question about processing caption_sharegtpvideo_300k

Hi authors, thanks for your great work.
When using the codes for stage3 training, it seems the codes cannot process the catpion_sharegptvideo_300k dataset.

The "img" item in the following config (instruction_data.py) seems make no difference.

   caption_sharegptvideo_300k=[
        f"{my_path}/video/caption/sharegptvideo/train_300k.json", 
        "{my_path}/shareGPTVideo/train_video_and_instruction/train_300k/videos",
        "video",
        "img", # read from image
    ],

Thanks for your reply.

Start with instruction_data.py and the caption_sharegptvideo_300k configuration entry. Trace how the dataset JSON, video directory, and "img" value are consumed during stage3 training, then reproduce the reported processing behavior. Done means the dataset is processed as intended or the limitation and required configuration are clearly documented.