Closed e96031413 closed 3 years ago
Hi, we did not try semi-supervised imbalanced learning on ImageNet-LT. The main reason is that we do not have a larger extra unlabeled dataset for it. FYI, CIFAR-10 has the TinyImages as the extra unlabeled dataset, as we descirbed in the paper.
I guess one way to include more unlabeled data could be using the unused data in the original ImageNet (since ImageNet-LT uses only a subset of ImageNet).
OK, Thanks for your reply
Hi, have you ever tried "Semi-Supervised Imbalanced Learning" on ImageNet-LT?
According to the experiment result in the paper, the performance with Semi-Supervised Imbalanced Learning seems better than Self-Supervised Imbalanced Learning on CIFAR-10-LT.
If I want to try this experiment, how can I modify the dataset/imagenet.py to dataset/imblance_imagenet.py (similar to imblance_cifar.py)?