ggplot.Rmd

---
title: "r4ds chapter 3 ggplot"
author: "Yaqiong Li"
date: "Dec 10, 2018"
---

```{r}
library(tidyverse)
```

```{r}
library(ggplot2)

head(mpg)
```

ggplot() creates a cooridnate systhem that you can add layers on.
```{r}
mpg %>% ggplot(aes(x=displ, y=hwy)) + geom_point()
```

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

# Ex 3.2.4
```{r}
ggplot(data = mpg)
```

```{r}
dim(mpg)
```

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

```{r}
mpg %>% ggplot(aes(x = class, y = drv)) + geom_point()
```

# 3.3
```{r}
mpg %>% ggplot(aes(x=displ, y=hwy)) + geom_point(aes(col = class))
```

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

Notes: to map an aesthetic to a variable, associate the name of the aesthetic to the name of the variable inside aes(). ggplot2 will automatically assign a unique level of aesthetic (color here) to each unique value of the variable, a process known as *scaling*.

Mapping an unordered varialbe (class) to an ordered aesthetic (size) is not a good idea, thus we got a warning for doing this.
```{r}
mpg %>% ggplot(aes(x = displ, y = hwy)) + geom_point(aes (size = class))
```

aes _alpha_ controls the transparency
aes _shape_ controls the shape
```{r}
mpg %>% ggplot() + geom_point(aes(x = displ, y = hwy, alpha = class))
```

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

If you want to manually set the aesthetic properties that do not convey information from the data set, it goes _outside_ of aes()
```{r}
mpg %>% ggplot() + geom_point(aes(x = displ, y = hwy), col = "red")
```

other manual aesthetic settings include
* color
* size (mm)
* the [shape](https://r4ds.had.co.nz/data-visualisation.html#fig:shapes) of points

# Ex 3.3.1
```{r}
glimpse(mpg)
```

A reference about [stroke](https://ggplot2.tidyverse.org/reference/geom_point.html)
```{r}
# For shapes that have a border, you can color inside and outside separately. Use the stroke aesthetic to modify the width of the border

mpg %>% ggplot(aes(x = displ, y = hwy)) + geom_point(shape = 21, color = "red", fill = "blue", size = 5, stroke = 2)
```

```{r}
mpg %>% ggplot(aes(x = displ, y = hwy, col = displ < 5)) + geom_point()
```

# 3.4 Common problems
* If the left hand of the console is a "+", it means that R doesn't think that you have completed an expression. Press *ESCAPE* to start over.

# 3.5 Facets
Facet can be used to split the plot by categorical variable.

Facet by single variable.
```{r}
mpg %>% ggplot(aes(x = displ, y = hwy)) + geom_point() + facet_wrap(~ class, nrow = 2)
```

Facet by combination of variables.
```{r}
mpg %>% ggplot(aes(x = displ, y = hwy)) +
  geom_point() +
  facet_grid(drv ~ cyl)
```

```{r}
mpg %>% ggplot(aes(x = displ, y = hwy)) +
  geom_point() +
  facet_grid( .~ cyl) # . means not into r*c
```

# Ex 3.5.1
```{r}
mpg %>% ggplot(aes(x = displ, y = hwy)) +
  geom_point() +
  facet_grid(drv ~.)
```

# 3.6 Geometric objects
```{r}
g <- mpg %>% ggplot(aes(x = displ, y = hwy))
g + geom_point()
```

```{r}
g + geom_smooth()
```

```{r}
g + geom_smooth(aes(linetype = drv))
```
```{r}
g + geom_point(aes(col = drv)) + geom_smooth(aes(linetype = drv))
```