Patrick Hiemsch

Where did you find the checkpoint? Is it created during training or once it finished training? There should be a /temp directory, but even after 45000 steps it did not...

So basically: EMA is the exponential moving average of the parameters, then it makes sense that the first few thousand EMA models just produce noise. The models called models just...

As I said look up EMA, the parameters of the exponential moving average will change very slowly compared to the „model“ parameters. Fix: train (far) longer. Or sample from model...

opt is not a model, it is the optimizer state if you want to restart training

I would be interested in the same. I guess 30000 rgb images is a lot so of course it will take a lot of time to train. But is there...