https://github.com/Alibaba-MIIL/ImageNet21K/blob/00ef9989825bbcb8dedc91eac18638c129eb5ad8/src_files/models/utils/factory.py#L54

Why do we need to load pretrained ImageNet-1K model if we are going to pretrain based on the imagenet21k dataset?

I did a full preprocessing using your script on the winter21_whole.tar.gz dataset. Is it normal to have tons of general fail messages? Will these fails of images impact final training results?

Dear author ： When I test resnet50 model you provide on imagenet21k-p val dataset, the ImageNet-21K-P semantic top-1 Accuracy is just 69%, but what you have claimed is 75.6%, but I indeed see the improvement on downstream task, what's the problem?

i tried to run visualize_detector.py, but there is a httperror

urllib.request.urlretrieve(url, filename)

Exception has occurred: HTTPError HTTP Error 403: Forbidden

how can it be solved?

https://github.com/Alibaba-MIIL/ImageNet21K/blob/72c822a7a30290f078a2611139a1f0bcdb668606/visualize_detector.py#L25

error:

Unknown model (vit_base_patch16_224_miil_in21k)

how to run with my own wight

1 machine 2 GPUs I cant use command line to start how to start in the wInternal code to use DDP

i cant find where to save mode, how to save in the DDP

Hi @mrT23, thanks for your great work! Currently I use timm train.py to finetune the 'vit_base_patch16_224_miil_in21k' model on cifar100, however I can't get the reported result 94.2%. Here is my running script.

python -m torch.distributed.launch --nproc_per_node=8 --master_port 6016 train.py \
/data/cifar-100-images/ \
-b=64 \
--img-size=224 \
--epochs=50 \
--color-jitter=0 \
--amp \
--lr=2e-4 \
--sched='cosine' \
--model-ema --model-ema-decay=0.995 --reprob=0.5 --smoothing=0.1 \
--min-lr=1e-8 --warmup-epochs=3 --train-interpolation=bilinear --aa=v0 \
--model=vit_base_patch16_224_miil_in21k \
--pretrained \
--num-classes=100 \
--opt=adamw --weight-decay=1e-4 \
--checkpoint-hist=1

I try several settings: adam, lr=2e-4, wd=1e-4 92.44% adamw, lr=2e-4, wd=1e-4 92.90% sgd, lr=2e-4, wd=1e-4 87.49% adamw, lr=4e-4, wd=1e-4 92.24% adamw, lr=2e-4, wd=1e-2 93.08% Could you give me some suggestions?

In the vit paper, it says:

The classification head is implemented by a MLP with one hidden layer at pre-training time and by a single linear layer at fine-tuning time

So if you are using timm package, they define the head like this: https://github.com/rwightman/pytorch-image-models/blob/d3f744065088ca9b6b3a0f968c70e90ed37de75b/timm/models/vision_transformer.py#L293

Did you reach the stats in your paper using single layer head or head of one hidden layer?

Hi, I try to transfer the ImageNet-21k model (resnet50_miil_21k) to ImageNet-1K according to the details provided in your paper and https://github.com/Alibaba-MIIL/ImageNet21K/blob/main/Transfer_learning.md. But I only get 79.5% top-1 acc on ImageNet-1K, lower than the 82.0% in your readme. Could you give me some suggestions?

My training command is as follows: python3 -m torch.distributed.launch --nproc_per_node=8 train.py
data/ImageNet-1k/
-b=128
--img-size=224
--epochs=100
--color-jitter=0
--amp
--sched='cosine'
--model-ema --model-ema-decay=0.995 --reprob=0.5 --smoothing=0.1
--min-lr=1e-8 --warmup-epochs=3 --train-interpolation=bilinear --aa=v0
--pretrained
--lr=2e-4
--model=resnet50
--opt=adam --weight-decay=1e-4

Could you provide the training parameters when using 8 * V100 with DDP? When I use the following command line python3 -u -m torch.distributed.launch --nnodes 1 --node_rank 0 --nproc_per_node 8 --master_port 2221 train_semantic_softmax.py --data_path /data/imagenet22kp_fall/ --model_name mobilenetv3_large_100 --epochs 80 --weight_decay 1e-4 --batch_size 1024 --lr 3e-4 --num_classes 11221 --tree_path ./data/imagenet21k_miil_tree.pth --model_path=./mobilenetv3_large_100.pth The accuracy is only 71.366%, which is far lower than 73.1% reported in the paper.

Hi,

Can you please share the link to the weights file of model trained on Stanford cars data set. I am unable to get the expected results using https://miil-public-eu.oss-eu-central-1.aliyuncs.com/model-zoo/ImageNet_21K_P/models/tresnet_l_v2_miil_21k.pth

Please guide.

Thanks,

Anyone here have trouble reaching the mentioned accuracy for ViT-B? For some reason, the best accuracy I can get is 77% top1 without KD. While in the paper they said they reach 81% top1 without KD and 84.4% top1 with KD. Anyone manage to get that accuracy? If so, can you tell me what hyperparameters did you use? Thanks!

What is the teacher model when using semantic softmax with KD? The figure in the paper is not clear on what the teacher is. Or in there is no code example on how to use the KD

Nowhere on the paper or code where it mentions what Dropout prob and momentum value used for SGD when fine tuning on ImageNet1k. Is is the same as https://arxiv.org/pdf/2010.11929.pdf ? Also can you provide the code for imagenet1k. I would like to see how the images are normalized for that stage?

Good morning, thank you very much for your work. Can you share the hyperparameters/training procedure that have been used to perform the finetuning of resnet50 to standard ImageNet?

Thank you very much in advance!