-
Notifications
You must be signed in to change notification settings - Fork 5.5k
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
Problem of BatchNorm in Fluid. #9273
Comments
can you paste the desc info of |
The above ProgrameDesc is about the regularization since I set L2 regularization 0.00005. Although, the un-trainable parameters skip the parameter updating process in https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/fluid/optimizer.py#L202 . But there is no need to append these ops in ProgrameDesc. |
@jacquesqiao the desc for batch_norm_op:
The above descs for scale_op and elementwise_add_op I pasted are added by transpiler, not in the model config. |
Now the moving mean and variance in
batch_norm
are created as parameters and settrainable
False:https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/fluid/layers/nn.py#L1512
But when see the ProgrameDesc proto string, there is still some calculation operators related to the moving mean and variance. For example, I print the ProgrameDesc in MobileNet-SSD of one GPU, the following proto string is about the moving variance(
batch_norm_x.w_2
is the moving variance):The text was updated successfully, but these errors were encountered: