![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/1701311/image.png)
Question about the Matlab Wasserstein GAN example
5 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
The original Wasserstein gan paper suggest removing the Critic's last dense layer activation function(sigmoid) such that the output value is not limited to fake or real. The posted example still uses sigmoid layer, am I right?
0 comentarios
Respuestas (1)
Malay Agarwal
el 22 de Mayo de 2024
Editada: Malay Agarwal
el 22 de Mayo de 2024
The diagram of the Discriminator model in the example (https://www.mathworks.com/help/deeplearning/ug/trainwasserstein-gan-with-gradient-penalty-wgan-gp.html) shows that the model does have a “sigmoid” layer at the end:
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/1701311/image.png)
This can also be confirmed by looking at how the Discriminator model is defined:
layersD = [
imageInputLayer(inputSize,Normalization="none")
convolution2dLayer(filterSize,numFilters,Stride=2,Padding="same")
leakyReluLayer(scale)
convolution2dLayer(filterSize,2*numFilters,Stride=2,Padding="same")
layerNormalizationLayer
leakyReluLayer(scale)
convolution2dLayer(filterSize,4*numFilters,Stride=2,Padding="same")
layerNormalizationLayer
leakyReluLayer(scale)
convolution2dLayer(filterSize,8*numFilters,Stride=2,Padding="same")
layerNormalizationLayer
leakyReluLayer(scale)
convolution2dLayer(4,1)
sigmoidLayer]; % Notice the sigmoid layer at the end
Hope this helps!
0 comentarios
Ver también
Categorías
Más información sobre Statistics and Machine Learning Toolbox en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!