ruddy 0.1.0__py3-none-any.whl

ruddy/__init__.py

@@ -0,0 +1,279 @@
+"""Ruddy: domain-agnostic statistical exploratory data analysis."""
+
+from ruddy._version import __version__
+from ruddy.analysis import (
+    AnalysisConfig,
+    ConfidenceIntervalResult,
+    UnifiedAnalysisResult,
+    analyze,
+    analyze_confidence_intervals,
+    analyze_groups,
+)
+from ruddy.anomaly import AnomalyResult, analyze_anomalies
+from ruddy.bayesian import BayesianEDAResult, analyze_bayesian_eda
+from ruddy.bivariate import (
+    BivariateResult,
+    ContingencyDiagnosticsResult,
+    DependenceResult,
+    GroupAnalysisResult,
+    PosthocResult,
+    analyze_bivariate,
+    analyze_contingency_diagnostics,
+    analyze_dependence,
+    analyze_grouped_responses,
+    analyze_posthoc,
+    distance_correlation,
+    resolve_groups,
+    resolve_responses,
+    summarize_annotation_coverage,
+    summarize_categorical_associations,
+    summarize_correlations,
+    summarize_general_dependence,
+    summarize_group_coverage,
+    summarize_grouped_categorical_responses,
+    summarize_grouped_numeric_responses,
+    summarize_numeric_categorical_comparisons,
+    summarize_partial_correlations,
+    summarize_response_catalog,
+)
+from ruddy.compositional import (
+    CompositionalResult,
+    aitchison_distance_matrix,
+    alr_transform,
+    analyze_composition,
+    closure,
+    clr_transform,
+    ilr_transform,
+    multiplicative_zero_replacement,
+    variation_matrix,
+)
+from ruddy.core import (
+    AlignmentMode,
+    AnalysisBlock,
+    AnnotationCoverage,
+    ColumnKind,
+    ColumnRole,
+    ComparisonTest,
+    CorrelationMethod,
+    OptionalDependencyError,
+    OutlierMethod,
+    PAdjustMethod,
+    ProjectionMethod,
+    ResultStatus,
+    ScalingMethod,
+)
+from ruddy.data import (
+    AlignedAnnotations,
+    FeatureMatrix,
+    TabularDataset,
+    align_annotation_source,
+    attach_annotations,
+)
+from ruddy.factorial import (
+    FactorialDesign,
+    FactorialResult,
+    FactorialTerm,
+    MarginalMeansResult,
+    MixedEffectsResult,
+    analyze_factorial,
+    analyze_marginal_means,
+    analyze_mixed_effects,
+    build_factorial_design,
+)
+from ruddy.multivariate import (
+    CollinearityResult,
+    CovarianceResult,
+    MahalanobisResult,
+    MANOVAResult,
+    MultivariateResult,
+    PermutationGroupResult,
+    analyze_collinearity,
+    analyze_covariance_structure,
+    analyze_mahalanobis,
+    analyze_manova,
+    analyze_multivariate,
+    analyze_permutation_group_structure,
+)
+from ruddy.profiling import (
+    ProfilingResult,
+    pairwise_completeness,
+    profile_columns,
+    profile_dataset,
+    summarize_missingness,
+    summarize_missingness_patterns,
+    summarize_overview,
+)
+from ruddy.projections import (
+    PCAResult,
+    PreparedFeatures,
+    ProjectionResult,
+    analyze_pca,
+    analyze_projection,
+    analyze_tsne,
+    analyze_umap,
+    prepare_features,
+)
+from ruddy.representation import (
+    AlignedRepresentationPair,
+    CCAResult,
+    RepresentationComparisonResult,
+    align_feature_matrices,
+    analyze_representation_similarity,
+    linear_cka,
+)
+from ruddy.results import Advisory, AnalysisProvenance, AnalysisResult
+from ruddy.statistics import (
+    BootstrapResult,
+    bootstrap_confidence_interval,
+    mean_confidence_interval,
+    odds_ratio_confidence_interval,
+    pearson_confidence_interval,
+    welch_mean_difference_confidence_interval,
+)
+from ruddy.univariate import (
+    DistributionDiagnosticsResult,
+    OutlierResult,
+    UnivariateResult,
+    UnivariateTables,
+    analyze_distribution_diagnostics,
+    analyze_outliers,
+    analyze_univariate,
+    summarize_categorical_statistics,
+    summarize_datetime_statistics,
+    summarize_dispersion_diagnostics,
+    summarize_normality_diagnostics,
+    summarize_numeric_quality,
+    summarize_numeric_statistics,
+    summarize_outliers,
+    summarize_univariate,
+)
+
+__all__ = [
+    "Advisory",
+    "AlignedAnnotations",
+    "AlignedRepresentationPair",
+    "AlignmentMode",
+    "AnalysisBlock",
+    "AnalysisConfig",
+    "AnalysisProvenance",
+    "AnalysisResult",
+    "AnnotationCoverage",
+    "AnomalyResult",
+    "BayesianEDAResult",
+    "BivariateResult",
+    "BootstrapResult",
+    "CCAResult",
+    "CollinearityResult",
+    "ColumnKind",
+    "ColumnRole",
+    "ComparisonTest",
+    "CompositionalResult",
+    "ConfidenceIntervalResult",
+    "ContingencyDiagnosticsResult",
+    "CorrelationMethod",
+    "CovarianceResult",
+    "DependenceResult",
+    "DistributionDiagnosticsResult",
+    "FactorialDesign",
+    "FactorialResult",
+    "FactorialTerm",
+    "FeatureMatrix",
+    "GroupAnalysisResult",
+    "MANOVAResult",
+    "MahalanobisResult",
+    "MarginalMeansResult",
+    "MixedEffectsResult",
+    "MultivariateResult",
+    "OptionalDependencyError",
+    "OutlierMethod",
+    "OutlierResult",
+    "PAdjustMethod",
+    "PCAResult",
+    "PermutationGroupResult",
+    "PosthocResult",
+    "PreparedFeatures",
+    "ProfilingResult",
+    "ProjectionMethod",
+    "ProjectionResult",
+    "RepresentationComparisonResult",
+    "ResultStatus",
+    "ScalingMethod",
+    "TabularDataset",
+    "UnifiedAnalysisResult",
+    "UnivariateResult",
+    "UnivariateTables",
+    "__version__",
+    "aitchison_distance_matrix",
+    "align_annotation_source",
+    "align_feature_matrices",
+    "alr_transform",
+    "analyze",
+    "analyze_anomalies",
+    "analyze_bayesian_eda",
+    "analyze_bivariate",
+    "analyze_collinearity",
+    "analyze_composition",
+    "analyze_confidence_intervals",
+    "analyze_contingency_diagnostics",
+    "analyze_covariance_structure",
+    "analyze_dependence",
+    "analyze_distribution_diagnostics",
+    "analyze_factorial",
+    "analyze_grouped_responses",
+    "analyze_groups",
+    "analyze_mahalanobis",
+    "analyze_manova",
+    "analyze_marginal_means",
+    "analyze_mixed_effects",
+    "analyze_multivariate",
+    "analyze_outliers",
+    "analyze_pca",
+    "analyze_permutation_group_structure",
+    "analyze_posthoc",
+    "analyze_projection",
+    "analyze_representation_similarity",
+    "analyze_tsne",
+    "analyze_umap",
+    "analyze_univariate",
+    "attach_annotations",
+    "bootstrap_confidence_interval",
+    "build_factorial_design",
+    "closure",
+    "clr_transform",
+    "distance_correlation",
+    "ilr_transform",
+    "linear_cka",
+    "mean_confidence_interval",
+    "multiplicative_zero_replacement",
+    "odds_ratio_confidence_interval",
+    "pairwise_completeness",
+    "pearson_confidence_interval",
+    "prepare_features",
+    "profile_columns",
+    "profile_dataset",
+    "resolve_groups",
+    "resolve_responses",
+    "summarize_annotation_coverage",
+    "summarize_categorical_associations",
+    "summarize_categorical_statistics",
+    "summarize_correlations",
+    "summarize_datetime_statistics",
+    "summarize_dispersion_diagnostics",
+    "summarize_general_dependence",
+    "summarize_group_coverage",
+    "summarize_grouped_categorical_responses",
+    "summarize_grouped_numeric_responses",
+    "summarize_missingness",
+    "summarize_missingness_patterns",
+    "summarize_normality_diagnostics",
+    "summarize_numeric_categorical_comparisons",
+    "summarize_numeric_quality",
+    "summarize_numeric_statistics",
+    "summarize_outliers",
+    "summarize_overview",
+    "summarize_partial_correlations",
+    "summarize_response_catalog",
+    "summarize_univariate",
+    "variation_matrix",
+    "welch_mean_difference_confidence_interval",
+]

ruddy/_version.py

@@ -0,0 +1,3 @@
+"""Package version."""
+
+__version__ = "0.1.0"

ruddy/analysis/__init__.py

@@ -0,0 +1,15 @@
+"""High-level analysis configuration and orchestration."""
+
+from ruddy.analysis.api import analyze, analyze_groups
+from ruddy.analysis.config import AnalysisConfig
+from ruddy.analysis.intervals import ConfidenceIntervalResult, analyze_confidence_intervals
+from ruddy.analysis.results import UnifiedAnalysisResult
+
+__all__ = [
+    "AnalysisConfig",
+    "ConfidenceIntervalResult",
+    "UnifiedAnalysisResult",
+    "analyze",
+    "analyze_confidence_intervals",
+    "analyze_groups",
+]

ruddy/analysis/api.py

@@ -0,0 +1,51 @@
+"""Public high-level analysis API."""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from ruddy.analysis.engine import analyze
+from ruddy.bivariate.groups import GroupAnalysisResult, analyze_grouped_responses
+from ruddy.core.enums import ComparisonTest, PAdjustMethod
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+    from ruddy.data import AlignedAnnotations, TabularDataset
+
+__all__ = ["analyze", "analyze_groups"]
+
+
+def analyze_groups(
+    dataset: TabularDataset,
+    *,
+    responses: Iterable[str] | None = None,
+    groups: Iterable[str] | None = None,
+    annotations: Iterable[AlignedAnnotations] = (),
+    comparison_tests: tuple[ComparisonTest | str, ...] = (
+        ComparisonTest.WELCH_T,
+        ComparisonTest.MANN_WHITNEY,
+        ComparisonTest.WELCH_ANOVA,
+        ComparisonTest.KRUSKAL_WALLIS,
+        ComparisonTest.CHI_SQUARE,
+        ComparisonTest.FISHER_EXACT,
+    ),
+    p_adjust: PAdjustMethod | str = PAdjustMethod.FDR_BH,
+    min_group_n: int = 3,
+    max_group_levels: int = 20,
+    max_category_levels: int = 50,
+    pairwise: bool = False,
+) -> GroupAnalysisResult:
+    """Run response-centric grouped EDA using statistical roles and data kinds."""
+    return analyze_grouped_responses(
+        dataset,
+        responses=responses,
+        groups=groups,
+        annotations=annotations,
+        comparison_tests=comparison_tests,
+        p_adjust=p_adjust,
+        min_group_n=min_group_n,
+        max_group_levels=max_group_levels,
+        max_category_levels=max_category_levels,
+        pairwise=pairwise,
+    )