search

OpenGVLab / Ask-Anything

[CVPR2024 Highlight][VideoChatGPT] ChatGPT with video understanding! And many more supported LMs such as miniGPT4, StableLM, and MOSS.
https://vchat.opengvlab.com/
MIT License
2.85k stars 230 forks source link

mistral版本三阶段训练代码问题 #179

Closed zhanghang-official closed 1 month ago

zhanghang-official commented 1 month ago

https://github.com/OpenGVLab/Ask-Anything/blob/cd1f925cff55a0a0dfa730bdac132035a900b140/video_chat2/dataset/it_dataset_mistral.py#L161 在调用load_and_transform_media_data_video函数时传入dynamic_config参数 https://github.com/OpenGVLab/Ask-Anything/blob/cd1f925cff55a0a0dfa730bdac132035a900b140/video_chat2/dataset/base_dataset.py#L69 但是load_and_transform_media_data_video函数定义并没有dynamic_config参数，会导致报错，请问dynamic_config参数什么作用，是否可以直接去掉dynamic_config。

Andy1621 commented 1 month ago

可以的，我更正一下代码

Andy1621 commented 1 month ago

代码更新过啦，这个参数是用来支持后续微调的，后面会更新一下~

zhanghang-official commented 1 month ago

好的 感谢！

dragen1860 commented 3 weeks ago

代码更新过啦，这个参数是用来支持后续微调的，后面会更新一下~

没有更新啊，最新的代码： commit 9648e8f14f19bd68ae7ad430a563a8550bcfb2bf (HEAD -> main, origin/main, origin/HEAD) Author: Kunchang Li 812487273@qq.com Date: Tue May 28 23:06:46 2024 +0800

Update README.md
Andy1621 commented 3 weeks ago

@zhanghang-official Hi! 我更新了大分辨率微调的代码，可以试用一下