class: center, middle, inverse, title-slide

# Boosted Trees in R
### Dr. D’Agostino McGowan

---

layout: true

<div class="my-footer">
<span>
Dr. Lucy D'Agostino McGowan
</span>
</div>

---

## Boosted trees in R

```r
library(tidymodels)
```

```r
*boost_spec <- boost_tree(
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
* set_engine("xgboost")
```

---

## Boosted trees in R

```r
library(tidymodels)
```

```r
boost_spec <- boost_tree( 
* mode = "classification",
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost") 
```

* Set the `mode` as you would with a bagged tree or random forest

---

## Boosted trees in R

```r
library(tidymodels)
```

```r
boost_spec <- boost_tree( 
  mode = "classification", 
* tree_depth = 1,
  trees = 1000,
  learn_rate = 0.001,
) %>% 
  set_engine("xgboost") 
```

* `tree_depth` here is the depth of each tree, let's set that to 1

---

## Boosted trees in R

```r
library(tidymodels)
```

```r
boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
* trees = 1000,
  learn_rate = 0.001,
) %>% 
  set_engine("xgboost") 
```

* `trees` is the number of trees that are fit, this is equivalent to `B`

---

## Boosted trees in R

```r
library(tidymodels)
```

```r
boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
* learn_rate = 0.001,
) %>% 
  set_engine("xgboost") 
```

* `learn_rate` is `\(\lambda\)`

---

## Fit the model

```r
model <- fit(boost_spec,
             HD ~ Age + Sex + ChestPain + RestBP + Chol + Fbs + 
               RestECG + MaxHR + ExAng + Oldpeak + Slope + Ca + Thal,               
             data = heart)
```

---

class: inverse

<div class="countdown" id="timer_5e9506b3" style="right:0;bottom:0;" data-warnwhen="0">
<code class="countdown-time"><span class="countdown-digits minutes">03</span><span class="countdown-digits colon">:</span><span class="countdown-digits seconds">00</span></code>
</div>

## <i class="fas fa-laptop"></i> `Boosting`

How would this code change if I wanted to tune `B` the number of bootstrapped training samples?

```r
boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost") 
```

Notes for current slide

Notes for next slide

Boosted Trees in RDr. D’Agostino McGowan1 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree(
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost")

2 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree( 
  mode = "classification",
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost")

Set the mode as you would with a bagged tree or random forest

3 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1,
  trees = 1000,
  learn_rate = 0.001,
) %>% 
  set_engine("xgboost")

tree_depth here is the depth of each tree, let's set that to 1

4 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000,
  learn_rate = 0.001,
) %>% 
  set_engine("xgboost")

trees is the number of trees that are fit, this is equivalent to B

5 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001,
) %>% 
  set_engine("xgboost")

learn_rate is \(\lambda\)

6 / 8

Fit the model

model <- fit(boost_spec,
             HD ~ Age + Sex + ChestPain + RestBP + Chol + Fbs + 
               RestECG + MaxHR + ExAng + Oldpeak + Slope + Ca + Thal,               
             data = heart)

7 / 8

03:00

`Boosting`

How would this code change if I wanted to tune B the number of bootstrapped training samples?

boost_spec <- boost_tree( 
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost")

8 / 8

Boosted trees in R

library(tidymodels)

boost_spec <- boost_tree(
  mode = "classification", 
  tree_depth = 1, 
  trees = 1000, 
  learn_rate = 0.001, 
) %>% 
  set_engine("xgboost")

2 / 8

Paused

Help

Keyboard shortcuts

↑, ←, Pg Up, k	Go to previous slide
↓, →, Pg Dn, Space, j	Go to next slide
Home	Go to first slide
End	Go to last slide
Number + Return	Go to specific slide
b / m / f	Toggle blackout / mirrored / fullscreen mode
c	Clone slideshow
p	Toggle presenter mode
t	Restart the presentation timer
?, h	Toggle this help

Esc	Back to slideshow