public void NetworkComputeTest()
{
//given
ILayer firstLayer = MockRepository.GenerateMock <ILayer>();
List <List <double> > firstLayerResults = new List <List <double> >
{
new List <double> {
1.0, 2.0, 3.0, 4.0
},
new List <double> {
1.0, 2.0, 3.0, 4.0
},
new List <double> {
1.0, 2.0, 3.0, 4.0
},
new List <double> {
1.0, 2.0, 3.0, 4.0
},
};
firstLayer.Stub(layer => layer.Compute(Arg <IEnumerable <IEnumerable <double> > > .Is.Anything, Arg <int> .Is.Anything)).Return(firstLayerResults);
firstLayer.Stub(layer => layer.Neurons).Return(new List <INeuron>
{
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>()
});
List <double> secondLayerResults = new List <double> {
1.0, 2.0, 3.0, 4.0, 5.0, 6.0
};
ILayer secondLayer = MockRepository.GenerateMock <ILayer>();
secondLayer.Stub(layer => layer.Compute(Arg <IEnumerable <IEnumerable <double> > > .Is.Anything)).Return(secondLayerResults);
secondLayer.Stub(layer => layer.Neurons).Return(new List <INeuron>
{
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
MockRepository.GenerateMock <INeuron>(),
});
int expectedNeuronNumber = secondLayer.Neurons.Count();
ILayerFactory layerFactory = MockRepository.GenerateMock <ILayerFactory>();
layerFactory.Stub(lf => lf.ConstructLayer(
Arg <double> .Is.Anything, Arg <int> .Is.Anything, Arg <int> .Is.Anything,
Arg <string> .Is.Equal("First"))
).Return(firstLayer);
layerFactory.Stub(lf => lf.ConstructLayer(
Arg <double> .Is.Anything, Arg <int> .Is.Anything, Arg <int> .Is.Anything,
Arg <string> .Is.Equal("Second"))
).Return(secondLayer);
List <List <double> > inputs = new List <List <double> > {
new List <double> {
1.0, 2.0
}, new List <double> {
1.0, 2.0
}
};
List <LayerDescription> layersDescriptions = new List <LayerDescription>
{
new LayerDescription {
Bias = 1.0, LayerFactory = layerFactory, Name = "First", NeuronNumber = 4
},
new LayerDescription {
Bias = 1.0, LayerFactory = layerFactory, Name = "Second", NeuronNumber = 6
}
};
Network network = new Network(2, layersDescriptions);
//when
List <double> results = network.Compute(inputs).ToList();
//then
CollectionAssert.AreEqual(secondLayerResults, results);
firstLayer.AssertWasCalled(layer => layer.Compute(Arg <IEnumerable <IEnumerable <double> > > .Is.Equal(inputs), Arg <int> .Is.Equal(expectedNeuronNumber)));
secondLayer.AssertWasCalled(layer => layer.Compute(Arg <IEnumerable <IEnumerable <double> > > .Is.Equal(firstLayerResults)));
}
