Chen-Ping Yu
Chen-Ping Yu
just tried the provided vanilla vgg, training does not converge at all, the training loss after 4 epochs is still at 6.9.
yes I'm also getting a similar issue, alexnetowtbn is giving me low accuracy, trying to train with -netType alexnet to see if at least alexnet gives good performances...
hmm, tried it again and now alexnetowtbn converges fine, got to 38th epoch and the top-1 validation accuracy is at 53.93%.
hmm, I've updated torch, nn, cutorch, and cudnn. But my version of imagenet-multiGPU was from a few months ago, I've just cloned the new version and just began a training...
alexnetowbn trained and converged fine, btw.