---
title: "ggplot2"
output:
  html_document: default
  pdf_document: default
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r, message=FALSE}
library(tidyverse) # ggplot2 and friends
```

Focuses on when data is given. 

# Visualisation

```{r}
mpg
```

```{r}
plot(hwy ~ displ, mpg)
```

```{r}
ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy))
```

```{r}
# Set black/white theme with larger font size than default
theme_set(theme_bw(base_size = 20))
```

```{r}
ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy))
```

```{r}
ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy, color = class))
```


ggplot2 (part of tidyverse) based on (layered) grammar of graphics.

```
ggplot(data = <DATA>) +
  <GEOM_FUNCTION>(mapping = aes(<MAPPINGS>))

ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) +
  <GEOM_FUNCTION>()
```

* Data + aesthetic mapping (`aes`) of how to perceive data
* Layers: `geom_`* + `stat_`*
* Scales (color, size, ...)
* Coordinate (cartesian, polar, log10, ...)
* Faceting (conditioning or latticing/trellising)
* Theme (font size, background color, ...)
* No suggestions of which plots to use. 
* No interactivity, only static graphics.
* Reference: http://r4ds.had.co.nz/data-visualisation.html
* Documentation: # http://ggplot2.tidyverse.org/reference/

> Exercises 3.2.4


```
ggplot(data = mpg) + 
    geom_histogram()
```

```{r}
ggplot(data = mpg, mapping = aes(x = hwy))

ggplot(data = mpg, mapping = aes(x = hwy)) + 
  geom_histogram()

ggplot(data = mpg, mapping = aes(hwy)) + 
  geom_histogram(binwidth = nclass.FD)

ggplot(mpg, aes(hwy)) + 
  geom_histogram()

ggplot(mpg) + 
  geom_histogram(aes(hwy))

# Error:
#ggplot() + 
#  geom_histogram(mpg, aes(hwy))

ggplot() + 
  geom_histogram(aes(hwy), mpg)

ggplot() + 
  geom_histogram(data = mpg, aes(hwy))
```

Generic template:

```
ggplot(data = <DATA>) +
  <GEOM_FUNCTION>(mapping = aes(<MAPPINGS>))
```

Mapping (`aes()`) in `ggplot()` is inherited to layers (`geom_*()`/`stat_*()`) if not overridden.

## Captions, labels, themes

```{r}
ggplot(mpg, aes(hwy)) + 
  geom_histogram() + 
  labs(title = "Fuel economy data", 
       subtitle = "1999 and 2008 for 38 popular models of car",
       x = "Highway miles per gallon", 
       y = "Count") +
  theme_dark() 
```

> 3.3.1 Exercises


## Faceting/conditioning/latticing/trellising

```{r}
ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy)) + 
  facet_wrap(~ class, nrow = 3)

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy)) + 
  facet_grid(drv ~ cyl, labeller = label_both)

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy)) + 
  facet_grid(drv ~ cyl)

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy)) + 
  facet_wrap(~ cyl + drv, labeller = label_both)


ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy, color = class)) + 
  facet_grid(drv ~ cyl, labeller = label_both)

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy, size = year, color = class)) + 
  facet_grid(drv ~ cyl, labeller = label_both) 

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy, size = year, color = class)) + 
  facet_grid(drv ~ cyl, labeller = label_both) + 
  theme(legend.position = "bottom") + 
  scale_color_discrete("Class of car") + 
  guides(color = guide_legend(title.position = "top",
                              ncol = 1))

ggplot(data = mpg) + 
  geom_point(mapping = aes(x = displ, y = hwy, size = year, color = class)) + 
  facet_grid(drv ~ cyl, labeller = label_both) +
  scale_size_continuous("Year", range = c(1, 10)) +
  scale_color_discrete("Type") +
  labs(title = "Title", 
       subtitle = "Subtitle",
       x = "Displacement [l]", 
       y = "Highway fuel economy [mpg]")
```

> 3.5.1 Exercises


## Bar plots

```{r}
ggplot(data = mpg, mapping = aes(x = class)) + 
  geom_bar()
```

```{r}
ggplot(mpg, aes(class)) + 
  geom_bar()
```

```{r}
p <- ggplot(mpg, aes(manufacturer)) + 
  geom_bar()
p
print(p)
1:5
print(1:5)
```

```{r}
p2 <- p +
  theme(axis.text.x = element_text(angle = 90, 
                                   hjust = 1, 
                                   vjust = 0.5))
p2
```

```{r}
p2 + scale_y_continuous(breaks = scales::pretty_breaks(n = 10))

#p2 + scale_y_continuous(breaks = c(0, 7, 17))
```

```{r}
library(scales)
p2 + scale_y_continuous(breaks = pretty_breaks(n = 10))
```

```{r}
ggplot(mpg, aes(class, fill = cyl)) + 
  geom_bar()

str(mpg$cyl)
table(mpg$cyl)

ggplot(mpg, aes(class, fill = factor(cyl))) + 
  geom_bar()

ggplot(mpg, aes(class, fill = factor(cyl))) + 
  geom_bar()

ggplot(mpg, aes(class, fill = factor(cyl))) + 
  geom_bar(position = position_dodge())
```

## Points / scatter plots

```{r}
ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_point()

ggplot(mpg, aes(x = displ, y = hwy, color = "blue")) + 
  geom_point()

ggplot(mpg, aes(x = displ, y = hwy), color = "blue") + 
  geom_point()

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_point(color = "blue")


ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy))

ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy, color = "blue"))

ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy), color = "blue")


ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy, size = 10))

ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy), size = 10)

ggplot(mpg) + 
  geom_point(aes(x = displ, y = hwy, size = class))
```

```{r}
ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_point()

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_point(alpha = 0.2)

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_count()

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_jitter()

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_jitter(alpha = 0.2) # control jitter with width and height parameters

ggplot(mpg, aes(x = displ, y = hwy)) + 
  geom_jitter(height = 0, width = 0.1)
```


## log-scale

```{r}
ggplot(cars, aes(x = speed, y = dist)) + 
  geom_point()

ggplot(cars, aes(x = speed, y = dist)) + 
  geom_point() +
  scale_x_log10() + 
  scale_y_log10()

library(scales)
ggplot(cars, aes(x = speed, y = dist)) + 
  geom_point() +
  scale_x_log10(breaks = trans_breaks("log10", function(x) 10^x),
                labels = trans_format("log10", math_format(10^.x))) +
  scale_y_log10(breaks = trans_breaks("log10", function(x) 10^x),
                labels = trans_format("log10", math_format(10^.x))) +
  annotation_logticks()

# http://ggplot2.tidyverse.org/reference/annotation_logticks.html
```

## Boxplots

```{r}
ggplot(mpg, aes(x = class, y = hwy)) + 
  geom_boxplot()

ggplot(mpg, aes(x = class, y = hwy, color = class)) + 
  geom_boxplot()

p <- ggplot(mpg, aes(x = class, y = hwy, color = class)) + 
  geom_boxplot()

p

pdf("fig1.pdf")
print(p)
dev.off()

ggsave("fig2.pdf", p)
ggsave("fig2.pdf", p, width = 8, height = 6)
```

> 3.8.1 Exercises