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.

Sign up for GitHub

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

Jump to bottom

different accuracy under different batchsize in inference #86

Closed
Muran337287 opened this issue Jul 15, 2019 · 3 comments
Closed

different accuracy under different batchsize in inference #86

Muran337287 opened this issue Jul 15, 2019 · 3 comments

Comments

@Muran337287
Copy link

I have trained only ASRN model with SynthText, and test in ICDAR2015-2077, but I got the different accuracy if dataloader has different batchsize,
batch_size=16, acc=61.964, batch_size=1, acc=61.820
@Canjie-Luo
Copy link
Owner

Can you give more details? Please do not set batch_size to 1 when evaluating the model. It is not supported in this repo. Sorry.

@Muran337287
Copy link
Author

The code is similar to val in main.py

@Muran337287
Copy link
Author

Why can not batch_size be 1?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@Canjie-Luo @Muran337287