MH_python

_probe = ffmpeg.probe(video_path) _stream = next((stream for stream in _probe['streams'] if stream['codec_type'] == 'video'), None) Although I added ffmpeg and ffprobe binaries to my venv/bin/, but still getting this error...

Downstream classification task checkpoint link not working

2

I am trying to download a downstream classification task model (VideoMAE-L K700) but got this instead: ``` This XML file does not appear to have any style information associated with...

Great works and research. My question is simply if is it possible to use only the visual/video part (already pretrained on video dataset like kinetics) for fine-tuning on long video...

Thank you for the nice compact work. We have started recently to face an ambiguous error casing the loss to become nan during the training. After enabling anomaly detection "...