-
1.Revitalizing optimization for 3d human pose and shape estimation: A sparse constrained formulation(2021)
code:No
2.Body meshes as points(2021)
regared as a two class classification task(if a grid…
-
@hiyouga I've encountered a consistent issue where the logits score returns -inf during offline inference using models from Hugging Face/vLLM, even using the default inference example. How to solve th…
-
I use the code below to extract the image features.
```python
model = load_model('/data/checkpoints/sapiens_lite_host/pretrain/checkpoints/sapiens_1b_epoch_173_torchscript.pt2', True)
model.to(torc…
-
[[Open issues - help wanted!]](https://github.com/vllm-project/vllm/issues/4194#issuecomment-2102487467)
**Update [9/8] - We have finished majority of the refactoring and made extensive progress fo…
-
Thx for the marvelous work!
It seems lightseq only integrates its operators into NLP models. Does it support vision transformers? What do we need to do for adopting lightseq in Vision Transformer wor…
-
### System Information
// example for python user
OpenCV C++ version: 3.4.15
Operating System / Platform: Windows 10
Python version: 3.9.10
Pytorch Version: 2.0.1
### Detailed description
…
-
### Model description
Dear huggingface team,
The fair team published an improved version of dinov2 [VISION TRANSFORMERS NEED REGISTERS](https://arxiv.org/abs/2309.16588). The models and checkpoi…
-
Is it working right now in any way?
-
I am getting `[1] 1774566 floating point exception python main_finetune.py --batch_size 16 --model vit_large_patch16 --epochs 50` when trying to run your finetuning script. I also slightly changed…
-
If run this code:
```python
model = Qwen2VLForConditionalGeneration.from_pretrained(
"Qwen/Qwen2-VL-7B-Instruct",
torch_dtype=torch.bfloat16,
attn_implementation="flash_attention_2",
…