MCPcopy Create free account
hub / github.com/KeepTryingTo/Pytorch-GAN / load_checkpoint

Function load_checkpoint

SRGAN/utils.py:47–56  ·  view source on GitHub ↗
(checkpoint_file, model, optimizer, lr)

Source from the content-addressed store, hash-verified

45
46
47def load_checkpoint(checkpoint_file, model, optimizer, lr):
48 print("=> Loading checkpoint")
49 checkpoint = torch.load(checkpoint_file, map_location=config.DEVICE)
50 model.load_state_dict(checkpoint["state_dict"])
51 optimizer.load_state_dict(checkpoint["optimizer"])
52
53 # If we don't do this then it will just have learning rate of old checkpoint
54 # and it will lead to many hours of debugging \:
55 for param_group in optimizer.param_groups:
56 param_group["lr"] = lr
57
58
59def plot_examples(low_res_folder, gen):

Callers 1

mainFunction · 0.90

Calls

no outgoing calls

Tested by

no test coverage detected