Upstream example coverage
Disposition of every notebook and script in Numerics-Python-Examples with respect to this port. “Ported” examples reproduce the upstream results (exactly where seeded and deterministic). “Recast” examples cover the same statistical content through this package’s public API, where the upstream code drove the C# classes directly in ways this API deliberately does not expose (notably custom prior lists and Python-callback likelihoods). “Out of scope” areas were never part of the C++ port, which targets the probability, estimation, and analysis layers of Numerics plus all of RMC-BestFit.
| Upstream | Topic | Status | Notes |
|---|---|---|---|
00_getting_started |
Loading the library, first distribution | Ported | pythonnet/DLL plumbing replaced by a plain package install |
01_distributions |
Distribution tour, sampling, Latin hypercube | Ported | all 10 families and the seeded draws reproduce exactly |
02_distribution_fitting |
MOM / MLE / L-moment fitting | Ported | fitted parameters reproduce exactly |
03_mcmc_basics |
RWMH fundamentals | Recast | fixed-parameter model and hand-picked priors not expressible; recast to mcmc_sample() |
04_mcmc_bayesian_inference |
Bayesian flood frequency | Ported | same priors, seed, and serial chains as upstream; acceptance streams bit-exact vs C#, posterior tables reproduce at displayed precision (documented ulp drift in the MAP optimizer) |
05_mcmc_adaptive |
Adaptive sampler comparison | Recast | PyMC benchmark sections dropped |
06_mcmc_diagnostics |
R-hat, ESS, autocorrelation | Recast | diagnostics on arbitrary external chains not exposed |
07_integration_and_root_finding |
Numerical integration, root finding | Out of scope | the Numerics Mathematics.Integration / RootFinding namespaces are not part of the port |
08_optimization |
Local and global optimizers | Out of scope | only the optimizers the estimation layer needs are ported (BFGS, Nelder-Mead, Brent, Powell, DE, MLSL), and they are internal |
09_statistics |
Statistical tests, transforms, outliers | Ported (partial) | MGBT, Box-Cox/Yeo-Johnson, plotting positions, bootstrap covered; t/Mann-Whitney/KS tests and Jenks breaks are not ported |
10_time_series |
TimeSeries container operations | Out of scope | the heavy Numerics TimeSeries container (interpolation, file I/O, USGS download) is a documented severance; AR/MA/ARIMA/ARIMAX models are ported |
11_machine_learning |
K-Means, random forest, KNN, GMM | Out of scope | the Numerics.MachineLearning namespace is not part of the port |
12_linear_models |
GLM regression | Out of scope | the Numerics.MachineLearning GLM layer is not part of the port |
examples/bayesian_regression.py |
DEMCzs linear regression | Recast | recast onto arimax_analysis with a linear trend |
examples/flood_frequency_analysis.py |
Distribution ranking, design flows | Ported | fits and return-period tables reproduce exactly; ranking by log-likelihood/AIC instead of the unported KS test |
examples/reliability_analysis.py |
Monte Carlo reliability | Ported | uses the LnNormal family in place of C# LogNormal.Base = e; results reproduce exactly |