Sense-GVT / DeCLIP

Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm
622 stars 31 forks source link

Detailed Preprocess #3

Closed fiona-lxd closed 2 years ago

fiona-lxd commented 2 years ago

What is the preprocess like ? Such as the resize shape, mean and std

zlccccc commented 2 years ago

The testing data preprocessing can be seen in [https://github.com/Sense-GVT/DeCLIP/blob/main/prototype/data/imagenet_dataloader.py#L105].