Combining several stacked Autoencoders

1 次查看（过去 30 天）

DS 2018-4-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/393033-combining-several-stacked-autoencoders

Is it possible to combine several stacked Autoencoders (AE) sequentially prior to a softmax layer. I implemented 2 stacked AEs following this tutorial https://www.mathworks.com/examples/neural-network/mw/nnet-ex33663708-train-stacked-autoencoders-for-image-classification (I did not included the softmax layer in the stacked architecture, thus the AEs are just pretrained and not fine tuned). Than I defined a DAG network as follows:

layers = [ fullyConnectedLayer(100,'Name','fc')
    softmaxLayer('Name','softmax')
    classificationLayer('Name','classOutput')]; %100 is the number of outputs from all AEs (50 + 50) combined
lgraph = layerGraph(layers);

Afterwards, I tried to add one of the AEs to the DAG:

lgraph = addLayers(lgraph,autonenc1);

It resulted in the following error message:

Error using nnet.internal.cnn.util.validateLayersForLayerGraph (line 20)
Expected input to be one of these types:
nnet.cnn.layer.Layer
Instead its type was network.

Is there another way to do this?