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

How to depict Fig.3 in your paper? #11

Closed
youngwanLEE opened this issue Mar 5, 2018 · 3 comments
Closed

How to depict Fig.3 in your paper? #11

youngwanLEE opened this issue Mar 5, 2018 · 3 comments

Comments

@youngwanLEE
Copy link

Hi,
I'm really impressed by your good work.

As shown Fig3, I want to depict the effective receptive field like you.

Could you share the method?

@GOATmessi8
Copy link
Owner

@youngwanLEE I use the same gradient bp method in "Understanding the effective receptive field
in deep convolutional neural networks".

@youngwanLEE
Copy link
Author

@ruinmessi Could you share the program?

@GOATmessi8
Copy link
Owner

@youngwanLEE That is a messy and tiny scripts. What I do is that forwarding one picture, then define a zero grad tensor except the center. Using "output.backward(Zero_grad)" to bp, and the effective rf data is "input.grad.data.cpu()", where input and output are the Variables of your module or network.

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