public NeuralNetworkImage(
Layer[] layers, IErrorFunction error_fnc,
IDataConvertor input_convertor, IDataConvertor output_convertor,
IRegularization regularization)
{
CheckImageError(layers, error_fnc);
this.layers = layers;
this.error_fnc = error_fnc;
this.input_convertor = input_convertor;
this.output_convertor = output_convertor;
this.regularization = regularization;
}
private NeuralNetworkImage CloseCurrentImage()
{
var image = new NeuralNetworkImage(
layers.ToArray(),
error_func, in_cvrt, out_cvrt, regularization);
last_layer_input_count = 0;
layers = null;
error_func = null;
in_cvrt = out_cvrt = null;
regularization = null;
return(image);
}
public void Serialize(IRegularization regularization)
{
ushort code;
byte[] parameters;
if (regularization == null)
{
code = 0;
parameters = null;
}
else
{
switch (regularization)
{
case RegularizationL1 _:
code = 1;
parameters = null;
break;
case RegularizationL2 _:
code = 2;
parameters = null;
break;
default:
var attr = GetAttribute(regularization.GetType());
if (!registered_functions_via_code.ContainsKey(attr.code | REGULARIZATION_SIGN))
{
throw new ArgumentException(
nameof(regularization), "this type of IRegularization is not registered.");
}
var serializer = registered_functions_via_code[attr.code | REGULARIZATION_SIGN];
code = serializer.Code;
parameters = serializer.Serialize(regularization);
if ((parameters?.Length ?? 0) != serializer.ParameterLength)
{
throw new Exception("invalid parameters' length.");
}
break;
}
}
// serialaize type and parameters
Serialize(code, parameters);
}
public NeuralNetworkInitializer SetCorrection(
IErrorFunction error_func, IRegularization regularization = null)
{
if (layers == null)
{
throw new Exception("The layers input is not set yet.");
}
if (layers.Count < 1)
{
throw new Exception("The layers output is not set yet.");
}
this.error_func = error_func ??
throw new ArgumentNullException(nameof(error_func),
"The error function is undefined.");
last_layer_input_count = -1;
this.regularization = regularization;
return(this);
}
