detail of experiment

[ctxya1207]

Thank you for your excellent work. I have a question. When using contrastive loss pre training AesNet, do you load the weights of the backbone network with ImageNet pre training?

[yipoh]

Thank you for your excellent work. I have a question. When using contrastive loss pre training AesNet, do you load the weights of the backbone network with ImageNet pre training?

Thanks for your question. For better performance, we loaded the original ImageNet weights before contrastive pre-training.

[ctxya1207]

Thank you for your excellent work. I have a question. When using contrastive loss pre training AesNet, do you load the weights of the backbone network with ImageNet pre training?

Thanks for your question. For better performance, we loaded the original ImageNet weights before contrastive pre-training.

Thanks for your answer, can you provide the pre-training log, I want to understand the downward trend of the loss

[ctxya1207]

What is the temperature hyperparameter τ and epoch?,thank you.