Class BuiltInSupervisedAutoMLModelBase<T, TInput, TOutput>

Namespace

AiDotNet.AutoML

Assembly

AiDotNet.dll

Base class for built-in supervised AutoML strategies that operate on tabular Matrix/Vector tasks.

public abstract class BuiltInSupervisedAutoMLModelBase<T, TInput, TOutput> : SupervisedAutoMLModelBase<T, TInput, TOutput>, IAutoMLModel<T, TInput, TOutput>, IFullModel<T, TInput, TOutput>, IModel<TInput, TOutput, ModelMetadata<T>>, IModelSerializer, ICheckpointableModel, IParameterizable<T, TInput, TOutput>, IFeatureAware, IFeatureImportance<T>, ICloneable<IFullModel<T, TInput, TOutput>>, IGradientComputable<T, TInput, TOutput>, IJitCompilable<T>

Type Parameters

T

The numeric type used for calculations.

TInput

The input data type.

TOutput

The output data type.

Inheritance

object

AutoMLModelBase<T, TInput, TOutput>

SupervisedAutoMLModelBase<T, TInput, TOutput>

BuiltInSupervisedAutoMLModelBase<T, TInput, TOutput>

Implements

IAutoMLModel<T, TInput, TOutput>

IFullModel<T, TInput, TOutput>

IModel<TInput, TOutput, ModelMetadata<T>>

IModelSerializer

ICheckpointableModel

IParameterizable<T, TInput, TOutput>

IFeatureAware

IFeatureImportance<T>

ICloneable<IFullModel<T, TInput, TOutput>>

IGradientComputable<T, TInput, TOutput>

IJitCompilable<T>

Derived

BayesianOptimizationAutoML<T, TInput, TOutput>

EvolutionaryAutoML<T, TInput, TOutput>

MultiFidelityAutoML<T, TInput, TOutput>

RandomSearchAutoML<T, TInput, TOutput>

Inherited Members

SupervisedAutoMLModelBase<T, TInput, TOutput>.Random

SupervisedAutoMLModelBase<T, TInput, TOutput>.EnsembleOptions

SupervisedAutoMLModelBase<T, TInput, TOutput>.CrossValidationOptions

SupervisedAutoMLModelBase<T, TInput, TOutput>.BudgetPreset

SupervisedAutoMLModelBase<T, TInput, TOutput>.ExecuteTrialAsync(ModelType, Dictionary<string, object>, TInput, TOutput, TInput, TOutput, CancellationToken)

SupervisedAutoMLModelBase<T, TInput, TOutput>.TrySelectEnsembleAsBestAsync(TInput, TOutput, TInput, TOutput, DateTime, CancellationToken)

SupervisedAutoMLModelBase<T, TInput, TOutput>.PickCandidateModelType()

SupervisedAutoMLModelBase<T, TInput, TOutput>.EnsureDefaultOptimizationMetric(TOutput)

AutoMLModelBase<T, TInput, TOutput>._trialHistory

AutoMLModelBase<T, TInput, TOutput>._searchSpace

AutoMLModelBase<T, TInput, TOutput>._candidateModels

AutoMLModelBase<T, TInput, TOutput>._constraints

AutoMLModelBase<T, TInput, TOutput>._lock

AutoMLModelBase<T, TInput, TOutput>._optimizationMetric

AutoMLModelBase<T, TInput, TOutput>._maximize

AutoMLModelBase<T, TInput, TOutput>._optimizationMetricExplicitlySet

AutoMLModelBase<T, TInput, TOutput>._earlyStoppingPatience

AutoMLModelBase<T, TInput, TOutput>._earlyStoppingMinDelta

AutoMLModelBase<T, TInput, TOutput>._trialsSinceImprovement

AutoMLModelBase<T, TInput, TOutput>._modelEvaluator

AutoMLModelBase<T, TInput, TOutput>.OptimizationMetric

AutoMLModelBase<T, TInput, TOutput>.MaximizeOptimizationMetric

AutoMLModelBase<T, TInput, TOutput>.Type

AutoMLModelBase<T, TInput, TOutput>.Status

AutoMLModelBase<T, TInput, TOutput>.BestModel

AutoMLModelBase<T, TInput, TOutput>.BestScore

AutoMLModelBase<T, TInput, TOutput>.TimeLimit

AutoMLModelBase<T, TInput, TOutput>.TrialLimit

AutoMLModelBase<T, TInput, TOutput>.SearchAsync(TInput, TOutput, TInput, TOutput, TimeSpan, CancellationToken)

AutoMLModelBase<T, TInput, TOutput>.SetSearchSpace(Dictionary<string, ParameterRange>)

AutoMLModelBase<T, TInput, TOutput>.SetCandidateModels(List<ModelType>)

AutoMLModelBase<T, TInput, TOutput>.SetOptimizationMetric(MetricType, bool)

AutoMLModelBase<T, TInput, TOutput>.GetTrialHistory()

AutoMLModelBase<T, TInput, TOutput>.GetFeatureImportanceAsync()

AutoMLModelBase<T, TInput, TOutput>.SuggestNextTrialAsync()

AutoMLModelBase<T, TInput, TOutput>.ReportTrialResultAsync(Dictionary<string, object>, double, TimeSpan)

AutoMLModelBase<T, TInput, TOutput>.EnableEarlyStopping(int, double)

AutoMLModelBase<T, TInput, TOutput>.SetConstraints(List<SearchConstraint>)

AutoMLModelBase<T, TInput, TOutput>.Train(double[][], double[])

AutoMLModelBase<T, TInput, TOutput>.Predict(double[][])

AutoMLModelBase<T, TInput, TOutput>.GetModelMetadata()

AutoMLModelBase<T, TInput, TOutput>.ShouldStop()

AutoMLModelBase<T, TInput, TOutput>.ValidateConstraints(Dictionary<string, object>, IFullModel<T, TInput, TOutput>)

AutoMLModelBase<T, TInput, TOutput>.CreateModelAsync(ModelType, Dictionary<string, object>)

AutoMLModelBase<T, TInput, TOutput>.EvaluateModelAsync(IFullModel<T, TInput, TOutput>, TInput, TOutput)

AutoMLModelBase<T, TInput, TOutput>.GetDefaultSearchSpace(ModelType)

AutoMLModelBase<T, TInput, TOutput>.Train(TInput, TOutput)

AutoMLModelBase<T, TInput, TOutput>.Predict(TInput)

AutoMLModelBase<T, TInput, TOutput>.SaveModel(string)

AutoMLModelBase<T, TInput, TOutput>.LoadModel(string)

AutoMLModelBase<T, TInput, TOutput>.Serialize()

AutoMLModelBase<T, TInput, TOutput>.Deserialize(byte[])

AutoMLModelBase<T, TInput, TOutput>.GetParameters()

AutoMLModelBase<T, TInput, TOutput>.SetParameters(Vector<T>)

AutoMLModelBase<T, TInput, TOutput>.ParameterCount

AutoMLModelBase<T, TInput, TOutput>.WithParameters(Vector<T>)

AutoMLModelBase<T, TInput, TOutput>.FeatureNames

AutoMLModelBase<T, TInput, TOutput>.GetFeatureImportance()

AutoMLModelBase<T, TInput, TOutput>.GetActiveFeatureIndices()

AutoMLModelBase<T, TInput, TOutput>.IsFeatureUsed(int)

AutoMLModelBase<T, TInput, TOutput>.SetActiveFeatureIndices(IEnumerable<int>)

AutoMLModelBase<T, TInput, TOutput>.Clone()

AutoMLModelBase<T, TInput, TOutput>.DeepCopy()

AutoMLModelBase<T, TInput, TOutput>.CreateInstanceForCopy()

AutoMLModelBase<T, TInput, TOutput>.SetModelEvaluator(IModelEvaluator<T, TInput, TOutput>)

AutoMLModelBase<T, TInput, TOutput>.ExtractMetricFromEvaluation(ModelEvaluationData<T, TInput, TOutput>)

AutoMLModelBase<T, TInput, TOutput>.ReportTrialFailureAsync(Dictionary<string, object>, Exception, TimeSpan)

AutoMLModelBase<T, TInput, TOutput>.ConfigureSearchSpace(Dictionary<string, ParameterRange>)

AutoMLModelBase<T, TInput, TOutput>.SetTimeLimit(TimeSpan)

AutoMLModelBase<T, TInput, TOutput>.SetTrialLimit(int)

AutoMLModelBase<T, TInput, TOutput>.EnableNAS(bool)

AutoMLModelBase<T, TInput, TOutput>.SearchBestModel(TInput, TOutput, TInput, TOutput)

AutoMLModelBase<T, TInput, TOutput>.Search(TInput, TOutput, TInput, TOutput)

AutoMLModelBase<T, TInput, TOutput>.GetResults()

AutoMLModelBase<T, TInput, TOutput>.Run(TInput, TOutput, TInput, TOutput)

AutoMLModelBase<T, TInput, TOutput>.SetModelsToTry(List<ModelType>)

AutoMLModelBase<T, TInput, TOutput>.DefaultLossFunction

AutoMLModelBase<T, TInput, TOutput>.ComputeGradients(TInput, TOutput, ILossFunction<T>)

AutoMLModelBase<T, TInput, TOutput>.ApplyGradients(Vector<T>, T)

AutoMLModelBase<T, TInput, TOutput>.SupportsJitCompilation

AutoMLModelBase<T, TInput, TOutput>.ExportComputationGraph(List<ComputationNode<T>>)

AutoMLModelBase<T, TInput, TOutput>.SaveState(Stream)

AutoMLModelBase<T, TInput, TOutput>.LoadState(Stream)

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Extension Methods

DistributedExtensions.AsDistributedForHighBandwidth<T, TInput, TOutput>(IFullModel<T, TInput, TOutput>, ICommunicationBackend<T>)

DistributedExtensions.AsDistributedForLowBandwidth<T, TInput, TOutput>(IFullModel<T, TInput, TOutput>, ICommunicationBackend<T>)

DistributedExtensions.AsDistributed<T, TInput, TOutput>(IFullModel<T, TInput, TOutput>, ICommunicationBackend<T>)

DistributedExtensions.AsDistributed<T, TInput, TOutput>(IFullModel<T, TInput, TOutput>, IShardingConfiguration<T>)

Remarks

This class centralizes built-in model construction and default candidate selection so different search strategies (random, Bayesian, evolutionary, etc.) can focus on "how to propose the next trial".

For Beginners: AutoML tries many "trials". Each trial picks a model family and some settings, then trains and scores it. Different strategies decide which settings to try next.

Constructors

BuiltInSupervisedAutoMLModelBase(IModelEvaluator<T, TInput, TOutput>?, Random?)

protected BuiltInSupervisedAutoMLModelBase(IModelEvaluator<T, TInput, TOutput>? modelEvaluator = null, Random? random = null)

Parameters

modelEvaluator IModelEvaluator<T, TInput, TOutput>

random Random

Methods

CreateModelAsync(ModelType, Dictionary<string, object>)

Creates a model instance for the given type and parameters

protected override Task<IFullModel<T, TInput, TOutput>> CreateModelAsync(ModelType modelType, Dictionary<string, object> parameters)

Parameters

modelType ModelType

parameters Dictionary<string, object>

Returns

Task<IFullModel<T, TInput, TOutput>>

EnsureDefaultCandidateModels(TInput, TOutput)

Applies built-in default candidate models when the user has not configured candidates explicitly.

protected void EnsureDefaultCandidateModels(TInput inputs, TOutput targets)

Parameters

inputs TInput

targets TOutput

GetDefaultSearchSpace(ModelType)

Gets the default search space for a model type

protected override Dictionary<string, ParameterRange> GetDefaultSearchSpace(ModelType modelType)

Parameters

modelType ModelType

Returns

Dictionary<string, ParameterRange>