MCPcopy Create free account
hub / github.com/CompVis/diff2flow / Decoder

Function Decoder

diff2flow/tiny_autoencoder.py:39–46  ·  view source on GitHub ↗
(latent_channels=4)

Source from the content-addressed store, hash-verified

37
38
39def Decoder(latent_channels=4):
40 return nn.Sequential(
41 Clamp(), conv(latent_channels, 64), nn.ReLU(),
42 Block(64, 64), Block(64, 64), Block(64, 64), nn.Upsample(scale_factor=2), conv(64, 64, bias=False),
43 Block(64, 64), Block(64, 64), Block(64, 64), nn.Upsample(scale_factor=2), conv(64, 64, bias=False),
44 Block(64, 64), Block(64, 64), Block(64, 64), nn.Upsample(scale_factor=2), conv(64, 64, bias=False),
45 Block(64, 64), conv(64, 3),
46 )
47
48
49class TinyAutoencoderKL(nn.Module):

Callers 1

__init__Method · 0.70

Calls 3

ClampClass · 0.85
convFunction · 0.85
BlockClass · 0.85

Tested by

no test coverage detected