IDomainDecompositionTrainingHistory<T>

Interfaces · AiDotNet.PhysicsInformed.Interfaces

Extended training history interface for domain decomposition PINN training.

Known implementations: DomainDecompositionTrainingHistory<T>

How It Works

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

Property Summary
List<List<T>> SubdomainLosses { get; } Gets the losses per subdomain per epoch.
List<T> InterfaceLosses { get; } Gets the interface continuity losses per epoch.
List<T> Losses { get; } Gets the total losses per epoch (combined from all subdomains).
List<T> PhysicsLosses { get; } Gets the PDE residual losses per epoch (sum across all subdomains).
int SubdomainCount { get; } Gets the number of subdomains.

Methods

Method Summary
void AddEpoch(T totalLoss, List<T> subdomainLosses, T interfaceLoss, T physicsLoss) Records metrics for a training epoch.