jancylee

the pre-training loss

10

The pre-training losses are always negative (like -199.03), is that normal?

positive pairs

5

As the two crops have different scales in the original image, and after the network, the two outputs have the same size. So how did you find the positive pairs,...

__init__() got an unexpected keyword argument 'prefetch'

4

When train the first stage, I met the error "__init__() got an unexpected keyword argument 'prefetch'". Have you changed somgthing about the environment ”open-mmlab“?