udacity/deep-learning-v2-pytorch

Normalization

saikrishnadas opened this issue · 2 comments

Each color channel was normalized separately, the means are [0.485, 0.456, 0.406] and the standard deviations are [0.229, 0.224, 0.225].

How did you find those normalization values and can i use them in all cases other than cat and dog ?

These are Imagenet's mean and std. Check this answer https://stackoverflow.com/a/57533806/6210807

Though we are a little late to respond, we suggest you reach out to Udacity support for user-specific issues.
Thanks