---
title: "gbm - Generalized Boosted Model"
output: html_document
vignette: >
  %\VignetteEncoding{UTF-8}
  %\VignetteIndexEntry{gbm - Generalized Boosted Model}
  %\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 example, we use the air quality data for regression.

### GBM

The `gbm` package in R stands for Generalized Boosted Models. It offers an efficient implementation of gradient boosting algorithms for classification, regression, and other machine learning tasks.

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



### Regression

```{r, eval=FALSE}

# load data
aq <- na.omit(airquality)

# build SVM model
gb <-  gbm(Ozone ~ ., 
           data = aq, 
           distribution = "gaussian")

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


<!-- #### Heatmap -->
<!-- ```{r, gb_r_heat,  out.width = '100%', eval=FALSE} -->
<!-- viviHeatmap(mat = vi) -->
<!-- ``` -->

<!-- ```{r, echo = F,  out.width = '100%'} -->
<!-- knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/gb_r_heat-1.png") -->
<!-- ``` -->
<!-- <center> -->
<!-- <caption><span id="fig1:fig1">Figure 1: </span>Heatmap of a GBM regression fit displaying 2-way interaction strength on the off diagonal and individual variable importance on the diagonal.</caption> -->
<!--   </center> -->

#### PDP
```{r, gb_r_pdp, out.width='100%', eval=FALSE}
pdpPairs(data = aq, 
         fit =  gb, 
         response = "Ozone", 
         nmax = 50, 
         gridSize = 4,         
         nIce = 10)
```

```{r, echo = F,  out.width = '100%'}
knitr::include_graphics("https://raw.githubusercontent.com/AlanInglis/vivid/master/vignettes/vig/gb_r_pdp-1.png")
```
<center>
<caption><span id="fig1:fig1">Figure 1: </span>Generalized pairs partial dependence plot for a GBM regression fit.</caption>
</center>



<!-- ### Classification -->

<!-- ```{r, eval=FALSE} -->

<!-- # Load the iris dataset -->
<!-- data(iris) -->

<!-- # Train  -->
<!-- gb <- gbm( -->
<!--   Species ~ .,  -->
<!--   data = iris,  -->
<!--   distribution = "multinomial" -->
<!-- ) -->


<!-- # vivi matrix -->
<!-- vi <- vivi(data = iris, fit = gb, response = 'Species') -->
<!-- ``` -->


<!-- #### Heatmap -->
<!-- ```{r, gb_c_heat,  out.width = '100%', eval=FALSE} -->
<!-- viviHeatmap(mat = vi) -->
<!-- ``` -->

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