https://github.com/mattwarkentin/ordered

Bindings for ordinal classification models for use with the 'parsnip' package, such as ordinal random forests by Hornung R. (2020) <doi:10.1007/s00357-018-9302-x> and others.
https://github.com/mattwarkentin/ordered

Last synced: 5 months ago
JSON representation

Bindings for ordinal classification models for use with the 'parsnip' package, such as ordinal random forests by Hornung R. (2020) <doi:10.1007/s00357-018-9302-x> and others.

• Host: GitHub
• URL: https://github.com/mattwarkentin/ordered
• Owner: mattwarkentin
• License: other
• Created: 2024-10-27T21:18:06.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2024-10-29T11:01:27.000Z (7 months ago)
• Last Synced: 2024-12-24T12:38:08.119Z (5 months ago)
• Language: R
• Size: 57.6 KB
• Stars: 6
• Watchers: 2
• Forks: 1
• Open Issues: 7
• Metadata Files:
  • Readme: README.Rmd
  • License: LICENSE
  • Code of conduct: CODE_OF_CONDUCT.md

Awesome Lists containing this project

README

        
---

output: github_document

---

```{r, include = FALSE}

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "#>",

  fig.path = "man/figures/README-",

  out.width = "100%"

)

```

# ordered

[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)

[![CRAN status](https://www.r-pkg.org/badges/version/ordered)](https://CRAN.R-project.org/package=ordered)

[![R-CMD-check](https://github.com/topepo/ordered/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/topepo/ordered/actions/workflows/R-CMD-check.yaml)

[![Codecov test coverage](https://codecov.io/gh/topepo/ordered/branch/main/graph/badge.svg)](https://app.codecov.io/gh/topepo/ordered?branch=main)

The goal of ordered is to enable additional classification models for ordinal outcomes (e.g., "low", "medium", "high"). While there are several model/engine combinations in the parsnip package that can be used, this package adds:

 - ordinal forests [Hornung R. (2020)](https://scholar.google.com/scholar?hl=en&as_sdt=0%2C7&q=10.1007%2Fs00357-018-9302-x&btnG=)

More will be added. 

There are some existing features in tidymodels packages that are useful for ordinal outcomes: 

 - The "partykit" engines for [parsnip::decision_tree()] and [parsnip:: rand_forest()] use the ordered nature of the factors to train the model. 

 - The yardstick package has [yardstick::kap()] for weighted and unweighted Kappa statistics (the former being of more interest). Also, [yardstick::classification_cost()] can utilize more complex cost structures and uses the class probabilities for estimation. 

## Installation

You can install the development version of ordered like so:

``` r

pak::pak("topepo/ordered")

```

## Example

Here is a simple example using computational chemistry data to predict the permeability of a molecule: 

```{r}

library(ordered)

library(dplyr)

data(caco, package = "QSARdata")

caco_dat <-

  inner_join(caco_Outcome, caco_Dragon, by = "Molecule") %>%

  as_tibble() %>%

  select(class = Class, mol_weight = QikProp_mol_MW,

         volume = QikProp_volume, ClogP)

  caco_train <- caco_dat[-(1:5), ]

  caco_test  <- caco_dat[ (1:5), ]

ord_rf_spec <- 

  rand_forest(mtry = 2, trees = 100) %>% # you should really use many more trees

  set_mode("classification") %>%

  set_engine("ordinalForest")

set.seed(382)

ord_rf_fit <- ord_rf_spec %>% fit(class ~ ., data = caco_train)

augment(ord_rf_fit, caco_test)

```

## Code of Conduct

Please note that the ordered project is released with a [Contributor Code of Conduct](https://contributor-covenant.org/version/2/1/CODE_OF_CONDUCT.html). By contributing to this project, you agree to abide by its terms.