search

junwenxiong / diff_sal

Offical implemention of the paper DiffSal: Joint Audio and Video Learning for Diffusion Saliency Prediction
19 stars 1 forks source link

pth file miss #3

Closed Full-Adder closed 4 months ago

Full-Adder commented 5 months ago

https://github.com/junwenxiong/diff_sal/blob/c9b8070a8fdaafb0bfddb1fb067f1038b42da94c/cfgs/audio_visual.py#L31

Nice work! I am trying to run your code but where can i get this .pth file?

Full-Adder commented 5 months ago

add also 'vggish': 'data/pretrained_models/vggish.pth',

Full-Adder commented 5 months ago

when i ues python=3.10, in datasets/spatial_transforms.py raise module ‘collections‘ has no attribute ‘Iterable‘, i fix these bug using collections.abc.Iterable

junwenxiong commented 4 months ago

Sorry for late reply. The pth link is here https://download.openmmlab.com/mmaction/v1.0/recognition/mvit/converted/mvit-small-p244_16x4x1_kinetics400-rgb_20221021-9ebaaeed.pth. And the vggish is from https://github.com/harritaylor/torchvggish.