Class InverseProblemResult<T>

Namespace

AiDotNet.PhysicsInformed.Interfaces

Assembly

AiDotNet.dll

Results from inverse problem optimization.

public class InverseProblemResult<T>

Type Parameters

T

The numeric type.

Inheritance

object

InverseProblemResult<T>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Properties

Converged

Whether the optimization converged.

public bool Converged { get; set; }

Property Value

bool

DataLoss

Final data loss (fit to observations).

public T DataLoss { get; set; }

Property Value

T

IterationsToConverge

Number of iterations until convergence.

public int IterationsToConverge { get; set; }

Property Value

int

ParameterCorrelations

Correlation matrix between parameters (for uncertainty analysis).

public T[,]? ParameterCorrelations { get; set; }

Property Value

T[,]

Remarks

High correlation indicates parameters are not independently identifiable.

ParameterHistory

History of parameter values during training.

public List<T[]>? ParameterHistory { get; set; }

Property Value

List<T[]>

Remarks

Useful for visualizing convergence and detecting oscillations.

ParameterNames

Names of the identified parameters.

public string[] ParameterNames { get; set; }

Property Value

string[]

ParameterUncertainties

Estimated uncertainties (standard deviations) for each parameter.

public T[]? ParameterUncertainties { get; set; }

Property Value

T[]

Remarks

Null if uncertainty estimation was not performed.

Parameters

The identified parameter values.

public T[] Parameters { get; set; }

Property Value

T[]

PhysicsLoss

Final physics loss (PDE residual).

public T PhysicsLoss { get; set; }

Property Value

T

TotalLoss

Total loss (data + physics + regularization).

public T TotalLoss { get; set; }

Property Value

T