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

Can I do prediction by setting "is_training==False" #15

Open
p1n0cch10 opened this issue Sep 19, 2020 · 0 comments
Open

Can I do prediction by setting "is_training==False" #15

p1n0cch10 opened this issue Sep 19, 2020 · 0 comments

Comments

@p1n0cch10
Copy link

p1n0cch10 commented Sep 19, 2020

Hi,

I am using the tf slim inception_v3 model. The flag is_training is set True when training.

I checked the code, and found that is_training is in arg_scope for the model.

I am confused that the calculating mechanics for training and prediction are different. I.e. the mean and the variance in batch normalization are calculated each batch when training, but the stored mean and variance are used when prediction.

How can I set the flags when prediction? Many thanks.

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