Module parallel_processing 

Expanded parallel processing for statistical computations

This module provides parallelized implementations of:

• Descriptive statistics (Welford’s algorithm, parallel quantiles, histograms)
• Hypothesis testing (permutation tests, bootstrap, cross-validation)
• Distribution fitting (parallel MLE, grid search)

All parallel code is feature-gated behind cfg(feature = "parallel") (which scirs2-core enables via scirs2-core/parallel = ["rayon"]).

Structs

CrossValidationResult

Result of parallel cross-validation.

GridSearchResult

Result of a parallel parameter grid search.

ParallelBootstrapResult

Result of a parallel bootstrap procedure.

ParallelHistogramResult

Result of a parallel histogram computation.

ParallelMLEResult

Result of parallel maximum-likelihood estimation.

PermutationTestResult

Result of a permutation test.

WelfordAccumulator

Mergeable accumulator for mean, variance, skewness, kurtosis via Welford’s method.

Functions

parallel_bootstrap

Run a parallel bootstrap procedure.

parallel_cross_validation

Run parallel k-fold cross-validation.

parallel_grid_search

Parallel grid search over distribution parameters to maximize log-likelihood.

parallel_histogram

Compute a histogram in parallel.

parallel_median

Compute the median in parallel.

parallel_mle_fit

Fit multiple distributions in parallel and return the best by AIC.

parallel_moments

Compute mean, variance, skewness, and kurtosis in parallel via Welford’s method.

parallel_permutation_test

Parallel permutation test for the difference in means between two groups.

parallel_quantile

Compute a single quantile using parallel sort.

parallel_welford_kurtosis

Compute excess kurtosis in parallel.

parallel_welford_mean

Compute mean in parallel via Welford’s online algorithm.

parallel_welford_skewness

Compute skewness in parallel.

parallel_welford_variance

Compute variance in parallel via Welford’s online algorithm.