TMultiLayerPerceptron has 1600 inputs

ROOT
1

Hi,
I want to analyze data by ANN method.
My data can be read as the following way.

TTree *signal=(TTree*)input->Get("Events");
Double_t pmt[40][40];
signal->SetBranchAddress("pmt",pmt);

I read the example “mlpHiggs.C” and the neural network structure is defined like this:

TMultiLayerPerceptron *mlp = new TMultiLayerPerceptron("@msumf,@ptsumf,@acolin:5:3:type", "ptsumf",simu,"Entry$%2","(Entry$+1)%2");

Does any one know how to define the neural network structure in my case (1600 inputs, 2 outputs)?

Thanks a lot.

2

Hi,

Do you still have this problem? I recommend you re-submit this post in the ROOT Math and Stats forum with more details (how did it fail for you, ROOT version, etc.).

Philippe.