public void AddOutputLayer_ShouldAddOutputLayer()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func1 = new ActivationFunction(x => x, x => 1);
ActivationFunction func2 = new ActivationFunction(x => 2 * x, x => 1);
builder.AddInputLayer(3, func1, false)
.AddOutputLayer(2, func2);
Assert.AreEqual(2, network1.Layers.Count);
Assert.IsNotNull(network1.InputLayer);
Assert.AreEqual(3, network1.InputLayer.Inputs.Length);
Assert.AreEqual(func1, network1.InputLayer.ActivationFunction);
Assert.IsNotNull(network1.InputLayer.Weights);
Assert.AreEqual(3, network1.InputLayer.Weights.GetLength(0));
Assert.AreEqual(2, network1.InputLayer.Weights.GetLength(1));
Assert.IsNotNull(network1.OutputLayer);
Assert.AreEqual(2, network1.OutputLayer.Inputs.Length);
Assert.AreEqual(func2, network1.OutputLayer.ActivationFunction);
Assert.IsNull(network1.OutputLayer.Weights);
}
public void GetNetwork_WithNetworkWithoutInputLayer_ShouldThrowException()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
AssertEx.AssertThrows <NeuralNetworkConfigurationException>(() => builder.GetNetwork(), "can't get not set up neural network");
}
public void AddOutputLayer_WithoutInputLayer_ShouldThrowException()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func = ActivationFunctions.Identity;
AssertEx.AssertThrows <NeuralNetworkConfigurationException>(() => builder.AddOutputLayer(2, func), "can't have only output layer");
}
public void AddInputLayer_CalledTwice_ShouldThrowException()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func = ActivationFunctions.Identity;
builder.AddInputLayer(3, func, true);
AssertEx.AssertThrows <NeuralNetworkConfigurationException>(() => builder.AddInputLayer(2, func, true), "can't have 2 input layers");
}
public void GetNetwork_WithNetworkWithoutOutputLayer_ShouldThrowException()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
ActivationFunction func = ActivationFunctions.Identity;
builder.AddInputLayer(3, func, false);
AssertEx.AssertThrows <NeuralNetworkConfigurationException>(() => builder.GetNetwork(), "can't get not set up neural network");
}
public void GetNetwork_WithSetUpNetwork()
{
NeuralNetwork neuralNetworkMock = new NeuralNetworkMock();
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CurrentNeuralNetwork = neuralNetworkMock;
var network = builder.GetNetwork();
Assert.IsNotNull(network);
Assert.AreSame(neuralNetworkMock, network);
Assert.IsTrue(((NeuralNetworkMock)neuralNetworkMock).InitilizeIsCalled);
}
public void AddHiddenLayer_CalledAfterAddOutpout_ShouldThrowException()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func = ActivationFunctions.Identity;
builder.AddInputLayer(3, func, false)
.AddOutputLayer(4, func);
AssertEx.AssertThrows <NeuralNetworkConfigurationException>(() => builder.AddHiddenLayer(5, func, false), "can't have hidden layer after the output layer");
}
public void AddInputLayerWithBias_ShouldAddBiasNode()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func = ActivationFunctions.Identity;
builder.AddInputLayer(3, func, true);
Assert.AreEqual(1, network1.Layers.Count);
Assert.IsNotNull(network1.InputLayer);
Assert.AreEqual(4, network1.InputLayer.Inputs.Length);
Assert.AreEqual(func, network1.InputLayer.ActivationFunction);
Assert.IsNull(network1.InputLayer.Weights);
}
public void AddLayers_ShouldSetProperLayersAsInputAndOutput()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew()
.AddInputLayer(2, new ActivationFunction(x => x, x => 1), false)
.AddHiddenLayer(3, new ActivationFunction(x => 2 * x, x => 1), false)
.AddHiddenLayer(4, new ActivationFunction(x => 3 * x, x => 1), false)
.AddOutputLayer(1, new ActivationFunction(x => 4 * x, x => 1));
NeuralNetwork network = builder.CurrentNeuralNetwork;
Assert.IsNotNull(network);
Assert.AreEqual(4, network.Layers.Count);
Assert.AreEqual(network.Layers[0], network.InputLayer);
Assert.AreEqual(network.Layers[3], network.OutputLayer);
}
public void CreateNew_ShouldPrepareANewInstance()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
Assert.IsNotNull(network1);
builder.CreateNew();
NeuralNetwork network2 = builder.CurrentNeuralNetwork;
Assert.IsNotNull(network2);
Assert.AreNotSame(network1, network2);
}
public void AddHiddenLayer_WithInputLayer_ShouldAddHiddenLayer()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func = ActivationFunctions.Identity;
builder.AddInputLayer(3, func, false)
.AddHiddenLayer(4, func, false);
Assert.AreEqual(2, network1.Layers.Count);
Assert.AreEqual(4, network1.Layers[1].Inputs.Length);
Assert.AreEqual(func, network1.Layers[1].ActivationFunction);
Assert.IsNull(network1.Layers[1].Weights);
Assert.IsNotNull(network1.InputLayer.Weights);
Assert.AreEqual(3, network1.InputLayer.Weights.GetLength(0));
Assert.AreEqual(4, network1.InputLayer.Weights.GetLength(1));
}
public void AddHiddenLayer_ThreeTimes_ShouldAddThreeHiddenLayers()
{
NeuralNetworkBuilderTestDouble builder = new NeuralNetworkBuilderTestDouble();
builder.CreateNew();
NeuralNetwork network1 = builder.CurrentNeuralNetwork;
ActivationFunction func1 = new ActivationFunction(x => x, x => 1);
ActivationFunction func2 = new ActivationFunction(x => 2 * x, x => 1);
ActivationFunction func3 = new ActivationFunction(x => 3 * x, x => 1);
ActivationFunction func4 = new ActivationFunction(x => 4 * x, x => 1);
builder.AddInputLayer(2, func1, false)
.AddHiddenLayer(3, func2, false)
.AddHiddenLayer(4, func3, false)
.AddHiddenLayer(5, func4, false);
Assert.AreEqual(4, network1.Layers.Count);
Assert.IsNotNull(network1.InputLayer.Weights);
Assert.AreEqual(2, network1.InputLayer.Weights.GetLength(0));
Assert.AreEqual(3, network1.InputLayer.Weights.GetLength(1));
NeuralNetworkLayer hidden1 = network1.Layers[1];
Assert.AreEqual(3, hidden1.Inputs.Length);
Assert.AreEqual(func2, hidden1.ActivationFunction);
Assert.IsNotNull(hidden1.Weights);
Assert.AreEqual(3, hidden1.Weights.GetLength(0));
Assert.AreEqual(4, hidden1.Weights.GetLength(1));
NeuralNetworkLayer hidden2 = network1.Layers[2];
Assert.AreEqual(4, hidden2.Inputs.Length);
Assert.AreEqual(func3, hidden2.ActivationFunction);
Assert.IsNotNull(hidden2.Weights);
Assert.AreEqual(4, hidden2.Weights.GetLength(0));
Assert.AreEqual(5, hidden2.Weights.GetLength(1));
NeuralNetworkLayer hidden3 = network1.Layers[3];
Assert.AreEqual(5, hidden3.Inputs.Length);
Assert.AreEqual(func4, hidden3.ActivationFunction);
Assert.IsNull(hidden3.Weights);
}
