---
title: "ranger - Random Forest"
output: html_document
vignette: >
  %\VignetteEncoding{UTF-8}
  %\VignetteIndexEntry{ranger - Random Forest}
  %\VignetteEngine{knitr::rmarkdown}
editor_options:
  chunk_output_type: console
---


```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "vig/"
)
options(rmarkdown.html_vignette.check_title = FALSE)
```

<!-- avoid border around images -->
<style>
img {
  border: 0;
  }
</style>

This guide is designed as a quick-stop reference of how to use some of the more popular machine learning R packages with `vivid`.  In the following examples, we use the air quality data for regression and the iris data for classification. 


```{r, message=FALSE}
library('vivid')
library("ranger")
```

### ranger - Random Forest

The `ranger` package in R is a fast implementation of Random Forests, leveraging optimized C++ code for efficiency.

### Regression

```{r, eval = F}
# load data
aq <- na.omit(airquality)

# build rf model
rf <- ranger(Ozone ~ ., data = aq)

# vivid
vi <- vivi(data = aq, fit = rf, response = 'Ozone')
```



#### Heatmap
```{r, rang_r_heat,  out.width = '100%', eval = F}
viviHeatmap(mat = vi)
```


```{r, echo = F,  out.width = '100%'}
knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/rang_r_heat-1.png")
```
<center>
<caption><span id="fig1:fig1">Figure 1: </span>Heatmap of a ranger random forest regression fit displaying 2-way interaction strength on the off diagonal and individual variable importance on the diagonal.</caption>
  </center>
  
  
#### PDP
```{r, rang_r_pdp, out.width='100%', eval = F}
pdpPairs(data = aq, 
         fit =  rf, 
         response = "Ozone", 
         nmax = 500, 
         gridSize = 20,         
         nIce = 100)
```


```{r, echo = F,  out.width = '100%'}
knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/rang_r_pdp-1.png")
```
<center>
<caption><span id="fig2:fig2">Figure 2: </span>Generalized pairs partial dependence plot for a ranger random forest regression fit.</caption>
</center>

### Classification

```{r, eval = F}

# Load the iris dataset
data(iris)

# Train 
rf <- ranger(Species ~ ., data = iris, probability = T)

vi <- vivi(data = iris, fit = rf, response = 'Species', class = 'setosa')
```

#### Heatmap
```{r, rang_c_heat,  out.width = '100%', eval = F}
viviHeatmap(mat = vi)
```


```{r, echo = F,  out.width = '100%'}
knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/rang_c_heat-1.png")
```
<center>
<caption><span id="fig3:fig3">Figure 3: </span>Heatmap of a ranger random forest classification fit displaying 2-way interaction strength on the off diagonal and individual variable importance on the diagonal.</caption>
  </center>

#### PDP
```{r, rang_c_pdp, out.width='100%', eval = F}
pdpPairs(data = iris, 
         fit =  rf, 
         response = "Species", 
         nmax = 50, 
         gridSize = 4,         
         nIce = 10,
         class = 'setosa')
```
```{r, echo = F,  out.width = '100%'}
knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/rang_c_pdp-1.png")
```
<center>
<caption><span id="fig4:fig4">Figure 4: </span>Generalized pairs partial dependence plot for a ranger random forest classification fit.</caption>
</center>