-
Very nice work, is the follow up intended to provide pre-trained models on ImageNet-21k for easy comparison with previous work.
-
The base tagging of the original imagenet21K is single label. I wonder how to get multi-label information for each image in ImageNet22K.
SJLeo updated
4 months ago
-
> Image-based filtering. We select a subset of examples whose visual content overlaps with ImageNet
classes. After applying English language (fasttext) and caption length filtering, we cluster the
i…
-
Hi guys, congrats on your ICLR23-top25% paper. The 'severity' benchmark in your work is impressing. However, up to now I can only find your released dummy dataset in this repo. If you can kindly relea…
-
I am trying to train ViT_base_16 on imagenet 1K (224x224) from an imagenet21K pretrained weights.
Following the training recipe from the paper (batch=512, 20K steps, 500 warmup steps, lr=0.01, cosin…
-
As title
-
* Name of dataset: ImageNet21k
* URL of dataset: https://github.com/Alibaba-MIIL/ImageNet21K, https://image-net.org/download-images.php
* License of dataset: [Same as ImageNet 2012](https://image-ne…
-
Thanks for your work！
I would like to ask will provide the model weights of "Pre-training on ImageNet-21K" . I want to run Cross-dataset evaluation directly, but I can't find it in existing documenta…
-
@yhcao6 could you please provide the imagenet21k sampled images for step3 training of detic, as imagenet21k is too large to download, thanks
-
I find no clue about this less used dataset. Could you give a link?