HybridODESDEScheduler<T>
Models & Types ยท AiDotNet.Diffusion.Schedulers
Hybrid ODE/SDE scheduler that transitions between deterministic and stochastic sampling.
Inheritance: NoiseSchedulerBase<T> โ HybridODESDEScheduler<T>
For Beginners
This scheduler starts deterministically (same seed = same image) for the main structure, then adds some controlled randomness for fine details. This gives you both reliable composition and natural-looking textures.
How It Works
Begins with deterministic ODE sampling for coarse structure, then switches to stochastic SDE sampling for fine detail generation. The transition point is configurable, balancing consistency with diversity.
Constructors
| Constructor | Summary |
|---|---|
HybridODESDEScheduler(SchedulerConfig<T> config, double transitionPoint) |
Methods
| Method | Summary |
|---|---|
Vector<T> Step(Vector<T> modelOutput, int timestep, Vector<T> sample, T eta, Vector<T> noise) |