Have you normalize the input image to [-1,1]

[mlszy928]

I trained the model with medical images normalized to [0,1], and have achieved satisfactory results. I'd like to ask if it is necessary to normlize the image to [-1,1]?

[sangyun884]

I haven't thought about that, but my intuition is that it shouldn't matter so much because neural networks can learn such simple affine transformations in the first layer if needed.