Validation & Test shouldn't randomly crop the input image

[nitinsurya]

I haven't read the paper yet, but looking into the code, random cropping on the validation/test set doesn't seem like a proper evaluation method, as random crop is usually done to avoid model overfitting on training data.

[nitinsurya]

Came across this RandomResizedCrop, which would probably yield even better results.

[huanmofazong]

Yes, Ramdomcrop should be removed here when in test and val , otherwise randomness will be generated when test or val