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

replace the convolution in Resnet with PConv #41

Open
wanderer1230 opened this issue Jun 2, 2023 · 1 comment
Open

replace the convolution in Resnet with PConv #41

wanderer1230 opened this issue Jun 2, 2023 · 1 comment

Comments

@wanderer1230
Copy link

Great work! I wonder if the model performance will be improved when replacing the convolution in Resnet with PConv.
Have you done similar experiments?

@JierunChen
Copy link
Owner

@wanderer1230 Hi, if the Conv is directly replaced by PConv without other modifications, the accuracy would drop. However, if the total FLOPs are aligned to be similar for the ResNet50 and ResNet50 w/ PConv (e.g., by increasing the network width), the ResNet50 w/ PConv would have higher accuracy (+0.6% in my experiment for 90 epochs' training).

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