Normalization seems to be unnecessary

[greentfrapp]

Hi there! Thanks so much for this! While playing around with the model, I realized that the initial normalization step seems to be unnecessary, since the model already has InstanceNormalization layers.

I tried running a few tests and it seems to give the same hash regardless of whether we normalize the inputs.

[AsuharietYgvar]

TBH I don't know how the layers work. But in Apple's implementation input images are normalized. Probably it's because Vision framework does it anyway.