rainley123

Do you solve this problem?

ok but i find that the model can run normally, although this problem occurs

your memory of GPU is not enough, you can change a smaller batch_size

> always inform me of _different tensor name not found_ error every time I press `run`. For example, > > > `NotFoundError (see above for traceback): Tensor name "scale5/block3/c/scale5/block3/c/moving_variance/biased" not...

And I also find that the structure of the pre-trained model is right, however it is nor equal to your code in inference()