so much!

Capture d’écran, le 2023-03-21 à 16 55 20
apethree commented 1 year ago

Same problem I am facing. #40 created a bug.

jtfields commented 1 year ago

It sounds like multi-label classification has been added as a potential enhancement. Does anyone know if the Multimodal Toolkit supports multi-class classification?

akashsaravanan-georgian commented 1 year ago

@jtfields multi-class classification is supported. You can try it out with this dataset.

Multi-label classification is currently not supported and is not likely to be added soon.