About Normalization using pre-trained vgg16 networks

[ptrblck]

@Ashima_Garg

Wouldn’t it be computationally expensive to store latent tensors of the entire dataset?

It depends how large the dataset is and how large each latent tensor is.
If you cannot store all tensors during training, you would have to calculate the stats on the fly.

@Ashima_Garg

Can you give a code snippet of the same.

Here is an example of using forward hooks.