Why did the 'scale' in the __init__ function cause an error during initialization?
cmd error:
(trackany) E:\track_anything\Track-Anything>python app.py --device cuda:0
Initializing BaseSegmenter to cuda:0
Hyperparameters read from the model weights: C^k=64, C^v=512, C^h=64
Single object mode: False
load pretrained SPyNet...
Loads checkpoint by http backend from path: https://download.openmmlab.com/mmediting/restorers/basicvsr/spynet_20210409-c6c1bd09.pth
Traceback (most recent call last):
File "E:\track_anything\Track-Anything\app.py", line 431, in
what's the gradio tension of this project? Mine is 4.15.0 but it express ERROR.
更新:把gradio版本降低后成功了。
Solution: pip install gradio==3.30
谢谢,对我很有帮助-爱来自瓷器
It is recommended to use
pip install gradio==3.39.0 . I have tried this based on #120