end-to-end fine-tuning

Question

end-to-end fine-tuning

Closed this issue 3 years ago · 4 comments

@lidq92 Thanks for your reply. When I train the network end to end ,that means I will train the Resnet-50,overfifting will occur,SROCC is 0.1.But if I just only train the regression net, the SROCC is 0.6. What should I do?Thank you very much!!

Answer 1 · 2021-03-23T14:45:15.000Z

@ciwei123 Please read the paper and follow the README. I did not get what you said both in the title and the comment.

Answer 2 · 2021-03-23T14:50:51.000Z

@ciwei123 Please change your issue's title to the correct one. I tried to understand your question, and if my understanding is correct, then one suggestion for end-to-end fine-tuning is using a smaller learning rate for the pretrained ResNet-50's weights.

Answer 3 · 2021-03-23T16:50:31.000Z

@lidq92 Thanks for your reply. I will try the method you said. I am sorry for my carelessness about the title.Thank you very much.

Answer 4 · 2021-03-24T09:30:39.000Z

@ciwei123 Usually 1/10*lr for the ResNet-50 backbone's weights. See lidq92/LinearityIQA and the corresponding paper's Figure 5(c).