Yonglong Tian
Yonglong Tian
I like the "Top Recent" function a lot. Seems I did not find an equivalent in arxiv sanity lite.
> Hi, I will work on the export of previous database in the next few days. I will plan to also add the "Top Recent" function. I didn't add it...
Hi, @ildoonet , Thanks for contributing this nice code! I wonder if the performance gap is related to two possible misalignments? (a) Seems you put RandAugment before RandomResizedCrop. If I...
Anyone still looks at this? @diweiqiang @yassouali @yrchen92 I think, from the perspective of gradients: ``` loss = -(torch.mean(t) - torch.mean(et) / ma_et.detach()) ``` and ``` loss = -(torch.mean(t) -...
I did not perform that experiments so not very sure about that detail, but that table will be further updated in a later version.
At least you need to modify the input-output format
Sry to hear it. I did not save the checkpoints for CIFAR datasets. Will try to upload a new one once I rerun it. Or if anyone else who see...
If not, you need to modify it accordingly. But the logic should not be hard.
I am not sure if I understand correctly. But if both vectors are normalized, the absolute `l2` euclidean distance is the same as cosine distance. On the other hand, if...
Would the comment here: https://github.com/HobbitLong/SupContrast/issues/64#issuecomment-1182845137 help answer the question? For the given example above, there will also be terms 2 & 3 that encourages larger A•A2 and A•A3 in the...