Skip to content
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

Question about the OS-block #17

Open
ShiinoMashiro opened this issue Dec 17, 2022 · 1 comment
Open

Question about the OS-block #17

ShiinoMashiro opened this issue Dec 17, 2022 · 1 comment

Comments

@ShiinoMashiro
Copy link

Hello,

I have read your paper and i have a few questions listed below about the OS-block part:

Q1. "the first two layers of OS-block have prime-sized kernels from 1 to pk", why it is size of kernels, not size of receptive fileds?

Q2. "the two layers can cover all even number receptive field sizes", does this mean the RF is calculated just by adding RF of layer1 and layer2 , will the batchnorm or relu part have influenced the size of RF?

I am looking forward to your reply.
Thanks!

@Wensi-Tang
Copy link
Owner

Hello, thanks for your attention!
Q1: For one layer, the receptive field sizes are the same as the kernel sizes. But for multiple inception layers, the receptive field sizes will be the combination of kernel sizes in each layer.
Q2: Based on the definition (search "size of the region in the input that produces the feature") of RF, the batch norm and ReLu will not influence the size.

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

No branches or pull requests

2 participants