Gamma, Beta in the model weight
Why the value of gamma_s3, gamma_s2, beta_s3, beta_s2 are all zeros in your provided model weights? If they are all zeros, meaning that they are not functional, right?
Hello, gamma and beta are normalization factors, which is initialized as 0 and gradually learns a weight.
I mean in your provided trained weights, they are still zeros.
This is a little strange and I will check it later.
The gradients of gamma_s3, gamma_s2, beta_s3, beta_s2 are None. Is that expected?
The gradients of
gamma_s3, gamma_s2, beta_s3, beta_s2areNone. Is that expected?
I think these parameters are not updated during training, and the parameters from the provided trained weights are all zeros.
I also found this problem :the gradients of gamma_s3, gamma_s2, beta_s3, beta_s2 are None ,and they are not updated during training.it will lead to poor training results