search

snap-research / Panda-70M

[CVPR 2024] Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers
https://snap-research.github.io/Panda-70M/
438 stars 15 forks source link

Error when running video caption model inference. #11

Closed Marlod390 closed 3 months ago

Marlod390 commented 4 months ago

Thank you for your great work. But when I ran the video caption model according to the tutorial, the following error occurred:

Traceback (most recent call last): File "/home/markusc/github/Panda-70M/captioning/inference.py", line 27, in <module> cfg = Config(args) File "/home/markusc/github/Panda-70M/captioning/video_llama/common/config.py", line 25, in __init__ user_config = self._build_opt_list(self.args.options) AttributeError: 'Namespace' object has no attribute 'options'

How can I fix the problem? Maybe line 25 user_config = self._build_opt_list(self.args.options) in config.py should be replaced with user_config = self._build_opt_list(self.args.cfg_path)?

tsaishien-chen commented 4 months ago

Thanks for letting me know the bug. I have fixed it by adding these.

johndpope commented 4 months ago

@tsaishien-chen - random question do you guys have a SAN to store all these videos? or just use a bucket? s3 or minio? I'm was thinking of buying a couple of 12tb drives and "gluing" them together with ReFS to have 36tb drive.. but looks fragile.