Expand description
Analysis utilities for the unified benchmarking system
This module provides helper constructors, default-value builders, and
pure analysis functions used by system.rs. Keeping them here prevents
system.rs from growing past the 2 000-line limit.
Structs§
- Fidelity
Stats - Simple fidelity statistics aggregated from a slice of raw fidelity values.
Functions§
- compute_
cost_ analysis - Aggregate platform results into a
CostAnalysisResult. - compute_
cost_ metrics - Compute
CostMetricsfrom the three benchmark result sets. - compute_
cross_ platform_ analysis - Rank platforms and produce a
CrossPlatformAnalysis. - compute_
fidelity_ statistics - Compute fidelity statistics from a slice of values in
[0, 1]. - compute_
performance_ metrics - Compute
PlatformPerformanceMetricsfrom the four benchmark result sets. - compute_
reliability_ metrics - Compute
ReliabilityMetricsfrom the three benchmark result sets. - compute_
resource_ analysis - Aggregate platform results into a
ResourceAnalysisResult. - default_
algorithm_ level_ results - Build a minimal, valid
AlgorithmLevelResults. - default_
circuit_ level_ results - Build a minimal, valid
CircuitLevelResults. - default_
gate_ level_ results - Build a minimal, valid
GateLevelResultsrepresenting a device that has been characterised with generic default values. - default_
scirs2_ analysis - Produce a fully-structured but analytically trivial
SciRS2AnalysisResult. This is the fallback when the scirs2 feature is not available or the per-platform data is too sparse for meaningful analysis. - default_
system_ level_ results - Build a minimal, valid
SystemLevelResults. - statistical_
summary_ from_ slice - Build a
StatisticalSummaryfrom a non-empty slice off64values. If the slice is empty, returns a zero summary. - zero_
statistical_ summary - Build a zero-valued
StatisticalSummary.