org.encog.util.normalize.output
Interface OutputField
-
- All Superinterfaces:
- java.io.Serializable
- All Known Implementing Classes:
- BasicOutputField, OutputEquilateral, OutputFieldDirect, OutputFieldEncode, OutputFieldGrouped, OutputFieldMultiplicative, OutputFieldRangeMapped, OutputFieldZAxis, OutputFieldZAxisSynthetic, OutputOneOf
public interface OutputField extends java.io.SerializableAn output field, this represents the actual output from the normalization. Output from the normalization class is usually input to a neural network. An output field may contain several subfields that will be generated. Call getSubfieldCount to determine how many fields will be generated. A simple field will return 1, indicating that this is a single field.
-
-
Method Summary
All Methods Instance Methods Abstract Methods Modifier and Type Method and Description doublecalculate(int subfield)Calculate the value for this field.intgetSubfieldCount()booleanisIdeal()voidrowInit()Init this field for a new row.voidsetIdeal(boolean ideal)Set whether this field is part of the ideal output for a network.
-
-
-
Method Detail
-
calculate
double calculate(int subfield)
Calculate the value for this field. Specify subfield of zero if this is a simple field.- Parameters:
subfield- The subfield index.- Returns:
- The calculated value for this field.
-
getSubfieldCount
int getSubfieldCount()
- Returns:
- The numebr of fields that will actually be generated by this field. For a simple field, this value is 1.
-
isIdeal
boolean isIdeal()
- Returns:
- Is this field part of the ideal data uses to train the neural network.
-
rowInit
void rowInit()
Init this field for a new row.
-
setIdeal
void setIdeal(boolean ideal)
Set whether this field is part of the ideal output for a network.- Parameters:
ideal- True if this field is ideal.
-
-
DMelt 3.0 © DataMelt by jWork.ORG