Struct MLQueryOptimizer 

pub struct MLQueryOptimizer { /* private fields */ }

ML-based query optimizer with adaptive learning

Uses SciRS2-core features for advanced query optimization:

• Statistical cardinality prediction with continuous learning
• Adaptive join ordering based on execution feedback
• Pattern-based optimization strategies
• Continuous learning from execution feedback

This is a foundation for future ML integration including:

• GPU-accelerated histogram processing
• Neural architecture search for join ordering
• Quantum-inspired graph optimization

Implementations

impl MLQueryOptimizer

pub fn new() -> Self

Create a new ML query optimizer

Initializes GPU context (if available), ML pipeline, neural architecture search, and quantum optimizer for advanced query optimization.

pub fn with_config(config: MLOptimizerConfig) -> Self

Create with custom configuration

pub fn predict_cardinality(&self, features: &PatternFeatures) -> Result<usize>

Predict query cardinality using trained model

Uses learned weights for cardinality prediction. Falls back to heuristic if ML model not yet trained.

pub fn optimize_join_order( &self, pattern_count: usize, features: &PatternFeatures, ) -> Result<Vec<usize>>

Optimize join order using adaptive strategy

Uses learned heuristics to determine optimal join ordering for the given query pattern. Returns recommended join order based on selectivity.

pub fn train_from_execution( &mut self, features: PatternFeatures, actual_cardinality: usize, execution_time_ms: f64, ) -> Result<()>

Train the optimizer from execution feedback

Adds execution data to training buffer and triggers retraining when sufficient samples are collected. This enables continuous learning and adaptation to workload characteristics.

pub fn optimize( &mut self, features: PatternFeatures, ) -> Result<MLOptimizationResult>

Get comprehensive optimization recommendation

Combines ML techniques (cardinality prediction, adaptive join ordering) to provide comprehensive optimization guidance for query execution.

pub fn training_stats(&self) -> Result<TrainingStats>

Get training statistics

pub fn performance_metrics(&self) -> PerformanceMetrics

Get performance metrics for the ML optimizer

Returns comprehensive performance statistics including:

• Number of predictions made
• Number of training operations

Trait Implementations

impl Default for MLQueryOptimizer

fn default() -> Self

Returns the “default value” for a type.