README.Rmd

---
title: "sparta: Sparse Tables"
output: github_document
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```

<!-- badges: start -->
[![R build status](https://github.com/mlindsk/sparta/workflows/R-CMD-check/badge.svg)](https://github.com/mlindsk/sparta/actions)
[![](https://www.r-pkg.org/badges/version/sparta?color=green)](https://cran.r-project.org/package=sparta)
<!-- badges: end -->

## About

The `sparta` package implements new methods (multiplication, marginalization etc.) for dealing with conditional probability tables (CPTs) and are especially useful when the CPTs are sparse.

## Installation

 - Current stable release from CRAN: 

```{r, eval = FALSE}
install.packages("sparta")
```

 - Current development version: 

```{r, eval = FALSE}
remotes::install_github("mlindsk/sparta")
```

## See Also
The `jti`, https://github.com/mlindsk/jti, package (on CRAN) implements belief propagation via the Junction Tree Algorithm which relies heavily on `sparta` tables. See also  https://github.com/mlindsk/pednoa for a usecase of `jti` where "ordinary/dense" table operations will fail.