re training neural network from previous state using trainNetwork

 I am training a deep neural network , using the following matlab function:

net = trainNetwork(XTrain,YTrain,layers,options);

 

could I use the trainNetwork command to retrain the network (not from scratch), using the last network state from previous training?

 

I am sharing some of the code:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

layers = [

 

imageInputLayer([1 Nin 5],"Name","imageinput")

convolution2dLayer([1 3],32,"Name","conv","Padding","same")%

%batchNormalizationLayer('Name','batchDown')

tanhLayer("Name","tanh1")

convolution2dLayer([1 3],32,"Name","conv","Padding","same","DilationFactor",[1 3])%

%batchNormalizationLayer('Name','batchDown1')

tanhLayer("Name","tanh2")

convolution2dLayer([1 3],2,"Name","conv","Padding","same","DilationFactor",[1 9])%

 

regressionLayer];

 

 

options = trainingOptions('adam', ...

'InitialLearnRate',0.001, ...

'MaxEpochs',50000, ...

'ExecutionEnvironment','parallel',...

'Verbose',false, ...

'Plots','training-progress');

 

 

Net1 = trainNetwork(XTrain1,YTrain1,layers,options);

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

 

now I would like to train again getting Net2 using new data, and starting the training from Net1 stage.

for example:

Net2 = trainNetwork(XTrain2,YTrain2,layers,options);

however its not clear to me how to start the train process for Net2 using Net1 stage.

ANSWER

Matlabsolutions.com provide latest MatLab Homework Help,MatLab Assignment Help for students, engineers and researchers in Multiple Branches like ECE, EEE, CSE, Mechanical, Civil with 100% output.Matlab Code for B.E, B.Tech,M.E,M.Tech, Ph.D. Scholars with 100% privacy guaranteed. Get MATLAB projects with source code for your learning and research.

You can do the following if Net1 is a SeriesNetwork:

 

 

Net2 = trainNetwork(XTrain2, YTrain2, Net1.Layers, options);

or if Net1 is a DAGNetwork:

 

 

Net2 = trainNetwork(XTrain2, YTrain2, layerGraph(Net1), options);

This will train using Net1 as the initial network.

 

If you would also like to prevent weights in certain layers from changing, you could use freezeWeights (see how to access it below). This function could be used to set the learning rates in those layers to zero. During training, trainNetwork does not update the parameters of the "frozen" layers.

 

edit(fullfile(matlabroot,'examples','nnet','main','freezeWeights.m'))