public Neuron(Layer entryNeurons) : this()
{
for (int i = 0; i < entryNeurons.Neurons.Count; i++)
{
var synapse = new Synapse(entryNeurons.Neurons[i], this);
entryNeurons.Neurons[i].Exit.Add(synapse);
Entry.Add(synapse);
}
}
public Neuron(IEnumerable <Neuron> entryNeuron) : this()
{
foreach (var x in entryNeuron)
{
var synapse = new Synapse(x, this);
x.Exit.Add(synapse);
Entry.Add(synapse);
}
}
public static Network NetworkImport(string path)
{
var data = GetNetwork(path);
if (data == null)
{
return(null);
}
Network network = new Network(true);
List <Neuron> neurons = new List <Neuron>();
network.LearningRate = data.LearningRate;
network.Momentum = data.Momentum;
foreach (var x in data.EntryLayer)
{
Neuron neuron = new Neuron
{
ID = x.ID,
Bias = x.Bias,
DeltaBias = x.DeltaBias,
Gradient = x.Gradient,
Value = x.Value
};
network.EntryLayer.Neurons.Add(neuron);
neurons.Add(neuron);
}
foreach (var x in data.HiddenLayers)
{
Layer neuronsHidden = new Layer();
foreach (var y in x)
{
Neuron neuron = new Neuron
{
ID = y.ID,
Bias = y.Bias,
DeltaBias = y.DeltaBias,
Gradient = y.Gradient,
Value = y.Value
};
neuronsHidden.Neurons.Add(neuron);
neurons.Add(neuron);
}
network.HiddenLayers.Add(neuronsHidden);
}
foreach (var x in data.ExitLayer)
{
Neuron neuron = new Neuron
{
ID = x.ID,
Bias = x.Bias,
DeltaBias = x.DeltaBias,
Gradient = x.Gradient,
Value = x.Value
};
network.ExitLayer.Neurons.Add(neuron);
neurons.Add(neuron);
}
foreach (var x in data.Synapses)
{
Synapse synapse = new Synapse {
ID = x.ID
};
Neuron entryNeuron = neurons.First(y => y.ID == x.EntryNeuronID);
Neuron exitNeuron = neurons.First(y => y.ID == x.ExitNeuronID);
synapse.EntryNeuron = entryNeuron;
synapse.ExitNeuron = exitNeuron;
synapse.Weight = x.Weight;
synapse.DeltaWeight = x.DeltaWeight;
entryNeuron.Exit.Add(synapse);
exitNeuron.Entry.Add(synapse);
}
return(network);
}
