private static void Main(string[] args)
{
// species a 2-layer neural network with one hidden layer of 20 neurons
var net = new Net();
// input layer declares size of input. here: 2-D data
// ConvNetJS works on 3-Dimensional volumes (width, height, depth), but if you're not dealing with images
// then the first two dimensions (width, height) will always be kept at size 1
net.AddLayer(new InputLayer(1, 1, 2));
// declare 20 neurons, followed by ReLU (rectified linear unit non-linearity)
net.AddLayer(new FullyConnLayer(20, Activation.Relu));
// declare the linear classifier on top of the previous hidden layer
net.AddLayer(new SoftmaxLayer(10));
// forward a random data point through the network
var x = new Volume(new[] {0.3, -0.5});
var prob = net.Forward(x);
// prob is a Volume. Volumes have a property Weights that stores the raw data, and WeightGradients that stores gradients
Console.WriteLine("probability that x is class 0: " + prob.Weights[0]); // prints e.g. 0.50101
var trainer = new Trainer(net) {LearningRate = 0.01, L2Decay = 0.001};
trainer.Train(x, 0); // train the network, specifying that x is class zero
var prob2 = net.Forward(x);
Console.WriteLine("probability that x is class 0: " + prob2.Weights[0]);
// now prints 0.50374, slightly higher than previous 0.50101: the networks
// weights have been adjusted by the Trainer to give a higher probability to
// the class we trained the network with (zero)
}
private static void Regression1DDemo()
{
var net = new Net();
net.AddLayer(new InputLayer(1, 1, 1));
net.AddLayer(new FullyConnLayer(20, Activation.Relu));
net.AddLayer(new FullyConnLayer(20, Activation.Sigmoid));
net.AddLayer(new RegressionLayer(1));
var trainer = new Trainer(net) { LearningRate = 0.01, Momentum = 0.0, BatchSize = 1, L2Decay = 0.001 };
// Function we want to learn
double[] x = { 0.0, 0.5, 1.0 };
double[] y = { 0.0, 0.1, 0.2 };
var n = x.Length;
// Training
do
{
RegressionUpdate(n, x, trainer, y);
} while (!Console.KeyAvailable);
// Testing
var netx = new Volume(1, 1, 1);
for (var ix = 0; ix < n; ix++)
{
netx.Weights = new[] { x[ix] };
var result = net.Forward(netx);
}
}
private static void Classify2DDemo()
{
var net = new Net();
net.AddLayer(new InputLayer(1, 1, 2));
net.AddLayer(new FullyConnLayer(6, Activation.Tanh));
net.AddLayer(new FullyConnLayer(2, Activation.Tanh));
net.AddLayer(new SoftmaxLayer(2));
var trainer = new Trainer(net) { LearningRate = 0.01, Momentum = 0.0, BatchSize = 10, L2Decay = 0.001 };
// Data
var data = new List<double[]>();
var labels = new List<int>();
data.Add(new[] { -0.4326, 1.1909 });
labels.Add(1);
data.Add(new[] { 3.0, 4.0 });
labels.Add(1);
data.Add(new[] { 0.1253, -0.0376 });
labels.Add(1);
data.Add(new[] { 0.2877, 0.3273 });
labels.Add(1);
data.Add(new[] { -1.1465, 0.1746 });
labels.Add(1);
data.Add(new[] { 1.8133, 1.0139 });
labels.Add(0);
data.Add(new[] { 2.7258, 1.0668 });
labels.Add(0);
data.Add(new[] { 1.4117, 0.5593 });
labels.Add(0);
data.Add(new[] { 4.1832, 0.3044 });
labels.Add(0);
data.Add(new[] { 1.8636, 0.1677 });
labels.Add(0);
data.Add(new[] { 0.5, 3.2 });
labels.Add(1);
data.Add(new[] { 0.8, 3.2 });
labels.Add(1);
data.Add(new[] { 1.0, -2.2 });
labels.Add(1);
var n = labels.Count;
// Training
do
{
Classify2DUpdate(n, data, trainer, labels);
} while (!Console.KeyAvailable);
// Testing
var netx = new Volume(1, 1, 1);
for (var ix = 0; ix < n; ix++)
{
netx.Weights = data[ix];
var result = net.Forward(netx);
var c = net.GetPrediction();
bool accurate = c == labels[ix];
}
}
void Update()
{
for (var i = 0; i < iteration; ++i)
{
foreach (Vector3 v in data)
{
trainer.Train(new Volume(new double[] { v.x, v.y }), v.z);
}
}
if (Input.GetMouseButtonDown(0))
{
Ray ray = Camera.main.ScreenPointToRay(Input.mousePosition);
var pos = ray.origin;
pos /= 10.0f;
pos.z = 0;
data.Add(pos);
}
if (Input.GetMouseButtonDown(1))
{
Ray ray = Camera.main.ScreenPointToRay(Input.mousePosition);
var pos = ray.origin;
pos /= 10.0f;
pos.z = 1;
data.Add(pos);
}
for (var i = 0; i < pixels.GetLength(0); ++i)
{
for (var j = 0; j < pixels.GetLength(1); ++j)
{
Volume action = net.Forward(new Volume(new double[] { i / 10f, j / 10f }));
double output = action.Weights[0];
if (output > 0.5)
{
pixels [i, j].GetComponent <MeshRenderer> ().material.color = new Color(1, 0, 0);
}
else
{
pixels [i, j].GetComponent <MeshRenderer> ().material.color = new Color(0, 0, 1);
}
}
}
}
public void learn(float r1, bool food)
{
// perform an update on Q function
if (r0 != -99f)
{
// learn from this tuple to get a sense of how "surprising" it is to the agent
double tderror = learnFromTuple(s0, a0, r0, s1, a1);
entropyNum++;
/*
*
* entropyAverage = entropyAverage + (1 / (double)entropyNum) * (entropy - entropyAverage);
*
*
* double entropyDifference = entropy - entropyAverage;
*
* if (entropyDifference < 0)
* entropyDifference = 0;
*
*/
/*
*
* if( entropyDifference * alpha > 0.5f )
* Debug.Log( entropyDifference * alpha );
*
*
*
*
* if (entropyNum >= 500)
* {
*
* entropyAverage = entropyNum = 0;
* }
*
*/
double[] state = new double[2];
state[0] = e.getEntropySingle(s0);
state[1] = e.getEntropySingle(s1);
/* printArrayDouble(s0);
*
* printArrayDouble(state);
*
* Debug.Log(s0.Length + " " + state.Length);
*
*/
double resultClass = 1f;
//bool sample = true;
if (t > 2)
{
Volume resultClassify = netClassify.Forward(new Volume(state));
resultClass = resultClassify.Weights[0];
//sample = resultClassify.Weights[0] <= resultClassify.Weights[1];
}
//Debug.Log(resultClass);
// decide if we should keep this experience in the replay
//if (food || Random.Range(0f, 1f) < resultClass)
//if (entropy > 0.7 || r0 != 0 || t == 0)
//if(sample)
if (t % experienceAddEvery == 0)
{
exp[expi] = new Experience(s0, a0, r0, s1, a1);
expi += 1;
if (expn < experienceSize)
{
expn += 1;
}
if (expi >= experienceSize)
{
expi = 0;
} // roll over when we run out
}
t += 1;
//Debug.Log("length: " + exp.Length);
// sample some additional experience from replay memory and learn from it
int k = 0;
while (k < learningStepsPerIteration)
{
Experience experience = exp[Random.Range(0, expn)];
double[] s = new double[2];
s[0] = e.getEntropySingle(experience.s0);
s[1] = e.getEntropySingle(experience.s1);
Volume resultClassify = netClassify.Forward(new Volume(s));
double result = 1f;
result = resultClassify.Weights[0];
if (Random.Range(0f, 1f) < result)
{
learnFromTuple(experience.s0, experience.a0, experience.r0, experience.s1, experience.a1);
k++;
}
}
}
r0 = r1; // store for next update
}
double[] forwardQ(double[] s)
{
Volume result = net.Forward(new Volume(s));
return(result.Weights);
}
