IEnumerator AnimationRoutine()
{
LayerNeuron inputLayer = layers[0];
LayerNeuron firstHiddenLayer = layers[1];
for (int j = 0; j < firstHiddenLayer.neurons.Length; j++)
{
for (int i = 0; i < inputLayer.neurons.Length; i++)
{
//On grossie le neuron assez de fois pour qu'il se fasse absorbé par les nexts layers
Neuron currentNeuron = inputLayer.neurons[i];
currentNeuron.Growth();
}
yield return(new WaitForSeconds(timeBetweenNeuronConnection));
}
for (int i = 0; i < layers.Length; i++)
{
LayerNeuron currentLayer = layers[i];
for (int j = 0; j < currentLayer.neurons.Length; j++)
{
Neuron currentNeuron = currentLayer.neurons[j];
int connectionPerNeuron = currentNeuron.incomingConnections.Length;
for (int k = 0; k < connectionPerNeuron; k++)
{
currentNeuron.PlayAnimationWithConnection(k);
yield return(new WaitForSeconds(timeBetweenNeuronConnection));
}
yield return(new WaitForSeconds(timeBetweenNeuronConnection));
}
yield return(new WaitForSeconds(timeBetweenNeuronConnection));
}
LayerNeuron outputLayer = layers[layers.Length - 1];
LayerNeuron lastHiddenLayer = layers[layers.Length - 2];
for (int j = 0; j < lastHiddenLayer.neurons.Length; j++)
{
for (int i = 0; i < outputLayer.neurons.Length; i++)
{
//On grossie le neuron assez de fois pour qu'il se fasse absorbé par les nexts layers
Neuron currentNeuron = outputLayer.neurons[i];
currentNeuron.Shrink();
}
yield return(new WaitForSeconds(timeBetweenNeuronConnection));
}
StartCoroutine(AnimationRoutine());
}
void CreateConnections()
{
//On commence a la première hidden layer, donc i = 1
for (int i = 1; i < layers.Length; i++)
{
LayerNeuron previousLayer = layers[i - 1];
LayerNeuron currentLayer = layers[i];
for (int j = 0; j < currentLayer.neurons.Length; j++)
{
//Create connections
Neuron currentNeuron = currentLayer.neurons[j];
currentNeuron.incomingConnections = new Connection[previousLayer.neurons.Length];
for (int k = 0; k < currentNeuron.incomingConnections.Length; k++)
{
CreateSingleConnection(i, previousLayer, currentLayer, currentNeuron, k);
}
}
}
}
void GenerateNeuralNetwork()
{
for (int i = 0; i < layerDimension.Length; i++)
{
GameObject layer = new GameObject("Layer" + i);
LayerNeuron layerNeuron = layer.AddComponent <LayerNeuron>();
layer.transform.SetParent(transform);
layer.transform.localScale = Vector3.one;
for (int j = 0; j < layerDimension[i]; j++)
{
Neuron newNeuron = Instantiate(prefabNeuron, layer.transform);
newNeuron.transform.localScale = Vector3.one;
newNeuron.transform.localPosition = CalculateNeuronPosition(i, j, layerDimension[i]);
}
layerNeuron.Initiate();
}
layers = GetComponentsInChildren <LayerNeuron>();
}
private void CreateSingleConnection(int i, LayerNeuron previousLayer, LayerNeuron currentLayer, Neuron currentNeuron, int k)
{
//On spawn autant de connection quia de neuron au previous layer
Connection newConnection = Instantiate(prefabConnection, currentNeuron.transform);
currentNeuron.incomingConnections[k] = newConnection;
newConnection.transform.localScale = Vector3.one;
//On connect tous les neurons du previous layer au current neuron
newConnection.from = previousLayer.neurons[k];
newConnection.to = currentNeuron;
float layerT = (float)i / layers.Length;
float neuronT = (float)k / currentLayer.neurons.Length;
Color layerColor = gradientLayer.Evaluate(layerT);
Color neuronColor = gradientNeuron.Evaluate(neuronT);
Color startColor = (layerColor * weightColorLayer + neuronColor * (1 - weightColorLayer));
newConnection.enableColor = startColor;
newConnection.disableColor = endColor;
newConnection.ConnectionFromTo();
}
