Algunas funciones favoritas de {broom}

class: center, middle, inverse, title-slide

# Algunas funciones favoritas de {broom}
### Alejandra Tapia Silva
### R-Ladies Talca
### 30/04/2021

---

class: center, middle

<div>
<style type="text/css">.xaringan-extra-logo {
width: 110px;
height: 128px;
z-index: 0;
background-image: url(https://rladies.org/wp-content/uploads/2016/12/R-LadiesGlobal.png);
background-size: contain;
background-repeat: no-repeat;
position: absolute;
top:1em;right:1em;
}
</style>
<script>(function () {
  let tries = 0
  function addLogo () {
    if (typeof slideshow === 'undefined') {
      tries += 1
      if (tries < 10) {
        setTimeout(addLogo, 100)
      }
    } else {
      document.querySelectorAll('.remark-slide-content:not(.title-slide):not(.inverse):not(.hide_logo)')
        .forEach(function (slide) {
          const logo = document.createElement('div')
          logo.classList = 'xaringan-extra-logo'
          logo.href = null
          slide.appendChild(logo)
        })
    }
  }
  document.addEventListener('DOMContentLoaded', addLogo)
})()</script>
</div>

## ¿Por qué funciones favoritas de {broom}?

![](https://media.giphy.com/media/3JUsda8zZiFvQD0ioo/giphy.gif)
---
class: center, middle

## `Porque son funciones que integran los datos y resultados del modelamiento`

---
class: center, middle

## `Esto facilita informar los resultados, crear gráficos y trabajar con un gran número de modelos a la vez`

---
class: left

## Paquetes

.center[
<img src=imgs/hex_tidyverse.png width="30%">
<img src=imgs/hex_tidymodels.png width="30%">
<img src=imgs/hex_broom.png width="30%"> 
]

---
class: Left, middle

## Funciones <img src="imgs/hex_broom.png" alt="Sharingan" width="10%" align="center"/>

###`>` tidy()

###`>` glance()

###`>` augment()

---
class: left, middle

## Cargar paquetes 
.left[

```r
library(tidyverse)
library(tidymodels)
library(broom)
```
]

---
class: center, middle
## El conjunto de datos

---
class: left, middle
## `Especies de tortugas en las distintas islas Galápagos` 🐢 🏝

```r
library(faraway)
glimpse(gala)
```

```
## Rows: 30
## Columns: 7
## $ Species   <dbl> 58, 31, 3, 25, 2, 18, 24, 10, 8, 2, 97, 93, 58, 5, 40, 347, …
## $ Endemics  <dbl> 23, 21, 3, 9, 1, 11, 0, 7, 4, 2, 26, 35, 17, 4, 19, 89, 23, …
## $ Area      <dbl> 25.09, 1.24, 0.21, 0.10, 0.05, 0.34, 0.08, 2.33, 0.03, 0.18,…
## $ Elevation <dbl> 346, 109, 114, 46, 77, 119, 93, 168, 71, 112, 198, 1494, 49,…
## $ Nearest   <dbl> 0.6, 0.6, 2.8, 1.9, 1.9, 8.0, 6.0, 34.1, 0.4, 2.6, 1.1, 4.3,…
## $ Scruz     <dbl> 0.6, 26.3, 58.7, 47.4, 1.9, 8.0, 12.0, 290.2, 0.4, 50.2, 88.…
## $ Adjacent  <dbl> 1.84, 572.33, 0.78, 0.18, 903.82, 1.84, 0.34, 2.85, 17.95, 0…
```

---
class: center, middle

### `Supongamos que queremos saber si existen influencia de la elevación de la isla en metros (Elevation) y el número de especies de tortugas que se encuentran en la isla (Species)`

.center[
<img src=imgs/turtle.png width="60%">
]

---
class: left, middle
.left[
## Ajustar un modelo lineal

```r
model_fit <- lm(Species ~ Elevation, data=gala)
```
]

---
class: left, middle
## tidy()

### Construye un tibble que resume información estadística sobre el ajuste del modelo

.left[

```r
tidy(model_fit) %>% tibble::as_tibble()
```

```
## # A tibble: 2 x 5
##   term        estimate std.error statistic    p.value
##   <chr>          <dbl>     <dbl>     <dbl>      <dbl>
## 1 (Intercept)   11.3     19.2        0.590 0.560     
## 2 Elevation      0.201    0.0346     5.80  0.00000318
```
]

---
class: left, middle
## glance()

### Construye un resumen de otras informaciones relacionadas al ajuste del modelo 
.left[

```r
glance(model_fit) %>% tibble::as_tibble()
```

```
## # A tibble: 1 x 12
##   r.squared adj.r.squared sigma statistic    p.value    df logLik   AIC   BIC
##       <dbl>         <dbl> <dbl>     <dbl>      <dbl> <dbl>  <dbl> <dbl> <dbl>
## 1     0.545         0.529  78.7      33.6 0.00000318     1  -172.  351.  355.
## # … with 3 more variables: deviance <dbl>, df.residual <int>, nobs <int>
```
]

---
class: left, middle
## augment()

### Agregar columnas con información relacionada a los datos que se modelaron
.left[

```r
augment(model_fit)
```

```
## # A tibble: 30 x 9
##    .rownames   Species Elevation .fitted .resid   .hat .sigma .cooksd .std.resid
##    <chr>         <dbl>     <dbl>   <dbl>  <dbl>  <dbl>  <dbl>   <dbl>      <dbl>
##  1 Baltra           58       346    80.8 -22.8  0.0334   80.0 1.50e-3    -0.295 
##  2 Bartolome        31       109    33.2  -2.22 0.0464   80.1 2.03e-5    -0.0289
##  3 Caldwell          3       114    34.2 -31.2  0.0459   79.9 3.97e-3    -0.406 
##  4 Champion         25        46    20.6   4.43 0.0535   80.1 9.45e-5     0.0579
##  5 Coamano           2        77    26.8 -24.8  0.0498   80.0 2.74e-3    -0.323 
##  6 Daphne.Maj…      18       119    35.2 -17.2  0.0454   80.0 1.19e-3    -0.224 
##  7 Daphne.Min…      24        93    30.0  -6.01 0.0480   80.1 1.55e-4    -0.0783
##  8 Darwin           10       168    45.1 -35.1  0.0411   79.8 4.44e-3    -0.455 
##  9 Eden              8        71    25.6 -17.6  0.0504   80.0 1.40e-3    -0.229 
## 10 Enderby           2       112    33.8 -31.8  0.0461   79.9 4.14e-3    -0.414 
## # … with 20 more rows
```
]

---
class: left, middle
## augment()

.pull-left[

```r
info_fit<-
  augment(model_fit)

info_fit %>% 
ggplot(aes(x=Elevation,y=Species))+ 
geom_jitter(alpha=.2)+
geom_line(aes(x=Elevation,y=.fitted))+
theme_bw()
```
]

.pull-right[
![](index_files/figure-html/unnamed-chunk-7-1.png)
]

---
class: left

##`Referencias`📖

### > David Robinson y + 122

https://github.com/tidymodels/broom

### > Max Kuhn, Hadley Wickham and RStudio

https://www.tidymodels.org/

---
class: center, middle

.center[<img src=imgs/R-Ladies_Talca_hex.png width="30%">]

##### Presentación creada con el paquete [**xaringan**](https://github.com/yihui/xaringan) de [**Yihui Xie**](https://github.com/yihui) y el tema [**rladies**](https://github.com/rbind/apreshill/blob/master/static/slides/rladies-demo-slides.Rmd) de [**Alison Hill**](https://github.com/apreshill)

<img src="imgs/logo_twitter.png" alt="Sharingan" width="6%" align="center"/>
`@aleants @RLadiesTalca` 
<img src="imgs/logo_instagram.png" alt="Sharingan" width="6%" align="center"/>
`@rladiestalca`