README.Rmd

---
output: 
  github_document
encoding: 'UTF-8'
---


```{r, include = FALSE}
knitr::opts_chunk$set(
  fig.path = "man/figures/",
  comment = NA
)
```

<!-- badges: start -->

[![R build status](https://github.com/m-clark/confusionMatrix/workflows/R-CMD-check/badge.svg)](https://github.com/m-clark/confusionMatrix/actions)
  [![Codecov test coverage](https://codecov.io/gh/m-clark/confusionMatrix/branch/master/graph/badge.svg)](https://codecov.io/gh/m-clark/confusionMatrix?branch=master)
  [![Lifecycle: maturing](https://img.shields.io/badge/lifecycle-experimental-blue.svg)](https://www.tidyverse.org/lifecycle/#experimental)

<!-- badges: end -->
 

# confusionMatrix

<br>

Given predictions and a target variable, provide numerous statistics from the resulting confusion matrix. The goal is to provide a wealth of summary statistics that can be calculated from a single confusion matrix, and return <span class="emph" style = "">**tidy**</span> results with as few dependencies as possible.




```{r demo}
library(confusionMatrix)

p = sample(letters[1:2], 250, replace = T, prob = 1:2)
o = sample(letters[1:2], 250, replace = T, prob = 1:2)

result = confusion_matrix(
  prediction = p,
  target = o,
  return_table = TRUE
)

result

result$Accuracy$`Frequency Table`
```

```{r demo-longer}
result = confusion_matrix(
  prediction = p,
  target = o,
  longer = TRUE
)

result
```


### Installation

To install from GitHub the <span class="pack">devtools</span> package is required.

```{r eval=FALSE}
devtools::install_github('m-clark/confusionMatrix')
```