Click or drag to resize

BackpropagationCalcError Method

Calculate new error values for the output layer.

Namespace:  RavingBots.MagicGestures.AI.Neural.Classic
Assembly:  Assembly-CSharp (in Assembly-CSharp.dll) Version: 0.0.0.0
Syntax
C#
Copy
public static float CalcError(
	MultilayerPerceptron mlp,
	float[] sampleOutput
)

Parameters

mlp
Type: RavingBots.MagicGestures.AI.Neural.ClassicMultilayerPerceptron
The network being trained.
sampleOutput
Type: SystemSingle
The expected output values.

Return Value

Type: Single
The mean sample error.
Remarks

The error value for each neuron on the output layer is set based on the derivative of that neuron's activation function (Activation) and the difference between expected output value and the actual one.

See Also