The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

PTT 1.0.1

Correctness

• Fixes undefined behavior in conditional hierarchical MAP partition output caused by using an unsigned counter with -1 as its initial sentinel.
• Uses 64-bit node offsets to avoid truncation when indexing larger trees.
• Rejects predictor/response dimension and shrinkage-state configurations that exceed the capacity of the internal partition representation.

Validation

• Adds regression checks for conditional partition boundaries and supported input capacities.
• Audits the remaining bit shifts and verifies the package examples, tests, and demos with undefined-behavior and address sanitizers.

PTT 1.0

PTT 1.0 is the first official release of the package.

Models

• Provides adaptive Pólya tree (APT), Markov APT, and optional Pólya tree (OPT) models for Bayesian nonparametric density estimation.
• Provides conditional APT and conditional OPT models for conditional-density estimation through recursive partitioning of the predictor space.
• Supports univariate and multivariate observations on the unit hypercube or a data-adaptive standardized hyperrectangle.

Inference and results

• Uses exact forward-backward inference on finite recursive partition trees.
• Computes log marginal likelihoods, posterior root shrinkage probabilities, posterior predictive densities, and hierarchical MAP partitions.
• Generates posterior partition samples using R’s random-number generator for reproducible simulation.
• Reports predictive densities in the physical coordinates of the selected sample space.

Package interface

• Supplies documented R interfaces backed by registered C++ routines using Rcpp and RcppArmadillo.
• Validates observations, prediction matrices, sample spaces, and model parameters before native computation.
• Includes plotting utilities for two-dimensional partitions and ROC curves.
• Includes package-level documentation, method references, citation metadata, API and numerical regression tests, and seven reproducible visual demos for marginal and conditional models in one and two dimensions.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.