Lab07_Visualizing-Data-Tips

# Lab07_Visualizing-Data-Tips
## Lab06_ggplot2-layered-graphics
### 曾子軒 Dennis Tseng
### 台大新聞所 NTU Journalism
### 2021/04/13

---

.remark-slide-content {
    padding: 1em 1em 1em 1em;
    font-size: 28px;
}

.my-one-page-font {
  padding: 1em 1em 1em 1em;
  font-size: 20px;
  /*xaringan::inf_mr()*/
}

</style>

# 今日重點

- 沒有作業
- 期中個人資料新聞
- AS05 檢討
- `ggplot2` layers
- Lab067 Practice

---

# 作業檢討

- [Lab06 範例解答](https://p4css.github.io/R4CSS_TA/Lab06_Homework_Visualizing-Date-Time_ref)、[AS05 範例解答](https://p4css.github.io/R4CSS_TA/AS05_Visualizing-Date-Time_ref.html)

---

# ggplot2 layer

- 組成
 - data + `ggplot(aes( ))` + `geom_**()` + ...
 - 資料層 + 美學層 + 幾何層 + ...

source: [Datacamp - Introduction to Data Visualization with ggplot2](https://www.datacamp.com/courses/data-visualization-with-ggplot2-1)

---

# ggplot2 layer

- 概念上
 - 圖表由 layer + scales + coordinate + facet + theme 所組成
- Layer 負責我們在圖表中看到的物件
 - Data
 - Aesthetic mappings
 - A statistical transformation (stat)
 - A geometric object (geom)
 - A position adjustment
 
---

# ggplot2 layer

- 概念上
 - 圖表由 layer + scales + coordinate + facet + theme 所組成
- Scales 負責控制變數轉換到美學的mapping
 - 每個變數都需要一個 scale, e.g. x, y, color, fill, etc.
 - 分為連續和類別，可以手動補值, e.g. `scale_x_continuous()`, `scale_fill_manual()`
 - 因為它控制轉換，所以也包含怎麼用圖例和座標軸向讀者解釋
 - 用英文比教好懂：Each scale is a function from a region in data space (the domain of the scale) to a region in aesthetic space (the range of the scale). The axis or legend is the inverse function: it allows you to convert visual properties back to data.

---

# ggplot2 layer

- Scales 負責控制變數轉換到美學的mapping

Argument name | Axis  | Legend
--------------|-------|------
name | Label |  Title
breaks | Ticks & grid line |  Key 
labels | Tick label |  Key label

---

# ggplot2 layer

- 概念上
 - 圖表由 layer + scales + coordinate + facet + theme 所組成
- Coordinate System 負責把物件位置對應到圖片的平面/表面(plane)上
 - 最常見到二維，但也有非直角坐標的像是 polar system
- Facet 負責將圖表依照特定變數切分成多格，實用所以拉出來談
- Theme 負責實質資料以外的內容
 - 內容非常多樣，舉凡字體、位置、框線等皆屬之
 - 有現成套件如 `library(ggthemes)` 可以直接調用

---

# ggplot2 layer

- 概念上
 - 圖表由 layer + scales + coordinate + facet + theme 所組成
- 實務上
 - data: 你要用的資料，可以很多
 - aesthetics: 選擇要映射的資料變數、怎麼映射，可以很多
 - geometries: 不同類型的圖表, 可以很多
 - coordinate: 有沒有要調整座標系統、座標軸
 - facet: 有沒有要用到 facet
 - scale: 資料變數映射過程中如何對應
 - legend and axis: 視覺變數如何對應回去資料變數
 - title and lab: 標題、次標、註解
 - theme: 通常有背景、框線、字體

---

# ggplot2 layer

```r
library(tidyverse)
diamonds %>% head(5)
```

```
## # A tibble: 5 x 10
##   carat cut     color clarity depth table price     x     y     z
##   <dbl> <ord>   <ord> <ord>   <dbl> <dbl> <int> <dbl> <dbl> <dbl>
## 1 0.23  Ideal   E     SI2      61.5    55   326  3.95  3.98  2.43
## 2 0.21  Premium E     SI1      59.8    61   326  3.89  3.84  2.31
## 3 0.23  Good    E     VS1      56.9    65   327  4.05  4.07  2.31
## 4 0.290 Premium I     VS2      62.4    58   334  4.2   4.23  2.63
## 5 0.31  Good    J     SI2      63.3    58   335  4.34  4.35  2.75
```

---
class: my-one-page-font

# ggplot2 layer: data & aes

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut)
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-4-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: geom

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut) +
  geom_point()
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-5-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: stat

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut) +
  geom_point() +
  geom_smooth(se = F)
```
]

```
## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-6-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: position adjustment

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut) +
  geom_point(position = 
               position_jitter(
                 height = 100, 
                 width = 0.1))
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-7-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: scale and guides

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut) +
  geom_point(position = position_jitter(height = 100, width = 0.1)) +
  scale_color_brewer(name = "cut of diamonds", 
                     palette = 1)
```
The brewer scales provides sequential, diverging and qualitative colour schemes from ColorBrewer.
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-8-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: scale and guides

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-9-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: scale and guides

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-10-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: labs

```r
ggplot(data = diamonds) +
  aes(x = carat, y = price, color = cut) +
  geom_point(position = position_jitter(height = 1), alpha = 0.3) +
  scale_color_brewer(name = "cut of diamonds", palette = 1) +
  scale_x_continuous(name = "carat of the diamond") +
  scale_y_continuous(name = "price of the diamond", 
                     limits = c(0, 22500), breaks = c(2500, 7500, 12500, 17500, 22500), 
                     labels = scales::comma) +
  labs(
    #x = "carat of the diamond",
    #y = "price of the diamond",
    #colour = "cut of the diamond",
    title = "Price and Carat of Diamonds are Highly Correlated",
    subtitle = "Cut of diamonds play a part also",
    caption = "source: library(tidyverse)"
  )
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-11-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: theme

```r
ggplot(data = diamonds, aes(x = carat, y = price, color = cut)) +
  geom_point(position = position_jitter(height = 1), alpha = 0.3) +
  scale_color_brewer(name = "cut of diamonds", palette = 1) +
  scale_x_continuous(name = "carat of the diamond") +
  scale_y_continuous(name = "price of the diamond", limits = c(0, 22500), breaks = c(2500, 7500, 12500, 17500, 22500), labels = scales::comma) +
  labs(title = "Price and Carat of Diamonds are Highly Correlated",
       subtitle = "Cut of diamonds play a part also",
       caption = "source: library(tidyverse)") + 
  theme(plot.title = element_text(face = "bold", size = 16),
    panel.background = element_blank(),
    legend.background = element_rect(fill = "white", size = 4, colour = "white"),
    panel.grid.major = element_line(colour = "grey70", size = 0.2))
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-12-1.png)
]

---

# 「讓我們從頭開始」

---

# ggplot2 layer

- 實務上
 - data: 你要用的資料，可以很多
 - aesthetics: 選擇要映射的資料變數、怎麼映射，可以很多
 - geometries: 不同類型的圖表, 可以很多
 - coordinate: 有沒有要調整座標系統、座標軸
 - facet: 有沒有要用到 facet
 - scale: 資料變數映射過程中如何對應
 - legend and axis: 視覺變數如何對應回去資料變數
 - title and lab: 標題、次標、註解
 - theme: 通常有背景、框線、字體

---
class: my-one-page-font

# ggplot2 layer: data

```
## Warning: package 'maps' was built under R version 4.0.2
```

```
## Warning: package 'sf' was built under R version 4.0.2
```

```r
df_city_taiwan %>% head(2)
```

```
## # A tibble: 2 x 6
##   name     country.etc    pop   lat  long capital
##   <chr>    <chr>        <int> <dbl> <dbl>   <int>
## 1 Changhwa Taiwan      227178  24.1  121.       0
## 2 Chaochou Taiwan       58839  22.6  121.       0
```

```r
sf_county %>% head(2)
```

```
## Simple feature collection with 2 features and 4 fields
## geometry type:  MULTIPOLYGON
## dimension:      XY
## bbox:           xmin: 119.9089 ymin: 24.30943 xmax: 124.5611 ymax: 26.38528
## geographic CRS: TWD97
##   COUNTYID COUNTYCODE COUNTYNAME         COUNTYENG
## 1        Z      09007     連江縣 Lienchiang County
## 2        G      10002     宜蘭縣      Yilan County
##                         geometry
## 1 MULTIPOLYGON (((119.9645 25...
## 2 MULTIPOLYGON (((121.9597 24...
```

---
class: my-one-page-font

# ggplot2 layer: data

```r
sf_county %>% 
  st_crop(xmin=119,xmax=123,ymin=20,ymax=26) %>%
  ggplot() +
  geom_sf() +
  theme_void()
```
]

```
## although coordinates are longitude/latitude, st_intersection assumes that they are planar
```

```
## Warning: attribute variables are assumed to be spatially constant throughout all
## geometries
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-15-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: data

```r
sf_county %>% 
  st_crop(xmin=119,xmax=123,ymin=20,ymax=26) %>%
  ggplot() +
  geom_sf() +
  geom_point(data=df_city_taiwan, aes(x=long, y=lat)) +
  theme_void()
```
]

```
## although coordinates are longitude/latitude, st_intersection assumes that they are planar
```

```
## Warning: attribute variables are assumed to be spatially constant throughout all
## geometries
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-16-1.png)
]
 
 
---

# ggplot2 layer: aesthetics

- aesthetics: 選擇要映射的資料變數、怎麼映射，可以很多
 - 位置（x / y axis）
 - 顏色（color）
 - 大小（size）
 - 透明程度（alpha）
 - 填滿（fill）
 - 形狀（shape）
 - 注意用變數映射與手動指定的差異

---
class: my-one-page-font

# ggplot2 layer: aesthetics

```r
diamonds %>%
  group_by(cut, color) %>%
  summarise(price = mean(price), carat = mean(carat)) %>%
  ungroup() %>%
  ggplot(aes(x = cut, y = price)) +
  geom_bar(stat = "identity")
```
]

```
## `summarise()` has grouped output by 'cut'. You can override using the `.groups` argument.
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-17-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: aesthetics

```r
diamonds %>%
  group_by(cut, color) %>%
  summarise(price = mean(price), carat = mean(carat)) %>%
  ungroup() %>%
  ggplot(aes(x = cut, y = price, fill = color)) +
  geom_bar(stat = "identity")
```
]

```
## `summarise()` has grouped output by 'cut'. You can override using the `.groups` argument.
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-18-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: aesthetics

```r
diamonds %>%
  group_by(cut, color) %>%
  summarise(price = mean(price), carat = mean(carat)) %>%
  ungroup() %>%
  ggplot(aes(x = cut, y = price, alpha = color)) +
  geom_bar(stat = "identity")
```
]

```
## `summarise()` has grouped output by 'cut'. You can override using the `.groups` argument.
```

![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-19-1.png)
]

---

# ggplot2 layer: geometries and statistics

- geometries: 不同類型的圖表, 可以很多
 - `point/jitter`
 - `line/path`
 - `bar/col`
 - `rect/tile/raster`
 - `density`
- statistics: 不同類型的圖表, 可以很多
 - uncertainty: `errorbar/linerange/smooth`
 - weighted data
 - distributions: `histogram/freqpoly/boxplot`

---

# ggplot2 layer: coordinate

- coordinate: 有沒有要調整座標系統、座標軸
 - 線性的: `cartesian/flip/fixed` 分別代表笛卡爾坐標系、翻轉、固定
 - 非線性的: `sf/polar/trans` 分別代表地圖投影、極性、轉換

---
class: my-one-page-font

# ggplot2 layer: coordinate

```r
diamonds %>%
  ggplot(aes(x = depth, y = table, color = carat)) +
  geom_point() +
  coord_fixed()
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-20-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: coordinate

```r
diamonds %>%
  ggplot(aes(x = carat, y = price)) +
  geom_point()
```
]

]

---
class: my-one-page-font

# ggplot2 layer: coordinate

```r
diamonds %>%
  ggplot(aes(x = carat, y = price)) +
  geom_point() +
  coord_trans(y = "log10")
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-22-1.png)
]
---

# ggplot2 layer: facet

- facet: 有沒有要用到 facet
 - 用來分層的 `grid/wrap`，以`facet_wrap()`為例子
 - x ~ y 用來放想要看得變數
 - nrow, ncol 用來擺放列行數量
 - scales 可以選`fixed`(固定), `free`(完全沒限制), `free_x/y`(只限制特定座標軸)

---
class: my-one-page-font

# ggplot2 layer: facet

```r
diamonds %>% 
  ggplot(aes(x = carat, y = price)) +
  geom_point()
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-23-1.png)
]
---
class: my-one-page-font

# ggplot2 layer: facet

```r
diamonds %>% 
  ggplot(aes(x = carat, y = price)) +
  geom_point() +
  facet_wrap(clarity ~ .)
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-24-1.png)
]

---
class: my-one-page-font

# ggplot2 layer: facet

```r
diamonds %>%
  ggplot(aes(x = carat, y = price)) +
  geom_point() +
  facet_wrap(clarity ~ ., scales = "free")
```
]

.pull-right[
![](Lab07_Tutorial_Visualizing-Data-Tips_files/figure-html/unnamed-chunk-25-1.png)
]

---

# ggplot2 layer: scale, and guides

- scale: 資料變數映射過程中如何對應
 - 不連續/連續: `scale_x_discrete()/scale_x_continous()` or y
 - 不連續/連續: `scale_fill_discrete()/scale_fill_continous()` or color
 - 不連續/連續: `scale_fill_brewer()/scale_fill_gradient()` or color
 - 有時候需要手動: `scale_fill_manual()` or color
 - 參數很多，`breaks`, `labels`, `limits` 是最常用到的，有很多方法可以做到同一件事
- guides(legend and axis): 視覺變數如何對應回去資料變數
 - 在裡面放 scale 對應的變數
 - 可以放名字，也可以手動改 label
---

# ggplot2 layer: scale, and guides

- title and lab: 標題、次標、註解
 - 同樣有很多方法可以做到同一件事
- theme: 通常有背景、框線、字體
 - 非常細節，通常有興趣才會研究，直接來練習看看！

---

# [Lab07](https://p4css.github.io/R4CSS_TA/Lab07_Homework_Visualizing-Data-Tips.html)