Pranav Rao comments

Repositories
Issues
Comments

Results 2 comments of


                                            Pranav Rao

Issue in Manual optimisation, during self.manual_backward call

I removed excess code, made a new Conda environment, installing just pytorch-lightning and tensorboard, and was able to replicate the same issue even with lightning version 2.2.3. I have edited...

RuntimeError when running basic GAN model (from tutorial at lightning.ai) with DDP

Doesn't setting `strategy == "ddp_find_unused_parameters_true"` make an extra forward pass, using more computation and time? As far I understant the tutorial, there doesn't seem to be any parameters with `requires_grad=True`...