Module ml 

ML-readiness evaluation module.

Validates that generated data is suitable for machine learning tasks including feature distributions, label quality, and graph structure.

Also provides baseline task definitions for benchmarking synthetic data.

Structs

AnomalyScoringAnalysis

Results of anomaly scoring analysis.

AnomalyScoringAnalyzer

Analyzer for anomaly scoring quality.

AnomalyScoringThresholds

Thresholds for anomaly scoring analysis.

BaselineConfig

Configuration for baseline evaluation.

BaselineEvaluation

Collection of baseline results for all tasks.

BaselineResult

Baseline task results.

BaselineSummary

Summary of baseline evaluation.

BaselineTask

ML baseline task definition.

ClassificationMetrics

Classification metrics for binary/multiclass tasks.

CrossModalAnalysis

Results of cross-modal consistency analysis.

CrossModalAnalyzer

Analyzer for cross-modal consistency.

CrossModalThresholds

Thresholds for cross-modal consistency analysis.

DistributionSample

A pair of value distributions to compare.

DomainGapAnalysis

Results of domain gap analysis.

DomainGapAnalyzer

Analyzer for domain gap between synthetic and reference distributions.

DomainGapDetail

Detail for a single distribution comparison.

DomainGapThresholds

Thresholds for domain gap analysis.

EmbeddingInput

Input for embedding readiness analysis.

EmbeddingReadinessAnalysis

Results of embedding readiness analysis.

EmbeddingReadinessAnalyzer

Analyzer for embedding readiness.

EmbeddingReadinessThresholds

Thresholds for embedding readiness analysis.

EntityModalData

Modal data for a single entity with tabular and graph feature vectors.

ExpectedMetrics

Expected performance metrics for a task.

FeatureAnalysis

Results of feature analysis.

FeatureAnalyzer

Analyzer for feature distributions.

FeatureQualityAnalysis

Results of feature quality analysis.

FeatureQualityAnalyzer

Analyzer for feature quality metrics.

FeatureQualityThresholds

Thresholds for feature quality analysis.

FeatureStats

Statistics for a single feature.

FeatureVector

A single feature vector with optional label values for importance estimation.

GnnGraphData

Input graph data for GNN readiness analysis.

GnnReadinessAnalysis

Results of GNN readiness analysis.

GnnReadinessAnalyzer

Analyzer for GNN readiness.

GnnReadinessThresholds

Thresholds for GNN readiness analysis.

GraphAnalysis

Results of graph analysis.

GraphAnalyzer

Analyzer for graph structure.

GraphMetrics

Basic graph metrics.

LabelAnalysis

Results of label analysis.

LabelAnalyzer

Analyzer for label quality.

LabelDistribution

Distribution for a single label class.

MLReadinessEvaluation

Combined ML-readiness evaluation results.

RankingMetrics

Ranking metrics for link prediction and recommendation.

RegressionMetrics

Regression metrics for continuous prediction tasks.

SchemeDetectabilityAnalysis

Results of scheme detectability analysis.

SchemeDetectabilityAnalyzer

Analyzer for scheme detectability.

SchemeDetectabilityThresholds

Thresholds for scheme detectability analysis.

SchemeRecord

A single scheme record with difficulty and detection score.

ScoredRecord

A single record with an anomaly score and ground truth label.

SplitAnalysis

Results of split analysis.

SplitAnalyzer

Analyzer for train/test splits.

SplitMetrics

Metrics for a single split.

TemporalFidelityAnalysis

Results of temporal fidelity analysis.

TemporalFidelityAnalyzer

Analyzer for temporal fidelity.

TemporalFidelityThresholds

Thresholds for temporal fidelity analysis.

TemporalRecord

A single temporal record with a timestamp and associated value.

Enums

BaselineAlgorithm

Baseline algorithm for a task.

MLTaskType

ML task type for benchmarking.

PerformanceGrade

Performance grade for baseline results.

Functions

get_accounting_baseline_tasks

Get predefined baseline tasks for synthetic accounting data.