Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Bug in building RegNetX #5

Open
sefibk opened this issue Jun 23, 2021 · 0 comments
Open

Bug in building RegNetX #5

sefibk opened this issue Jun 23, 2021 · 0 comments

Comments

@sefibk
Copy link

sefibk commented Jun 23, 2021

Hi,
Thanks for a GREAT repo!
I think there might be a bug in the creation of RegnetX here:

ls_group_width = ls_group_width.astype(np.int) * bottleneck_ratio

Why would you multiply the group_width by the bottleneck_ratio?

I will demonstrate through an example:

group_width = 16
block_width = 32
bottleneck_ratio = 2

With these set of parameters I would assume a bottleneck block will be created with 1/2 the channels in the bottleneck and 1 group convolution (i.e. standard convolution)
However: l.25 changes the groups to ls_group_width = ls_group_width.astype(np.int) * bottleneck_ratio => group_width = 32 making this block impossible and having the model FAIL!

Is this intentional or a bug?

Thank you very much

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant