Skip to main content

Advertisement

Table 4 Detailed architecture of UpscaleGAN \(64^{3}\rightarrow 256^{3}\). \(d=64\). The parameter shape of the inception convolution written InConv is too large be written in the table

From: Cosmological N-body simulations: a challenge for scalable generative models

OperationParameter sizeOutput shape
 Generator
Input \(z\,\mathcal{N}(0,1)\)(n,32,32,32,1)
Input smooth image(n,32,32,32,1)
Input borders(n,32,32,32,7)
Concatention(n,32,32,32,9)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,1)
ReLu(n,32,32,32,1)
 Discriminator
Input generated image(n,32,32,32,1)
Input borders(n,32,32,32,7)
Reshape to a cube(n,64,64,64,1)
Input smooth image (n,64,64,64,1)
Concatenation (+ diff) (n,64,64,64,3)
InConv 3D (Sride 2)*(n,32,32,32,2d)
LReLu (α = 0.2) (n,32,32,32,2d)
InConv 3D (Sride 1)*(n,32,32,32,2d)
LReLu (α = 0.2) (n,32,32,32,2d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 1)*(n,32,32,32,d)
LReLu (α = 0.2) (n,32,32,32,d)
InConv 3D (Sride 2)*(n,16,16,16,d)
LReLu (α = 0.2) (n,16,16,16,d)
InConv 3D (Sride 2)*(n,8,8,8,d)
LReLu (α = 0.2) (n,8,8,8,d)
Reshape(n,16384)
Compute PSD(n,1914)
Concatenate(n,18298)
Dense(18,298,64)(n,64)
LReLu (α = 0.2) (n,64)
Dense(6416)(n,16)
LReLu (α = 0.2) (n,16)
Dense(16,1)(n,1)