ValueError: Sample larger than population or is negative

[Larerr]

Thanks to the author for his wonderful work!
I am executing the function Three-step 3D Feature Extraction (Scene) and after Step1 I get the following output:

But when I execute the
python chatcaption.py --specific_scene f6e9ec5953854dff94176c36b877c519
I get the following error:

Traceback (most recent call last):
  File "chatcaption.py", line 108, in <module>
    main()
  File "chatcaption.py", line 85, in main
    caption_result = caption_images_from_dir(
  File "/data3/hongbo.wang/3D-LLM/3DLanguage_data/ChatCaptioner_based/caption.py", line 455, in caption_images_from_dir
    image_list = random.sample(image_list, 4)
  File "/home/hongbo.wang/anaconda3/envs/3DLLM/lib/python3.8/random.py", line 363, in sample
    raise ValueError("Sample larger than population or is negative")
ValueError: Sample larger than population or is negative

After debugging, it was found to be in . /3DLanguage_data/ChatCaptioner_based/caption.py, and found that there was a problem with the function caption_images_from_dir.

    for view in views.
        image_list = glob.glob(f"{image_dir}/rgb_{view}*.png")
        image_list = [os.path.basename(path) for path in image_list]
        image_list = sorted(image_list, key=lambda i: int(re.findall(r"\d+", i)[0]))
        image_list = random.sample(image_list, 4)

This code is looking for a file with the name "rgb_outside*"** from the output file of Step1, but . /3DLanguage_data/ChatCaptioner_based/objaverse_render/render.py doesn't have a file with "rgb" in its name, which causes the random.sample(image_list, 4), the image_list is empty, so I get an error.

Please ask the author what should be done to solve this situation, thanks a million!