如何在GAN深度学习中获得非空层?

1次观看(过去30天)

显示旧的注释

SREERAJ斗士 2022年2月15日

0
链接

直接链接到这个问题

https://de.mathworks.com/matlabcentral/answers/1650095-how-to-get-nonempty-layer-in-gan-deep-learning

我有密码。然而，它并没有在我的深度学习GAN代码中给我分层:

                          imds = imageDatastore(“C: \用户桌面\ \ Sreeraj \ \ \博士Mahindra \ DL”，“IncludeSubfolders”,真的,“FileExtensions”，(“使用”)，“LabelSource”，“foldernames”）;
                         
                          augmenter = imageDataAugmenter(RandXReflection=true);
                         
                          augimds = augmentedImageDatastore([64 64]，imds,DataAugmentation=augmenter);
                         
                          filterSize = 5;
                         
                          numFilters = 32;
                         
                          numLatentInputs = 10;
                         
                          projectionSize = [4 4 512];
                         
                          layersGenerator = [
                         
                          featureInputLayer (numLatentInputs)
                         
                          fullyConnectedLayer (prod (projectionSize))
                         
                          transposedConv2dLayer (filterSize 4 * numFilters)
                         
                          batchNormalizationLayer
                         
                          reluLayer
                         
                          transposedConv2dLayer (2 * numFilters filterSize,跨步= 2,裁剪=“相同”）
                         
                          batchNormalizationLayer
                         
                          reluLayer
                         
                          transposedConv2dLayer (filterSize numFilters,跨步= 2,裁剪=“相同”）
                         
                          batchNormalizationLayer
                         
                          reluLayer
                         
                          transposedConv2dLayer (filterSize 3步= 2,裁剪=“相同”）
                         
                          sigmoidLayer];
                         
                          lgraphGenerator = layerGraph(layergenerator);