Levigty/AimCLR

How to fine-tune

Opened this issue · 1 comments

How to fine-tune the model after the linear evaluation is completed? I don't see the corresponding yaml file in the project

tht106 commented

Fine-tune and linear evaluation are different evaluation methods from my understanding, where the fine-tune method retrains the whole model while the linear evaluation method only got the liner layer fine-tuned but got the encoder part fixed.