static async Task Main(string[] args)
{
var csvReader = new CsvReader();
var trainSets = await csvReader.Read(TrainFile);
var testSets = await csvReader.Read(TestFile);
var nn = new Network(784, 30, 10);
var bp = new BackpropagationAlgorithm(nn);
var error = 0.0;
var count = 0;
do
{
count++;
var sw = Stopwatch.StartNew();
error = bp.Train(trainSets, testSets);
sw.Stop();
Console.WriteLine(sw.ElapsedMilliseconds);
Console.WriteLine($"Count: {count} - MSE: {error}.");
} while (error > 0.005);
Console.WriteLine(nn.Save());
Console.ReadLine();
}
private void startTraining()
{
stopSignal = false;
if (Multithreading)
{
backpropagationAlgorithm = new ParallelBackpropagation(
NeuralNet, DataItems, LearningRate, Momentum, ErrorThreshold, Environment.ProcessorCount);
}
else
{
backpropagationAlgorithm = new Backpropagation(
NeuralNet, DataItems, LearningRate, Momentum, ErrorThreshold);
}
thread = new Thread(new ThreadStart(trainNetwork));
thread.Start();
}
public static void TestBackpropagation()
{
var NN = NetworkCreator.Perceptron(2, 1, new [] { 20, 20, 20 });
var nums = 100000;
var data = new Dictionary <List <double>, List <double> >();
var rand = new Random();
var dif = 1000;
var shell = 500;
for (var i = 0; i < nums; i++)
{
var first = rand.Next(dif) - shell;
var second = rand.Next(dif) - shell;
data.Add(new List <double>()
{
first, second
},
new List <double>()
{
(first + second + 2 * shell) / 2d / dif,
});
}
BackpropagationAlgorithm.Teach(NN, data);
NN.Print();
for (var i = 0; i < 100; i++)
{
var first = rand.Next(dif) - shell;
var second = rand.Next(dif) - shell;
NN.Work(new List <double>()
{
first, second
});
var res = NN.GetResults()[0];
Console.WriteLine(first + " " + second + " . " + (first + second) + " " + (Math.Round((res * dif - shell) * 2)));
Console.ReadLine();
}
}
