yuangongnd / uavm Goto Github PK

Hi,

I can reproduce the results of finetuning on Audioset 20K with the multimodal setting.
However, I tried to get the visual-only baseline from the multimodal pre-trained weights. I can only get 13.11 (missingvideoonly setting) and 12.7 (videoonly).

Pretrained ViT-B-16 on ImageNet-21K with multimodal script can get ~9.5mAP.
Pretrained CAV-MAE with audioonly script can get ~17.6 mAP.

Do you have the script for training videoonly baseline?

Thank you

“Download the AudioSet dataset, create json datafiles to include the wav path, see samples at [here], generate sample weight file using ./datafile/gen_weight_file.” in VGGSound Recipe and AudioSet Recipe cannot be accessed.

Could you please tell me if the json file here is in the same format as the json file in your other project AST, because I have run your AST code, and if it is the same as the json here, can I just copy it over?