Open desti-nation opened 1 year ago
From the error, I can see that the mismatch of the pos embedding shape is not matching the one from pretrained paramters. And from your input the img size is 50. I guess you need to resize your video width and height to 224 which is where the K400 Vivit weights are trained on.
When I want to finetune my dataset based on pretrained kinetics vivit model, the errors occured. I am new to pytorch, may I know How could solve the following errors? Thanks.
command
Errors: