public void ActivateTest()
{
NeuralNetwork nn = new NeuralNetwork(2, 2, 1);
Assert.AreNotEqual(new double[] { 0d }, nn.Activate(new float[] { 0f, 1f }));
Assert.AreNotEqual(new double[] { 1d }, nn.Activate(new float[] { 0f, 1f }));
Assert.AreEqual(1, nn.Activate(new float[] { 0f, 1f }).Length);
}
static void Main(string[] args)
{
Random random = new Random(13);
NeuralNetwork cNN = new NeuralNetwork(random, 0.2);
var conv = new ConvolutionLayer(new RectifiedLinearUnit(0.01), 8, 3, 3);
conv.IsSame = true;
cNN.AddNewLayer(new Shape(28, 28), conv);
cNN.AddNewLayer(new MaxPooling(2, 2));
cNN.AddNewLayer(new ConvolutionLayer(new RectifiedLinearUnit(0.01), 16, 3, 3));
cNN.AddNewLayer(new MaxPooling(2, 2));
cNN.AddNewLayer(new ConvolutionLayer(new RectifiedLinearUnit(0.01), 32, 3, 3));
cNN.AddNewLayer(new UnPooling(2, 2));
cNN.AddNewLayer(new ConvolutionLayer(new RectifiedLinearUnit(0.01), 16, 3, 3));
cNN.AddNewLayer(new MaxPooling(2, 2));
cNN.AddNewLayer(new Flatten());
cNN.AddNewLayer(new FeedForwardLayer(20, new RectifiedLinearUnit(0.01)));
cNN.AddNewLayer(new FeedForwardLayer(2, new SoftmaxUnit()));
Console.WriteLine(cNN);
GraphCPU graph = new GraphCPU(false);
NNValue nValue = NNValue.Random(28, 28, 2, random);
NNValue nValue1 = NNValue.Random(28, 28, 2, random);
NNValue outp = new NNValue(new double[] { 0, 1 });
NNValue outp1 = new NNValue(new double[] { 1, 0 });
DataSetNoReccurent data = new DataSetNoReccurent(new NNValue[] { nValue, nValue1 }, new NNValue[] { outp, outp1 }, new CrossEntropyWithSoftmax());
TrainerCPU trainer = new TrainerCPU(TrainType.MiniBatch, new Adam());
trainer.BatchSize = 2;
trainer.Train(10000, 0.001, cNN, data, 2, 0.0001);
double[] dbs = cNN.Activate(nValue, graph).DataInTensor;
double[] dbs1 = cNN.Activate(nValue1, graph).DataInTensor;
}
public void TrainTest()
{
NeuralNetwork nn = new NeuralNetwork(2, 2, 1);
nn.Train(new Tuple <float[], double[]>[] { new Tuple <float[], double[]>(new float[] { 1, 0 }, new double[] { 1 }), new Tuple <float[], double[]>(new float[] { 1, 1 }, new double[] { 0 }), new Tuple <float[], double[]>(new float[] { 0, 1 }, new double[] { 1 }), new Tuple <float[], double[]>(new float[] { 0, 0 }, new double[] { 0 }) }, 1000, 0.1f);
Assert.AreEqual(1D, nn.Activate(new float[] { 0f, 1f })[0], 0.2);
}
public static string generateOutput(NeuralNetwork network, char start, int length)
{
Matrix input = new Matrix(FieldLetterTranslator.traslateToField(start));
Graph g = new Graph(false);
Random rnd = new Random();
string result = "";
for (int i = 0; i < length; i++)
{
Matrix output = network.Activate(input, g);
//char act = FieldLetterTranslator.traslateToLetter(output.W);
char act = 'a';
try
{
act = FieldLetterTranslator.Letters[Util.PickIndexFromRandomVector(output, rnd)];
}
catch (Exception ex)
{
Console.WriteLine(ex.Message);
}
input = new Matrix(FieldLetterTranslator.traslateToField(act));
g = new Graph(false);
result += act;
}
return(result.Replace('$', '\n'));
}
private void activateNeuralNetwork(List <double> nnInputs)
{
for (int i = 0; i < nnInputs.Count; i++)
{
brain.PutSignalToNeuron(i, nnInputs[i]);
}
brain.Activate();
}
private static void Main(string[] args)
{
Console.WriteLine("App running");
const int countOutputs = 4;
var data = FileReader.Read(@"D:\auto.csv", countOutputs);
Console.WriteLine("Data from file loaded.");
var network = new NeuralNetwork(0.01, new SigmoidFunction(), data.Input.GetLength(1), 4, countOutputs);
Console.WriteLine("Neural network created.");
var dataSets = data.Input;
var expectedResults = data.Output;
if (ArrayHelper.Each(dataSets).Any(x => x > 1 || x < 0))
{
//scaling data [0...1]
dataSets = network.Scaling(dataSets);
Console.WriteLine("Data scaled.");
}
Console.WriteLine("Start train...");
network.Train(expectedResults, dataSets, CountEpoch);
Console.WriteLine("End train.");
//Compare data after training
var results = new List <List <double> >();
for (var i = 0; i < expectedResults.GetLength(0); i++)
{
var input = ArrayHelper.GetRow(dataSets, i).ToList();
var result = network.Activate(input);
results.Add(result);
}
Console.WriteLine("Results:");
for (var i = 0; i < results.Count; i++)
{
for (var j = 0; j < results[i].Count; j++)
{
var expected = Math.Round(expectedResults[i, j]);
var actual = Math.Round(results[i][j], 1);
Console.WriteLine($"{expected} = {actual}; {expected == actual}");
}
Console.WriteLine();
}
Console.ReadKey();
}
static void Main(string[] args)
{
Console.SetWindowSize(width, height);
Console.SetBufferSize(width, height + 500);
Design.Headline(Application.ProductName.Replace('_', ' '));
int inputSize = 10;
BaseTrans[] inputs = new BaseTrans[inputSize];
NeuralNetwork nn = new NeuralNetwork();
Random r = new Random();
for (int i = 0; i < inputSize; i++)
{
inputs[i] = new BaseTrans();
inputs[i].Value = r.Next(100);
nn.AddInput(inputs[i]);
}
BaseRec r1 = new BaseRec();
//BaseReciever r2 = new BaseReciever
nn.AddOutput(r1);
nn.AddOutput(r1);
//nn.AddOutput(r2);
//int iterationNum = 10000;
//double sum = 0
nn.Build();
LearningSet ls = nn.GenerateLearningSet();
foreach (BaseTrans input in inputs)
{
ls.GiveInput(input, r.Next(100));
}
ls.ExpectOutput(r1, 100);
nn.Practice(ls);
nn.Activate();
//Console.WriteLine(r1.Value);
//sum += r1.Value;
//Console.WriteLine(r1.Value);
//Console.WriteLine(sum / iterationNum);
//Design.End();
}
public void Forward(int[] inp)
{
GraphCPU graph = new GraphCPU(false);
int indOld = 10;
network.ResetState();
for (int i = 0; i < inp.Length; i++)
{
NNValue valueMatrix = new NNValue(DataSetSeq2Seq.GetValue(inp[i], 11));
network.Activate(valueMatrix, graph);
}
for (int i = 0; i < inp.Length; i++)
{
NNValue valueMatrix = new NNValue(DataSetSeq2Seq.GetValue(indOld, 11));
indOld = GetInd(network.Activate(valueMatrix, graph));
Console.Write(indOld);
}
Console.WriteLine();
}
public void Run()
{
Stopwatch watch = Stopwatch.StartNew();
NeuralNetwork network = new NeuralNetwork();
network.AddLayer("input", new InputLayer(2), BaseLayer.TYPE.INPUT);
network.AddLayer("hidden", new CoreLayer(8, ACTIVATION.SIGMOID, BaseLayer.TYPE.HIDDEN), BaseLayer.TYPE.HIDDEN);
network.AddLayer("output", new CoreLayer(1, ACTIVATION.SIGMOID, BaseLayer.TYPE.OUTPUT), BaseLayer.TYPE.OUTPUT);
network.AddConnection("input", "hidden", Connection.INIT.GLOROT_UNIFORM);
network.AddConnection("hidden", "output", Connection.INIT.GLOROT_UNIFORM);
/*
* Optimizer optimizer = new BackProp(network, 1e-5f, 0.99f, true)
* {
* Alpha = 0.1f
* };
*/
Optimizer optimizer = new RMSProp(network)
{
Alpha = 0.1f
};
optimizer.InitBatchMode(4);
Vector[] input = new Vector[4];
Vector[] target = new Vector[4];
//Vector output = null;
input[0] = Vector.Build(2, new float[] { 0f, 0f });
input[1] = Vector.Build(2, new float[] { 0f, 1f });
input[2] = Vector.Build(2, new float[] { 1f, 0f });
input[3] = Vector.Build(2, new float[] { 1f, 1f });
target[0] = Vector.Build(1, new float[] { 0f });
target[1] = Vector.Build(1, new float[] { 1f });
target[2] = Vector.Build(1, new float[] { 1f });
target[3] = Vector.Build(1, new float[] { 0f });
for (int e = 0; e < 200; e++)
{
//Console.Write("Start ");
//BasePool.Instance.Check();
float err = 0;
for (int i = 0; i < 4; i++)
{
err += optimizer.Train(input[i], target[i]);
}
Console.WriteLine(err);
//Console.Write("End ");
//BasePool.Instance.Check();
}
Console.WriteLine();
for (int i = 0; i < 4; i++)
{
Console.WriteLine(network.Activate(input[i])[0]);
Vector.Release(input[i]);
Vector.Release(target[i]);
}
optimizer.Dispose();
Console.Write("Finish ");
BasePool.Instance.Check();
watch.Stop();
Console.WriteLine(watch.ElapsedMilliseconds);
}