the loss is nan

[cena001plus]

i have try different lr from 0.00001 to 0.1 , but thr loss is nan , why?

[LuoXubo]

same problem, have you solved it?

[GHGluck]

I also had the same problem.

[hli1221]

Hi, thank you very much for report this problem.
Normally, the pixel loss should be decreased with increasing iteration number. However, the pixel loss was increasing. I suggest you can check the input and the output, this problem may occur if they are not in the same domain.

[GHGluck]

Thank you for your reply. I tried to modify the input data and found that sometimes the training can be successful, but when I increase the data, the loss will still be none. I don't know if it is because the network cannot fit my training data.