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 "07-CSRNet_all_ep_39_mae_32.6.pth" #119

Open
JayzWho opened this issue Oct 15, 2023 · 0 comments
Open

Question about "07-CSRNet_all_ep_39_mae_32.6.pth" #119

JayzWho opened this issue Oct 15, 2023 · 0 comments

Comments

@JayzWho
Copy link

JayzWho commented Oct 15, 2023

Excuse me, I have a problem that is recently I've been going through some old codes of a senior student and I came cross settings of "07-CSRNet_all_ep_39_mae_32.6.pth" which I haven't found. I tried to replace it with our trained model of CSRNet on SHTech-B which is a "PartBBest.pth.tar". The results were that the pre-train model had some mis-matched factors and it failed when loading with
"RuntimeError: Error(s) in loading state_dict for WeakNetwork: Missing key(s) in state_dict: "frontend.0.weight", "frontend.0.bias", "frontend.2.weight", "frontend.2.bias", "frontend.5.weight", "frontend.5.bias", "frontend.7.weight", "frontend.7.bias", "frontend.10.weight", "frontend.10.bias", "frontend.12.weight", "frontend.12.bias", "frontend.14.weight", "frontend.14.bias", "frontend.17.weight", "frontend.17.bias", "frontend.19.weight", "frontend.19.bias", "frontend.21.weight", "frontend.21.bias", "backend.0.weight", "backend.0.bias", "backend.2.weight", "backend.2.bias", "backend.4.weight", "backend.4.bias", "backend.6.weight", "backend.6.bias", "backend.8.weight", "backend.8.bias", "backend.10.weight", "backend.10.bias", "output.weight", "output.bias". Unexpected key(s) in state_dict: ""."

My question is: First, where can I get the "07-CSRNet_all_ep_39_mae_32.6.pth" file? Is it an open source file that can be directly downloaded or is it a result of running through a model?
Second, I think our own "pth.tar" was supposed to work, since it's just a pre-trained CSRNet that's required. Are the runtime error of missing factors not its fault? If so, how can I make changes to solve this mismatch? Thank you.

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

1 participant