TfannNetwork class
A structure of a complete fully-connected network.
Constructors
-
TfannNetwork(List<
TfannLayer> layers) -
TfannNetwork.full(List<
int> layersDefinition, List<ActivationFunctionType> activation)
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
-
layers
↔ List<
TfannLayer> -
getter/setter pair
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
calculateMeanAbsoluteError(
List< TrainSetInputOutput> data) → FVector - Given a single training pair, calculate the network's mean-absolute-error per output node
-
calculateMeanSquareError(
List< TrainSetInputOutput> data) → FVector - Given a single training pair, calculate the network's mean-square-error.
-
feedForward(
FVector input) → FVector - Get the network's output vector from the input vector.
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
save(
String filename) → Future< void> - Saves the network to file.
-
toJson(
) → Object - Returns Json object.
-
toString(
) → String -
A string representation of this object.
inherited
-
train(
TrainSet trainSet, {double learningRate = 0.04, double maxErrClipAbove = 0.0, double skipIfErrBelow = 0.0, bool skipIfOutput(FVector)?}) → TrainArtifacts - Train network with a single training pair, for a single epoch.
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited
Static Methods
-
fromFile(
String filename) → TfannNetwork? -
fromJson(
dynamic jsonObject) → TfannNetwork? - Creates TfannNetwork from Json object.