Skip to main content

Table 3 CAE-time architecture for Burger’s and Stoker’s test cases

From: Reduced-order modeling for stochastic large-scale and time-dependent flow problems using deep spatial and temporal convolutional autoencoders

Layer

Nb of filters

Kernel size

Activation function

Output shape

Encoder-time

    

Input

–

–

–

\(104\times L_{x}\)

Conv-pool

32

3–2

PReLU

\(52\times 32\)

Conv-pool

64

3–2

PReLU

\(26\times 64\)

Conv-pool

128

3–2

PReLU

\(13\times 128\)

Flatten

–

–

–

\(1\,664\)

Dense (output)

–

–

PReLU

\(L_{t}=10\)

Decoder-time

    

Input

–

–

–

\(L_{t}=10\)

Dense

–

–

PReLU

1664

Reshape

–

–

–

\(13\times 128\)

Conv-Upsamp

128

3–2

PReLU

\(26\times 128\)

Conv-Upsamp

64

3–2

PReLU

\(52\times 64\)

Conv-Upsamp

32

3–2

PReLU

\(104\times 32\)

Conv (output)

1

3

PReLU

\(104\times L_{x}\)