search

dccuchile / beto

BETO - Spanish version of the BERT model
Creative Commons Attribution 4.0 International
492 stars 63 forks source link

Finetuning on custom spanish dataset #19

Closed anirindg closed 3 years ago

anirindg commented 3 years ago

I have a custom spanish dataset which I want to classify into specific labels. However, in the repo, there are no specific instructions ( apart from the weight files ), on how to do that ( specifically, using TF2 ). Can you be more specific ?

Smolky commented 3 years ago

Hello. In case it helps, I address this problem fine-tunning the model using HuggingFace https://huggingface.co/transformers/training.html

Particularly, using the Trainer class is quite straightforward.

josecannete commented 3 years ago

Hello @anirindg, thank for your interest in the project.

The documentation of the Transformers library from HuggingFace is really good, with lots of examples. I encourage you to explore it.

The link provided by @Smolky is probably what you need. This other one might be helpful too: https://huggingface.co/transformers/custom_datasets.html

I hope it helps!

anirindg commented 3 years ago

Thank you , I referred to the same and it is working.