Interface IDomainDecompositionTrainingHistory<T>

Namespace

AiDotNet.PhysicsInformed.Interfaces

Assembly

AiDotNet.dll

Extended training history interface for domain decomposition PINN training.

public interface IDomainDecompositionTrainingHistory<T>

Type Parameters

T

The numeric type.

Remarks

For Beginners: Domain decomposition divides a large problem domain into smaller subdomains. Each subdomain has its own neural network, and interface conditions ensure continuity between neighboring subdomains.

This interface tracks:

• Per-subdomain losses
• Interface continuity losses
• Overall convergence metrics

Properties

InterfaceLosses

Gets the interface continuity losses per epoch.

List<T> InterfaceLosses { get; }

Property Value

List<T>

Losses

Gets the total losses per epoch (combined from all subdomains).

List<T> Losses { get; }

Property Value

List<T>

PhysicsLosses

Gets the PDE residual losses per epoch (sum across all subdomains).

List<T> PhysicsLosses { get; }

Property Value

List<T>

SubdomainCount

Gets the number of subdomains.

int SubdomainCount { get; }

Property Value

int

SubdomainLosses

Gets the losses per subdomain per epoch. Outer list: epochs, Inner list: subdomain losses.

List<List<T>> SubdomainLosses { get; }

Property Value

List<List<T>>

Methods

AddEpoch(T, List<T>, T, T)

Records metrics for a training epoch.

void AddEpoch(T totalLoss, List<T> subdomainLosses, T interfaceLoss, T physicsLoss)

Parameters

totalLoss T

Combined loss from all subdomains and interfaces.

subdomainLosses List<T>

Individual losses per subdomain.

interfaceLoss T

Interface continuity loss.

physicsLoss T

Total PDE residual loss.