Closed sgmath12 closed 5 years ago
In common setting, we normalize the data something like this (img - img.mean)/(img.std)
But in your code, you didn't divide std.
Is there any reason for that?
Because vgg16 doesn't, which is used as a pretrained-model for feature extraction
In common setting, we normalize the data something like this (img - img.mean)/(img.std)
But in your code, you didn't divide std.
Is there any reason for that?