Lab08_Tutorial_Text-Processing.Rmd

---
title: "Lab06_Visualizing-Text-Data"
subtitle: "Lab06_ggplot2-string-text"
author: "曾子軒 Dennis Tseng"
institute: "台大新聞所 NTU Journalism"
date: "2021/04/13"
output:
  xaringan::moon_reader:
    css: [default, metropolis, metropolis-fonts]
    lib_dir: libs
    nature:
      highlightStyle: github
      highlightLines: true
      countIncrementalSlides: false
      self_contained: true
---
```{r setup, cache = F, echo=F}
knitr::opts_chunk$set(error = TRUE)
```

<style type="text/css">

.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

- 分組
- AS06 Preview
- dplyr: more
- Lab08 Practice

---

class: inverse, center, middle

# 分組

---

class: inverse, center, middle

# [AS06](https://p4css.github.io/R4CSS_TA/AS06_Visualizing-Date-Time.html)

---

# dplyr 的未竟之業 - select()

- 懶人的福音，幫助你快速選 column
- operator: `:`, `!`, `&`, `|`, `c()`
- selection helpers: 
      - specific columns: `everything()`, `last_col()`
      - matching patterns: `starts_with()`, `ends_with()`, `contains()`, `matches()`, `num_range()`
      - character vector: `all_of()`, `any_of()`
      - 搭配 function: `where()`

---

```{r message=F, warning=F}
library(tidyverse)

df_marriage <- read_csv("data/Lab04/109Q4_county_marriage.csv") %>%
   mutate(across(where(is.character), ~iconv(.,from = "BIG5", to = "UTF8"))) %>% slice(-1) %>% mutate(across(matches("MARRY"), ~as.integer(.))) %>%
  `colnames<-`(str_to_lower(colnames(.)))

head(df_marriage, 3)
```

---

# dplyr 的未竟之業 - select()

- operator: `:`, `!`, `&`, `|`, `c()`

```{r message=F, warning=F}
df_marriage %>% slice(1)
df_marriage %>% select(county:marry_cnt) %>% slice(1)
df_marriage %>% select(1:2, 4) %>% slice(1)
df_marriage %>% select(!marry_cnt) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- specific columns: `everything()`, `last_col()`

```{r message=F, warning=F}
df_marriage %>% select(info_time, everything()) %>% slice(1)
df_marriage %>% select(-county_id, everything(), county_id) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- specific columns: `everything()`, `last_col()`

```{r message=F, warning=F}
df_marriage %>% select(last_col()) %>% slice(1)
df_marriage %>% select(1:last_col(1)) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- matching patterns: `starts_with()`, `ends_with()`, `contains()`, `matches()`, `num_range()`

```{r message=F, warning=F}
df_marriage %>% select(starts_with("marry")) %>% slice(1)
df_marriage %>% select(starts_with(c("marry_cp", "county"))) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- matching patterns: `starts_with()`, `ends_with()`, `contains()`, `matches()`, `num_range()`

```{r message=F, warning=F}
df_marriage %>% select(contains("marry")) %>% slice(1)
df_marriage %>% select(contains("marry.*cnt")) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- matching patterns: `starts_with()`, `ends_with()`, `contains()`, `matches()`, `num_range()`
- 注意！ `matches()` 放正規表示式

```{r message=F, warning=F}
df_marriage %>% select(matches("marry.*cnt")) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- selection helpers: 
      - character vector: `all_of()`, `any_of()`

```{r message=F, warning=F}
vars <- c("marry_m_cnt", "marry_f_cnt")
vars2 <- c("marry_m_cnt", "marry_f_cnt", "divorce_m_cnt", "divorce_f_cnt")

df_marriage %>% select(all_of(vars)) %>% slice(1)
df_marriage %>% select(all_of(vars2)) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- selection helpers: 
      - character vector: `all_of()`, `any_of()`

```{r message=F, warning=F}
df_marriage %>% select(any_of(vars)) %>% slice(1)
df_marriage %>% select(any_of(vars2)) %>% slice(1)
```

---

# dplyr 的未竟之業 - select()

- selection helpers: 
      - 搭配 function: `where()`
      - 通常跟 `across()` 一起使用

---

# dplyr 的未竟之業 - across()

- 懶人的福音，幫助你對不同 column 使用 function
- Apply a function (or functions) across multiple columns
- 動詞裡面放 `across(.cols = everything(), .fns = NULL, ..., .names = NULL)`
- 先選你要的欄位，接著指定函數
- 欄位部分可以活用上面的教的 selection 方法，函數可以使用完整的或匿名函數

---

# dplyr 的未竟之業 - across()

```{r message=F, warning=F}
df_marriage %>% mutate(across(matches("marry_"), ~(./100))) %>% slice(1)
df_marriage %>% summarise(across(where(is.numeric), ~sum(.)))
```


---

# Anonymous Function 匿名函數

- function
      - 平常寫函數
      - 但為了方便也可以不要寫完整，一次性使用
      - `.` 點點代表前面的變數/資料
- 匿名函數的形式
      - `~ function(x){x + 5}`
      - `~ as.integer(.) + 5`

---

# dplyr 的未竟之業 - across()

```{r message=F, warning=F}
df_marriage %>% slice(1)
df_marriage %>% mutate(across(starts_with("county"), ~str_c(., "-bad"))) %>% slice(1)
```

---

# dplyr 的未竟之業 - across()

```{r message=F, warning=F}
df_marriage %>% mutate(across(matches("marry") & -matches("marry_cp"), ~(./marry_cnt))) %>% slice(1)
df_marriage %>% select(matches("marry_.*_cnt"), -starts_with("county"), matches("marry")) %>% 
  mutate(across(matches("marry") & -matches("marry_cp"), ~(./marry_cnt))) %>% slice(1)
```
 
---

# dplyr 的未竟之業 - across()

```{r message=F, warning=F}
df_marriage <- read_csv("data/Lab04/109Q4_county_marriage.csv") %>%
   mutate(across(where(is.character), ~iconv(.,from = "BIG5", to = "UTF8"))) %>%
  slice(-1) %>% 
  mutate(across(matches("MARRY"), ~as.integer(.))) %>%
  `colnames<-`(str_to_lower(colnames(.)))
```

---

# dplyr 的未竟之業 - rowwise(), c_across()


```{r message=F, warning=F}
df_marriage %>% rowwise() %>% mutate(
    marry_sum = sum(c_across(marry_cp_cnt:marry_f_cnt)),
    marry_mean = mean(c_across(marry_cp_cnt:marry_f_cnt))
 ) %>% ungroup()
```

---

class: inverse, center, middle

# [Lab08](https://p4css.github.io/R4CSS_TA/Lab06_Homework_Visualizing-Date-Time.html)