Expand description
Execution engine for Soma computational graphs.
Provides the runtime components that execute compiled plans:
GraphSession— the primary orchestrator: Graph + filters → compile → executeexecutor— walks [ExecutionPlan] trees (sequence, parallel, cached, remote)cache— LRU memory cache, local disk cache, tiered cachesampler— hyperparameter samplers (Grid, Random, Bayesian/TPE)pruner— early stopping strategies (Median, Percentile)stream— chunk-based streaming executorStudyRunner— orchestrates optimization studies
Re-exports§
pub use cache::LocalCache;pub use cache::MemoryCache;pub use cache::TieredCache;pub use event_bus::EventBus;pub use executor::Context;pub use executor::GraphInfo;pub use executor::RemoteExecutor;pub use executor::execute;pub use executor::resolve_input;pub use filter_library::FilterLibrary;pub use graph_session::GraphSession;pub use graph_session::graph_fit;pub use graph_session::graph_predict;pub use graph_session::graph_run;pub use pbt_runner::FnPbtExecutor;pub use pbt_runner::PbtConfig;pub use pbt_runner::PbtExecutor;pub use pbt_runner::PbtRunner;pub use pbt_runner::PopulationMember;pub use pruner::MedianPruner;pub use pruner::PercentilePruner;pub use pruner::Pruner;pub use sampler::BayesianSampler;pub use sampler::GridSampler;pub use sampler::RandomSampler;pub use sampler::Sampler;pub use stream::FittedFilter;pub use stream::StreamExecutor;pub use study_runner::FnTrialExecutor;pub use study_runner::StudyRunner;pub use study_runner::TrialExecutor;pub use study_runner::TrialOutcome;
Modules§
- cache
- Cache implementations for the Soma runtime.
- event_
bus - Broadcast event bus for runtime observability.
- executor
- Plan executor — walks
ExecutionPlantrees and runs filter nodes. - filter_
library - Unified filter registry — holds implementations, metadata, and trained states.
- graph_
session - Graph session — the primary orchestrator for Graph → Compile → Execute.
- pbt_
runner - Population-Based Training runner.
- pruner
- Early stopping strategies for optimization studies.
- sampler
- Hyperparameter samplers for optimization studies.
- stream
- Streaming executor — processes data in chunks through fitted filters.
- study_
runner - Study runner — orchestrates hyperparameter optimization.