Closed jfb54 closed 3 years ago
Indeed, the transforms that I've put in my example don't scrupulously follow the ones from the original dataset. This should not make a big difference though. Still, I will modify to follow those at https://github.com/google-research/meta-dataset/blob/a38e8cea0f60e3aa1edb4424f5574c10e355f3d4/meta_dataset/data/decoder.py#L74
Awesome. Thanks!
Instead of the ImageNet friendly normalize transform:
use the following:
as Meta-Dataset guarantees that all images will be normalized in the range -1 to 1.