Lizhaoqing123
commented
1 year ago Thanks for the simple and elegant implementation! I tried running your code as is, on Multi-MNIST data, and failed to reproduce results. I ran main_multi_mnist.py without changing any hyper parameter (learning rate (0.0005) , batch size (256), number of epochs (100)). For comparison, I created a version with no pcgrad: 1. comment out line 57: ## optimizer = PCGrad(optimizer) 2. replace line 72: optimizer.pc_backward(losses) -> torch.sum(torch.stack(losses)).backward()
I run each version 7 times. my results (averaging left-digit and right-digit accuracy) are:
Without PCGrad: average accuracy 89.5%, max accuracy 89.9%, standard deviation 0.38 With PCGrad: average accuracy 89.5%, max accuracy 89.8%, standard deviation 0.20
Can you come up with an explanation? Many thanks, Noa Garnett
My result is same as yours, did you finally solve this problem?
NoaGarnettDia
yangmin666
NoaGarnett
Thanks for the simple and elegant implementation! I tried running your code as is, on Multi-MNIST data, and failed to reproduce results. I ran main_multi_mnist.py without changing any hyper parameter (learning rate (0.0005) , batch size (256), number of epochs (100)). For comparison, I created a version with no pcgrad:
I run each version 7 times. my results (averaging left-digit and right-digit accuracy) are:
Without PCGrad: average accuracy 89.5%, max accuracy 89.9%, standard deviation 0.38 With PCGrad: average accuracy 89.5%, max accuracy 89.8%, standard deviation 0.20
Can you come up with an explanation? Many thanks, Noa Garnett