public async Task Organizing()
{
var networkConfig = new NetworkConfiguration(3, 1);
var network = new KohonenNetwork <Logistic>(networkConfig);
var learningConfig = new LearningConfiguration
{
ThetaFactorPerEpoch = 0.95,
OrganizingAlgorithm = new Organizing <Logistic>(network, 0.777)
};
var learning = new UnsupervisedLearning(network, learningConfig);
var inputs = _getInputs();
await learning.LearnAsync(inputs, 25);
network.Input(_control[0]);
var res0 = await network.GetOutputIndexAsync();
network.Input(_control[1]);
var res1 = await network.GetOutputIndexAsync();
network.Input(_control[2]);
var res2 = await network.GetOutputIndexAsync();
Assert.NotEqual(res0, res1);
Assert.Equal(res1, res2);
}
public void ANN()
{
//var dbn = new DeepBeliefNetwork(new Accord.Neuro.ActivationFunctions.GaussianFunction(), 1, new int[] { 3, 5, 5, 10, 3 });
var dbn = DeepBeliefNetwork.CreateGaussianBernoulli(50, new int[] { 3, 5, 5, 10, 3 });
var dblNode = new UnsupervisedLearning(new DeepBeliefNetworkLearning(dbn) {
Algorithm = (h, v, i) => new ContrastiveDivergenceLearning(h,v)
{
LearningRate = 0.1,
Momentum = 0.5,
Decay = 0.001
}
});
var cout = new ConsoleOut("ANNOut.txt");
dbn.UpdateVisibleWeights();
dblNode.Process.LinkTo(cout.Output);
var data = new double[1000000][];
for (int i = 0; i < 1000000; i++)
{
var curr = new double[50];
for (int j = 0; j < 50; j++)
{
curr[j] = Math.Sin((double)j);
}
data[i] = curr;
}
dblNode.Process.Post(new MachineLearningData<double[][], double>(data));
System.Threading.Thread.Sleep(1000);
}
public void Learning()
{
var networkConfig = new NetworkConfiguration(3, 5);
var learningConfig = new LearningConfiguration
{
ThetaFactorPerEpoch = 0.95,
DefaultRepeatsNumber = 25
};
var network = new KohonenNetwork <Logistic>(networkConfig);
var learning = new UnsupervisedLearning(network, learningConfig);
var inputs = _getInputs();
learning.Learn(inputs);
network.Input(_control[0]);
var res0 = network.GetOutputIndex();
network.Input(_control[1]);
var res1 = network.GetOutputIndex();
network.Input(_control[2]);
var res2 = network.GetOutputIndex();
Assert.NotEqual(res0, res1);
Assert.Equal(res1, res2);
}
public void DBN()
{
var dbn = DeepBeliefNetwork.CreateGaussianBernoulli(1024, 10);
new GaussianWeights(dbn).Randomize();
dbn.UpdateVisibleWeights();
var dblNode = new UnsupervisedLearning(new DeepBeliefNetworkLearning(dbn));
var dataNorm = new Normalizer();
var fromFile = new TextReader();
var dataFile = new System.IO.FileInfo(@"C:\Users\Dan\Documents\Visual Studio 2015\Projects\AccordNETSamples\framework-master\Samples\Neuro\Deep Learning\Resources\optdigits-tra.txt");
// POST DataFile -> TextReader -> ? -> Normalizer -> UnsupervisedLearning
var dataProc = new TransformBlock<Instance<string>, IObservable<List<Instance<double>>>>(inst =>
{
var reader = new System.IO.StringReader(inst.Data);
var samples = new Subject<List<Instance<double>>>();
var epochSamples = new List<MachineLearningData<double[], double>>();
var buffer = new char[(32 + 1) * 32]; // 32 chars + \n
var count = 0;
while (true)
{
var read = reader.ReadBlock(buffer, 0, buffer.Length);
var label = reader.ReadLine();
if (read < buffer.Length || label == null)
break;
var currMLS = new MachineLearningData<double[], double>();
currMLS.Class = Convert.ToInt32(label);
count++;
}
return samples.AsObservable();
});
}
