e-prop learning for MNIST classification.
Opened this issue · 0 comments
LiYuan-0709 commented
I tried to apply this PyTorch-implemented e-prop algorithm to the MNIST classification task. However, I only got an accuracy of 54%. I want to know what I should do to improve it for this digit classification task. Or, what I should do to make it adjust to a new task quickly because there are so many hyperparameters?