-
@Neerajj9 In the following python code, https://github.com/Neerajj9/Stacked-Attention-Networks-for-Visual-Question-Answering/blob/master/save_QA.py, at line number 102, where can I find the vqa_final_…
-
in the training process, I found that if I set batch size > 1, the loss sometimes will be nan
```
tensor(nan, device='cuda:0', grad_fn=)
```
and some logits also nan
```
(Pdb) p llm_out.logits…
-
Thank you very much for your work. I have a question about
`image_features, pre_image_features = vision_tower(images, attention_mask=flatten_vit_attention_mask)`. What does this line do, and how doe…
-
Hi, thanks for the nice work and great repo!
I changed config to train_with_clip=1 to include ClipLoss.
Then, I am getting the following error in the eval step:
![image](https://user-images.githu…
-
### Motivation
Hi.
I saw a new version of the LAION dataset just came out (Aug 30th, 2024). The samples from the CSAM dataset were removed from LAION.
Do you intend to re-train your InternViT model…
-
I have installed matmulfreellm with Triton for Windows via triton-2.0.0-cp310-cp310-win_amd64.whl which makes matmulfreellm work on the 'configuration' file but fail on the 'generate' file. The 'gener…
-
[Show, attend and tell: Neural image caption generation with visual attention](http://proceedings.mlr.press/v37/xuc15.html)
Inspired by recent work in machine translation and object detection, we int…
-
### Describe the issue
Unable to build the ONNX Runtime our of release candidate branch on Windows against CUDA 12.6
### Urgency
This issue is vital if release plans to support CUDA 12.6
### Targe…
-
Hey John! Here's the curriculum that I've worked on in the past. It's a bit less focused on language models as a sole topic, and more on modern ML from a broad perspective.
- Essential Concepts of …
zmaas updated
2 weeks ago
-
The current banner in the project serves as a vital element but may benefit from enhancements to improve its visual appeal, messaging, or functionality. This task aims to elevate the banner's overall …