Weighted Contrastive Learning Loss

Data augmentation is an important paradigm for boosting the generalization capability of deep learning in image classification tasks. …