I found a problem in the section USING PRETRAINED MODELS, the author did not have an error because the data is running CPU the whole time, but if the model and R3M are placed on the GPU, because the r3M is placed on the cpu by default, when sampling,
r3M = ReplayBuffer(storage=storage, transform=r3m)
it will automatically transform the data, at this time the data is on the CPU, and the R3M is on the GPU, it will report an error
I found a problem in the section USING PRETRAINED MODELS, the author did not have an error because the data is running CPU the whole time, but if the model and R3M are placed on the GPU, because the r3M is placed on the cpu by default, when sampling,
r3M = ReplayBuffer(storage=storage, transform=r3m)
it will automatically transform the data, at this time the data is on the CPU, and the R3M is on the GPU, it will report an error