loss_function_search

loss_function_search copied to clipboard

What great work! Thank you very much for publishing your code, and I got into trouble when reproducing your code. Could you explain the meaning of variables "p_bins" and "a"?...

lulongfei-luffy

Thanks for your outstanding work. After reading your paper, I carefully analyze your code. I found out that you used pytorch api function `prob = F.softmax(pred, dim=1)`. Based on my...

jianyuheng

Hello! There are two piecewise linear search spaces in AM-LFS, which are \t(x) and \tau(x). But in your article only \tau(x) is considered, why not consider \t(x)?

GuHongyang

Validation dataset split?

1

Hello! First of all, thank you very much for releasing the code. My question is, is the validation dataset split from the training data? Or directly use the test data...

GuHongyang

Loss parameters optimization

1

Hello! Thanks for your work, very interesting reformulation of margin-based softmax ! My question is why to use REINFORCE for searching of ```a```? Can it be optimized together with parameters...

mnikitin