PyTorch: Difference between revisions

Jump to navigation Jump to search
m
no edit summary
m (added missing import statement "import torch.nn.functional as F")
mNo edit summary
Line 856: Line 856:
         which can cause issues due to updating logs too frequently."""
         which can cause issues due to updating logs too frequently."""


     trainer = pl.Trainer(gpus=2, num_nodes=1,accelerator='ddp', max_epochs = args.max_epochs, progress_bar_refresh_rate=0)  
     trainer = pl.Trainer(accelerator="gpu", devices=2, num_nodes=1, strategy='ddp', max_epochs = args.max_epochs, enable_progress_bar=False)  


     transform_train = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
     transform_train = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
cc_staff
282

edits

Navigation menu