Input normalization

[MathiasGilson]

why do you normalize the input with this function ?

https://github.com/aitorzip/Keras-ICNet/blob/bcc13bf5dd39abd3496dd669b245b91c31d7ff16/model.py#L16

It looks like an input value of 255 (let’s say in the green channel) would be normalized to 0.5 instead of 1 in that same channel. I don’t understand why, can you help me ?

[pgr2015]

@MathiasGilson I guess the reason is because he want to rescale to range from 0-255 to -0.5-+0.5. The intuition behind this is similar to batchnormalization, which helps training to converge faster due to normalized linear projection.