volo icon indicating copy to clipboard operation
volo copied to clipboard

VOLO: Vision Outlooker for Visual Recognition

Results 29 volo issues
Sort by recently updated
recently updated
newest added

According to my understanding, the permute order should be (2, 0, 1) but not (2, 1, 0), otherwise, the mul(a, v) calculation may have transposed position correspondings. ![image](https://user-images.githubusercontent.com/16267477/132193848-17faa94b-1600-4d17-8e0e-2e821729df80.png) Please correct...

When I run the code of volo, I get a debug error as follows: /pytorch/aten/src/ATen/native/cuda/ScatterGatherKernel.cu:312: operator(): block: [0,0,0], thread: [6,0,0] Assertion `idx_dim >= 0 && idx_dim < index_size && "index...

Traceback (most recent call last): File "main.py", line 960, in main() File "main.py", line 670, in main optimizers=optimizers) File "main.py", line 779, in train_one_epoch label_size=args.token_label_size) File "/opt/conda/lib/python3.6/site-packages/tlt/data/mixup.py", line 90, in...

Could you share the training script for training volo_d5 ↑512? how long will it take on 8 GPUs?

Hello, I was trying to compute the class weight "balanced". I see that there are two arguments: ``` parser.add_argument('--dense-weight', type=float, default=0.5, help='Token labeling loss multiplier (default: 0.5)') parser.add_argument('--cls-weight', type=float, default=1.0,...

Thanks for your great work! After reading the paper, I have a question: Can I think volo as a "pixel-wise conditional conv" network? The reasons are: - The weighted average...

Hi, Thanks for your work. What's main difference between VOLO and DynamicConv? Though `Convolution` is not explicitly used, **Convolution is equivalent with Unfold + Matrix Multiplication + Fold (or view...

question