Yucheng Han

图片问题

2

我这里显示的图片有问题，请问是什么情况

Well, I'm not familiar with reinforcement learning and colab. I just wonder whether it's possible to run dopamine just on my computer. I can run the training part, but I...

The input shape

2

I think this tool may be quite useful, but there are still something not convenient for me... I want to get the data of my model, but since my model's...

Hope there will be instructions about downloading from the HuggingFace Dataset

2

I am trying to download MMIC-IT from HuggingFace. However, the instructions are not clear. I downloaded it in the following ways: from datasets import load_dataset configs = ['LA_Images', 'LACONV_Preview', 'LACONV_Instructions',...

About the pretrained model

4

It seems that the pretrained models provided through the link are just 0 B. I don't know if anyone else meets this problem.

For example, your provided link for "Chain of Thought" is linked to FLAN. However, the file size is different from COT data in FLAN Collection (https://huggingface.co/datasets/conceptofmind/cot_submix_original/tree/main/data)

I am confused by the generation method and the source. For example, I do not know what the src of xP3 means: xP3 | 78883588 | ML | MT |...

grounding_dino_demo.py performs wrong when set device to other GPUs.

1

As described in the title, if setting the device to "cuda:2" there are no results by dino. Still not sure where the bug occurs.