Sahil Singla

icon indicating a website link https://singlasahil14.github.io icon indicating an email [email protected]

icon company University of Maryland icon location College Park icon location I am a PhD student at the University of Maryland, College Park. My research interests are in robust and reliable machine learning.

Results 3 comments of Sahil Singla

Loss becomes nan after a while

What would be the exact changes in the code?

Random accuracy for a trained model when using load_state_dict

Update: I realized that is I remove `model.eval()` from the code, it works fine. So my guess is that batch_norm is not being handled correctly in the eval mode.

Random accuracy for a trained model when using load_state_dict

Update: Although without model.eval(), the accuracy numbers are much higher, they are still 4-5% below the original 76.46% numbers. So for some reason composer trained models do not show good...