pytorch_ema
pytorch_ema copied to clipboard
Tiny PyTorch library for maintaining a moving average of a collection of parameters.
Hi, I would like to know how to use pytorch-ema for training in pytorh-lightning?
Thank for your this awesome lib.. I had a little question: do we need to use `,model.eval()` inside the `with` context manager? So, which of these are legit please? ```...
Hi, just looking for some clarification on the documentation. I'm new to PyTorch and have tried to figure this out by reading the source code but the mutability of the...
Thank you for your selfless sharing. In the author's application, EMA has been applied to all parameters that can be trained, but there are some statistics, such as the mean...
here is my code: from torch_ema import ExponentialMovingAverage model = ... optimizer = ... scheduler = ... ema_model = ExponentialMovingAverage(parameters=pg, decay=0.9999) for epoch in range(args.epochs):# train print('epoch:',epoch,'Current learning rate:', optimizer.param_groups[0]['lr'])...