how to connect new network to existing model

2 次查看(过去 30 天)
I have combined output of 2 networks. I need to connect combined output with 3rd network. I am unable to do it. I have to connect add_1 to deconv1 layer. please guide.
load netop.mat;
load trained_net.mat;
lgraph = layerGraph;
lgraph = addLayers(lgraph,visual_net_layes);
lgraph = addLayers(lgraph,audio_net_layers);
add = additionLayer(2,'Name','add_1')
lgraph = addLayers(lgraph,add);
lgraph = connectLayers(lgraph,'fc1','add_1/in1');
lgraph = connectLayers(lgraph,'fc2','add_1/in2');
% plot(lgraph)
newlayers = [
transposedConv2dLayer(11,96,'Stride',4,'Name','deconv1');
batchNormalizationLayer
reluLayer
transposedConv2dLayer(11,96,'Stride',4);
batchNormalizationLayer
reluLayer
transposedConv2dLayer(11,96,'Stride',4);
batchNormalizationLayer
reluLayer
averagePooling2dLayer(2,'Stride',2)
fullyConnectedLayer(2)
softmaxLayer
classificationLayer];
%lgraph = layerGraph(newlayers);
lgraph = addLayers(lgraph,newlayers);
figure
plot(lgraph)

采纳的回答

Vinayak Choyyan
Vinayak Choyyan 2023-4-10
Hi Shilpa Sonawane,
As per my understanding, you are facing an issue when trying to combine 3 deep neural networks.
From the code provided, I see that you are trying to combine the output of ‘add_1’ to the input of ‘deconv1’ layer. Please try out the following code.
lgraph=connectLayers(lgraph,'add_1/out','devcon1');
For more information on ‘additionLayer’, please check out the following documentation https://www.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.additionlayer.html
I hope this resolves the issue you are facing.

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Image Data Workflows 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by