NativeBackpropagation<P extends NativeSample> class

Backpropagation trainer accelerated by a native CPU/Metal backend.

Drop-in replacement for Backpropagation on Float32x4 networks. When no native backend is available it behaves exactly like Backpropagation.

Inheritance

• Object
• Training<double, Float32x4, SignalFloat32x4, Scale<double>, P>
• Backpropagation<double, Float32x4, SignalFloat32x4, Scale<double>, P>
• NativeBackpropagation

Constructors

NativeBackpropagation(ANN<double, Float32x4, SignalFloat32x4, Scale<double>> ann, SamplesSet<P> samplesSet, {NativeBackend backend = NativeBackend.auto, String? subject})

Properties

activeBackend → NativeBackend

The backend actually in use (NativeBackend.none when running pure-Dart).

no setter, inherited

algorithmName → String

The training algorithm name.

final, inherited

ann → ANN<double, Float32x4, SignalFloat32x4, Scale<double>>

The ANN to train.

final, inherited

backend → NativeBackend

final

bestTrainingError → double

The lowest training error seen by checkBestTrainingError since the last resetBestTraining.

no setter, inherited

elapsedTime → Duration?

no setter, inherited

enableSelectInitialANN ↔ bool

If true will select the initial ANN calling selectInitialANN.

getter/setter pair, inherited

endTime → DateTime?

The end time of the last training session or null if not finished yet.

no setter, inherited

globalError → double

Returns the current training global error (set by train).

no setter, inherited

globalLearnError → double

The global error while updating weights.

no setter, inherited

hashCode → int

The hash code for this object.

no setter, inherited

initialAnnEpochs ↔ int

Number of epochs to perform in the ANNs in the selection pool.

getter/setter pair, inherited

initialAnnPoolSize ↔ int

The initial ANN pool size.

getter/setter pair, inherited

isNativeAccelerated → bool

Whether a native backend is active for this trainer.

no setter, inherited

lastGlobalError → double

no setter, inherited

lastGlobalLearnError → double

The previous global error while updating weights.

no setter, inherited

learningRate → double

Returns the current learning rate of the Backpropagation.

no setter, inherited

learningRateEntry → Float32x4

no setter, inherited

logEnabled ↔ bool

If true logging will be enabled.

getter/setter pair, inherited

logger → TrainingLogger

final, inherited

logProgressEnabled ↔ bool

If true logging of progress will be enabled.

getter/setter pair, inherited

momentum → double

Returns the current momentum rate of the Backpropagation.

no setter, inherited

momentumEntry → Float32x4

no setter, inherited

noImprovementLimit → int

Limit of epochs without improvements. Used to trigger strategies.

no setter, inherited

noImprovementRatio ↔ double

Minimal improvement ratio.

getter/setter pair, inherited

parameters → String

no setter, inherited

random → Random

no setter, inherited

requestedBackend → NativeBackend

The requested backend (defaults to NativeBackend.auto).

no setter

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

samples → List<P>

Returns the samples of samplesSet

no setter, inherited

samplesSet → SamplesSet<P>

The samples set for training.

final, inherited

samplesSubject → String

Returns the subject of samplesSet

no setter, inherited

signalInstance → SignalFloat32x4

no setter, inherited

startTime → DateTime?

The start time of the last training session or null if reset.

no setter, inherited

subject → String

The training subject. Defaults to samplesSet.subject.

final, inherited

totalFailedEpochs → int

no setter, inherited

totalTrainedEpochs → int

Returns the total number of epochs of all the training session. A call to reset won't reset this value.

no setter, inherited

totalTrainingActivations → int

Returns the total number of activations of all the training session. A call to reset won't reset this value.

no setter, inherited

trainedEpochs → int

Returns the number of epochs of the last training session.

no setter, inherited

trainingActivations → int

Returns the number of activations of the last training session.

no setter, inherited

trainingSamplesSize → int

no setter, inherited

Methods

activateNative(SignalFloat32x4 input) → List<double>?

Runs a native forward pass (inference) for input and returns the output layer values, or null when no native backend is active.

inherited

backPropagateLastLayerError(Layer<double, Float32x4, SignalFloat32x4, Scale<double>> layer, int layerIndex, SignalFloat32x4 expected) → void

inherited

backPropagateMiddleLayerError(Layer<double, Float32x4, SignalFloat32x4, Scale<double>> layer, int layerIndex) → void

inherited

checkBestTrainingError(double trainingError) → void

inherited

computeEntryWeightUpdate(Float32x4 weight, Float32x4 weightLastUpdate, Float32x4 gradient, Float32x4 previousGradient, SignalFloat32x4 previousUpdateDeltas, SignalFloat32x4 noImprovementCounter, int weightsEntryIndex, Float32x4 neuronOutput) → Float32x4

inherited

computeEntryWeightUpdateSIMD(Float32x4 weight, Float32x4 weightLastUpdate, Float32x4 gradient, Float32x4 previousGradient, SignalFloat32x4 previousUpdateDeltas, SignalFloat32x4 noImprovementCounter, int weightsEntryIndex, Float32x4 neuronOutput) → Float32x4

Implementation of the weight update for an entry (SIMD).

inherited

computeGlobalError(List<P> samples) → double

inherited

computeWeightUpdate(double weight, double weightLastUpdate, num gradient, num previousGradient, List<num> previousUpdateDeltas, List<num> noImprovementCounter, int weightIndex, double neuronOutput) → double

Implementation of the weight update.

inherited

createLearningRateStrategy() → ParameterStrategy<double, Float32x4, SignalFloat32x4>

inherited

createMomentumStrategy() → ParameterStrategy<double, Float32x4, SignalFloat32x4>

inherited

createWeightStateBuffers({double fill = 0}) → List<List<SignalFloat32x4>>

Allocates a per-weight state buffer group with the same shape as the network weights (per layer -> per source neuron -> Signal over the target neurons), initialized to fill. The group is registered so reset reinitializes it. Optimizer subclasses use this for their persistent state (e.g. Adam's first/second moments).

inherited

generateRandomValue(double range) → double

inherited

generateRandomValuePositive(double range) → double

inherited

generateRandomWeightUpdate(double range, double min, double max, double multiplier) → double

inherited

generateRandomWeightUpdateByFactor(double weight, double factor, {double zeroPoint = 0.01, double multiplier = 1.0}) → double

inherited

initializeParameters() → void

Initialize training parameters.

inherited

initializeTraining() → void

inherited

learn(List<P> samples, double targetGlobalError) → bool

Learn the training of sample. Called by train.

inherited

loadGradients(List state) → void

Restores gradients previously produced by saveGradients.

inherited

loadOptimizerState(List state) → void

Restores optimizer state previously produced by saveOptimizerState.

inherited

logError(String message, [dynamic error, StackTrace? stackTrace]) → void

inherited

logInfo(String message) → void

inherited

logProgress(String message) → void

inherited

logWarn(String message) → void

inherited

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

recordTrainingBlock(double globalError, int epochs) → void

Records the outcome of an externally-driven training block (e.g. the async WebGPU trainers, which run epochs off the synchronous _trainImpl loop) so the public bookkeeping getters (globalError, trainedEpochs, totalTrainedEpochs, trainingActivations) reflect it. Also updates the best-training snapshot via checkBestTrainingError.

inherited

reset() → void

Reset this instance for a future training sessions.

inherited

resetBestTraining() → void

Discards the best weights/error tracked by checkBestTrainingError.

inherited

restoreGlobalLearnErrors(double last, double current) → void

Restores the epoch error-tracking used by iRProp+ backtracking (and by the learning-rate/momentum strategies). Used when resuming from a checkpoint.

inherited

saveGradients() → List<List<List<double>>>

Serializes the current accumulated gradients (layer.gradients) — the value that becomes previousGradient on the next epoch's resetGradients. Needed for an exact checkpoint resume of optimizers that read the previous gradient (Quickprop, iRProp+).

inherited

saveOptimizerState() → List<List<List<double>>>

Serializes the registered optimizer state buffers (for checkpointing). Each buffer group is flattened to a list of per-source-neuron value lists (layer-major, matching the allocation order).

inherited

selectInitialANN(List<P> samples, double targetGlobalError, [Random? random]) → void

Selects the initial ANN.

inherited

setLearningRate(double learningRate) → void

inherited

setMomentum(double momentum) → void

inherited

toString() → String

A string representation of this object.

inherited

train(int epochs, double targetGlobalError) → double

Train the samples for n epochs and returns the last global error.

inherited

trainUntilGlobalError({double? targetGlobalError, int epochsBlock = 50, int maxEpochs = 1000000, double maxEpochsLimitRatio = 3, int maxRetries = 5, double retryIncreaseMaxEpochsRatio = 1.50, Random? random}) → bool

Train the ann until targetGlobalError, with maxEpochs per training session and a maxRetries when a training session can't reach the target global error.

inherited

updateGlobalLearnError(double globalLearnError) → void

Publishes the epoch's learn error, rolling the previous value.

inherited

updateLayerWeights(Layer<double, Float32x4, SignalFloat32x4, Scale<double>> layer, int layerIndex) → void

inherited

updateParameters() → void

Update training parameters.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited