-
-
Notifications
You must be signed in to change notification settings - Fork 65
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
Tweak GoogLeNet to match the torchvision implementations #205
Conversation
The |
We have
I'd tried this during the refactor, but it added too much clutter to the docstring and I had to handle other norm layers manually (although for |
I think the rational here is that conv will have keywords customized more than batch norm, so it gets preference for pass-through keywords. This avoids having different subsets of keywords going to different layers which I think is confusing. The normalization layer then has to use the slightly more verbose syntax of a closure: Though as Abhirath mentioned, using |
Would doing this work? - Because |
I didn't quite follow. |
So you are suggesting tweaking the
|
Yep, that seems perfect! |
Thank you for all the help @theabhirath! I had one more question though, the current implementation of GoogLeNet does not have |
Sure, go ahead! The closer we are to paper parity the better 😄 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM! Could you just run the formatter locally once and commit so that the alignment/spacing for the code matches the rest of the repository? Otherwise everything seems great, thank you so much for the contribution!
Changed the
Conv
layers toconv_norm
layers.However, I couldn't find a way to specify eps for
BatchNorm
. The default value that flux uses is 1f-5, and torchvision version of GoogleNet has eps set to 0.001.Closes #196
PR Checklist
Also ps: Unrelated, but the link to 'contributing docs' in the readme seems to be broken.