Plz fix run_vila.py line 65 output variable(s)

[ziyaosg]

Hi!

Directly running

python -W ignore llava/eval/run_vila.py \
    --model-path Efficient-Large-Model/VILA1.5-40b \
    --conv-mode hermes-2 \
    --query "<video>\n Please describe this video." \
    --video-file "demo.mp4"

gives ValueError: too many values to unpack (expected 2) for line 65 in run_vila.py:

images, num_frames = opencv_extract_frames(video_file, args.num_video_frames).

After closer inspection, I realized only the images are outputted by this function, not the number of frames. So simply changing this line to

images = opencv_extract_frames(video_file, args.num_video_frames)

fixed the issue on my end.

Please help check if this is the expected behavior and fix the output if necessary. Thank you!

[Lyken17]

now we recommend to use vila-infer to run the inference

vila-infer \
    --model-path Efficient-Large-Model/NVILA-15B \
    --conv-mode auto \
    --text "Please describe the video" \
    --media https://huggingface.co/datasets/Efficient-Large-Model/VILA-inference-demos/resolve/main/OAI-sora-tokyo-walk.mp4