public void FeedForwardTest()
{
var outputs = new double[] { 0, 0, 1, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1 };
var inputs = new double[, ]
{
// Result - sick = 1
// healthy = 0
//Wrong Temperature T
//Good age A
//Smoking S
//Good food F
//T A S F
{ 0, 0, 0, 0 },
{ 0, 0, 0, 1 },
{ 0, 0, 1, 0 },
{ 0, 0, 1, 1 },
{ 0, 1, 0, 0 },
{ 0, 1, 0, 1 },
{ 0, 1, 1, 0 },
{ 0, 1, 1, 1 },
{ 1, 0, 0, 0 },
{ 1, 0, 0, 1 },
{ 1, 0, 1, 0 },
{ 1, 0, 1, 1 },
{ 1, 1, 0, 0 },
{ 1, 1, 0, 1 },
{ 1, 1, 1, 0 },
{ 1, 1, 1, 1 }
};
//arrange
var topology = new Topology(4, 1, 0.1, 2);
var neuralNetwork = new NeuralNetWork(topology);
var results = new List <double>();
//act
var difference = neuralNetwork.Learn(outputs, inputs, 10000);
for (int i = 0; i < outputs.Length; i++)
{
var row = NeuralNetWork.GetRow(inputs, i);
results.Add(neuralNetwork.FeedForward(row).Output);
}
//assert
for (int i = 0; i < results.Count; i++)
{
var expected = Math.Round(outputs[i], 2);
var actual = Math.Round(results[i], 2);
Assert.AreEqual(expected, actual);
}
}
public void DataSetTest()
{
//Read
var inputs = new List <double[]>();
var outputs = new List <double>();
using (var sr = new StreamReader("datasets_33180_43520_heart.csv"))
{
var header = sr.ReadLine();
while (!sr.EndOfStream)
{
var row = sr.ReadLine();
var temp = row.Split(',');
var values = row.Split(',').Select(v => Convert.ToDouble(v)).ToList();
var output = values.Last();
var input = values.Take(values.Count - 1).ToArray();
outputs.Add(output);
inputs.Add(input);
}
}
var inputSignals = new double[inputs.Count, inputs[0].Length];
for (int i = 0; i < inputSignals.GetLength(0); i++)
{
for (int j = 0; j < inputSignals.GetLength(1); j++)
{
inputSignals[i, j] = inputs[i][j];
}
}
//arrange
var topology = new Topology(outputs.Count, 1, 0.1, outputs.Count / 2);
var neuralNetwork = new NeuralNetWork(topology);
var results = new List <double>();
//Act
var difference = neuralNetwork.Learn(outputs.ToArray(), inputSignals, 10);
for (int i = 0; i < outputs.Count; i++)
{
results.Add(neuralNetwork.FeedForward(inputs[i]).Output);
}
//Assert
for (int i = 0; i < results.Count; i++)
{
var expected = Math.Round(outputs[i], 2);
var actual = Math.Round(results[i], 2);
Assert.AreEqual(expected, actual);
}
}
public void RecognizeImageTest()
{
//Arrange
var size = 1000;
var parasitizedPath = @"E:\Downloads_chrome\cell_images\Parasitized\";
var unparasitizedPath = @"E:\Downloads_chrome\cell_images\Uninfected\";
var converter = new PictureConverter();
var testparasitizedImageInput = converter.Convert(@"E:\Projects C#\Project1 - Lessons\FirstNeuralNetwork\NeuralNetworkClassesTests\Images\Parasitized.png");
var testunparasitizedImageInput = converter.Convert(@"E:\Projects C#\Project1 - Lessons\FirstNeuralNetwork\NeuralNetworkClassesTests\Images\Unparasitized.png");
var topology = new Topology(testparasitizedImageInput.Count, 1, 0.1, testparasitizedImageInput.Count / 2);
var nuralNetwork = new NeuralNetWork(topology);
double[,] parasitizedInputs = GetData(parasitizedPath, converter, testparasitizedImageInput, size);
double[,] unparasitizedInputs = GetData(unparasitizedPath, converter, testunparasitizedImageInput, size);
//Act
nuralNetwork.Learn(new double[] { 1 }, parasitizedInputs, 2);
nuralNetwork.Learn(new double[] { 0 }, unparasitizedInputs, 2);
var par = nuralNetwork.FeedForward(testparasitizedImageInput.Select(t => (double)t).ToArray());
var unpar = nuralNetwork.FeedForward(testunparasitizedImageInput.Select(t => (double)t).ToArray());
//Assert
Assert.AreEqual(1, Math.Round(par.Output, 2));
Assert.AreEqual(0, Math.Round(unpar.Output, 2));
}