wzhouad / ATLOP

Source code for paper "Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling", AAAI 2021
197 stars 39 forks source link

The main purpose of the function: get_label #16

Closed vhientran closed 3 years ago

vhientran commented 3 years ago

Hi @wzhouad ,

Thanks so much for releasing your source code. I only wonder about the main purpose of the function get_label() in the file losses.py in calculating the final loss. Could you please explain it? Thanks for your help!

wzhouad commented 3 years ago

Hi,

The function is for transforming class probabilities to predicted labels. It is used only in inference.

vhientran commented 3 years ago

Hi @wzhouad ,

I got it. Thank you very much!

vhientran commented 3 years ago

Hi @wzhouad ,

Sorry for disturbing you, but I just have one more question. In the training phase, the parameters of the pre-trained BERT will be fine-tuned or fixed? Thank you!

wzhouad commented 3 years ago

They are finetuned.

vhientran commented 3 years ago

Many thanks!