MCPcopy Create free account
hub / github.com/OpenImagingLab/4DSloMo / forward

Method forward

lpipsPyTorch/modules/networks.py:53–63  ·  view source on GitHub ↗
(self, x: torch.Tensor)

Source from the content-addressed store, hash-verified

51 return (x - self.mean) / self.std
52
53 def forward(self, x: torch.Tensor):
54 x = self.z_score(x)
55
56 output = []
57 for i, (_, layer) in enumerate(self.layers._modules.items(), 1):
58 x = layer(x)
59 if i in self.target_layers:
60 output.append(normalize_activation(x))
61 if len(output) == len(self.target_layers):
62 break
63 return output
64
65
66class SqueezeNet(BaseNet):

Callers

nothing calls this directly

Calls 2

z_scoreMethod · 0.95
normalize_activationFunction · 0.85

Tested by

no test coverage detected