Pytorch dataloader transforms and keras 3.0 model training

In most pytorch training, one of the augmentations applied to both the train and validation set is Normalization especially when doing transfer learning like so for Imagenet dataset;

transforms.Normalization(
                mean=[0.485, 0.456, 0.406],
                std=[0.229, 0.224, 0.225]
            )

I want to know if it is a good idea to apply a similar transform when using pytorch dataloaders with keras 3.0 model architecture for training?

Hi @Atia, Thanks for posting this issue.

You can train a Keras 3 model with pytorch dataloader objects and you can try to use transform.Normalization but if it doesn’t work, you can use keras.layers.Normalization for similar results.

Let me know if that works for you.