-
Hi,
Thanks for the publishing the nice code. I can't find code in this repo that makes the slide attention visualizations similar to that seen in Figures 2 and 3 of the paper. Is this available so…
-
Thank you for developing this!
## Context
Due to lenghty computation time and in order to speed things up, I thought about using the `flash_attention_2` and smaller floating points `torch.float16`…
-
Hi, thanks for your amazing work.
We notice that in your newly released version, you use co-attention from MCAN to achieve a much higher performance, I wonder that will this part of code be release…
-
is there newest precompile whl for win10 torch2.5 python3.10?
i only find for torch2.4
-
**Describe the feature**
class Template里面可以做padding,但是Qwen2VLTemplateMixin, InternLMXComposer2Template里面只有im_mask，没有，input_ids的attention_mask,（有PADDING的情形）
能不能把padding attention_mask都放回去呀。
http…
-
Hit this error when running https://huggingface.co/genmo/mochi-1-preview. Repro script as followed.
```python
import torch
from torch.nn.attention import sdpa_kernel
device = torch.device("cud…
-
### Problem Description
Hello,
Model https://huggingface.co/databricks/dbrx-instruct is not working with Flash attn on ROCM, working on NVIDIA 100
The current model is not working with AMD MI25…
-
Hello @Cjl-MedSeg 🤗
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daliy papers: https://huggingface.co/papers/2411.12814. The pa…
-
- https://arxiv.org/abs/2104.06399
- 2021
本論文では、CoaT（Co-scale conv-attentional image Transformers）を発表する。
CoaTは、Co-scaleとConv-attentional機構を備えたTransformerベースの画像分類器である。
co-scale機構は、Transformerのエンコ…
e4exp updated
3 years ago
-
### Discussed in https://github.com/comfyanonymous/ComfyUI/discussions/5334
Originally posted by **Benix-laj** October 23, 2024
So I've just installed ComfyUI and ComfyUI manager, and download…