-
Notifications
You must be signed in to change notification settings - Fork 10
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Results not same as mentioned in paper #1
Comments
Hi, Thanks JamesChuanggg for sharing San code in the torch. Here is the training log, which I got after so many iterations,validation loss: 17.691484603882 accuracy 0.268 |
Hi,
Thanks for sharing the code. I ran the code with default settings and only got 45% accuracy (in the paper authors have mentioned over 57% accuracy). I noticed that the optim method was set to 'rmsprop' by default, I changed it to 'sgd' since in the paper they have used sgd, and the accuracy dropped to 22%.
I was wondering if you've managed to get accuracy better than this, if yes, could you share your settings, optim method, learning rate, no of iters, etc.
The text was updated successfully, but these errors were encountered: