孙浩然

icon company THU icon location Beijing icon location 怕什么真理无穷，进一寸有进一寸的欢喜

Results 1 comments of 孙浩然

Train_loss and test_loss become NaN after using model.prune()

I have the same problem. I also use KANs to do five classification tasks, but after I complete the first training stage and run ```model = model.prune()```, I get **nan**...