public static double[] prediction(NeuralNetwork network, double[] inputValues)
{
FSharpList <double> ivs = ListModule.OfSeq(inputValues);
return(NeuralNetworkModule
.prediction(network.contents, ivs)
.ToArray());
}
public static NeuralNetwork initWithSeed(int seed, int[] layers)
{
FSharpList <int> ls = ListModule.OfSeq(layers);
LazyList <LazyList <Neuron.Neuron> > _contents =
NeuralNetworkModule.initWithSeed(seed, ls);
return(new NeuralNetwork(_contents));
}
public static NeuralNetwork fit(
NeuralNetwork network,
double learningRate,
double[] inputValues,
double[] expectedOutput)
{
FSharpList <double> ivs = ListModule.OfSeq(inputValues);
FSharpList <double> eos = ListModule.OfSeq(expectedOutput);
LazyList <LazyList <Neuron.Neuron> > _contents =
NeuralNetworkModule.fit(network.contents, learningRate, ivs, eos);
return(new NeuralNetwork(_contents));
}
public static double[] errors(
NeuralNetwork network,
double learningRate,
double[] inputValues,
double[] expectedOutput)
{
FSharpList <double> ivs = ListModule.OfSeq(inputValues);
FSharpList <double> eos = ListModule.OfSeq(expectedOutput);
return(NeuralNetworkModule
.errors(network.contents, learningRate, ivs, eos)
.ToArray());
}
public static NeuralNetwork customizedInit(
int[] layers,
Func <int, ActivationFunction> activationF,
Func <int, Double> weightInitF)
{
FSharpList <int> ls = ListModule.OfSeq(layers);
FSharpFunc <int, Activation.Function> actF =
(Converter <int, Activation.Function>)(x => activationF(x).contents);
FSharpFunc <int, Double> weightF =
(Converter <int, Double>)(x => weightInitF(x));
LazyList <LazyList <Neuron.Neuron> > _contents =
NeuralNetworkModule.customizedInit(ls, actF, weightF);
return(new NeuralNetwork(_contents));
}
public static NeuralNetwork ofJson(string json)
{
LazyList <LazyList <Neuron.Neuron> > _contents = NeuralNetworkModule.ofJson(json);
return(new NeuralNetwork(_contents));
}
public static string toJson(NeuralNetwork network)
{
return(NeuralNetworkModule.toJson(network.contents));
}
