render compile

2026-05-21 01:33:30 +08:00
parent 2c2eaa0d0d
commit 204a12a58f
2379 changed files with 142545 additions and 814 deletions
@@ -0,0 +1,115 @@
+---
+title: "Lesson 8"
+format: html
+---
+
+
+```{r}
+require(tidyverse)
+library(nycflights13)
+
+flights |>
+  select(3:6) |>
+  head(3) |>
+  rename_all(~ gsub("_", "", .))
+
+
+flights |>
+  select(3:6) |>
+  head(3) |>
+  rename_with(toupper, .cols = 2:4)
+
+# 每月10号-15号，dep_delay > 100 的航班
+
+# 每月哪个出发地origin的 dep_delay总时长最长
+
+flights |>
+  group_by(month, origin) |>
+  summarize(n = n(), total_dep_delay = mean(dep_delay, na.rm = TRUE)) |>
+  slice_max(total_dep_delay)
+
+# arrange(month, origin, desc(total_dep_delay))
+
+# 每月随机抽取一天，随机抽取三个航班
+
+slice_sample(n = 1)
+
+flights |>
+  tidyr::nest(ymddf = -c(year, month, day)) |>
+  group_by(year, month) |>
+  slice_sample(n = 5) |>
+  unnest(ymddf) |>
+  group_by(year, month, day) |>
+  slice_sample(n = 3)
+
+
+flights |>
+  tidyr::nest(ymddf = -c(year, month, day)) |>
+  group_by(year, month) |>
+  slice_sample(n = 1) |>
+  mutate(
+    ymddf = purrr::map(ymddf, \(x) {
+      x |>
+        slice_sample(n = 3)
+    })
+  )
+
+# 每月 每个出发地 周末的平均dep_delay 与 工作日的平均dep_delay 差值最大的3个航班
+
+flights |>
+  mutate(date = ymd(paste(year, month, day))) |>
+  mutate(weekday = wday(date)) |>
+  mutate(isworkday = if_else(between(weekday, 2, 6), "Yes", "No")) |>
+  group_by(year, month, origin, flight, isworkday) |>
+  summarize(mean_delay = mean(dep_delay, na.rm = TRUE)) |>
+  tidyr::nest(diffdelaydf = c(isworkday, mean_delay)) |>
+  filter(
+    purrr::map(diffdelaydf, \(x) {
+      nrow(x)
+    }) >
+      1
+  ) |>
+  mutate(
+    diffdelay = purrr::map_dbl(diffdelaydf, \(x) {
+      x |>
+        arrange(isworkday) |>
+        pull(mean_delay) |>
+        diff()
+    })
+  ) |>
+  group_by(year, month, origin) |>
+  slice_max(diffdelay, n = 3)
+
+
+wday(today())
+
+weekday()
+
+
+slice_sample(n = 1)
+
+
+flights |>
+  group_by(month) |>
+  slice_sample(n = 1)
+```
+
+
+
+```{r}
+p1 <- flights |>
+  group_by(year, month) |>
+  summarize(n = n()) |>
+  ggplot(aes(month, n)) +
+  geom_point(shape = 21, size = 6, color = "black", fill = "red") +
+  geom_line()
+
+ggsave("./a.pdf")
+ggsave("./a.png")
+
+require(patchwork)
+
+p1 / p1
+```
+
+
@@ -0,0 +1 @@
+../../_extensions
@@ -0,0 +1,309 @@
+---
+title: "Data Transform"
+format:
+  dwsd-revealjs:
+    logo: _extensions/drwater/dwsd/inst/ucaslogo.png
+---
+
+```{r}
+#| echo: false
+source("../../coding/_common.R")
+library(nycflights13)
+library(tidyverse)
+```
+
+
+## `tidyverse`风格数据分析总体流程
+
+![](../../image/data-science/transform.png) 
+
+
+## [dplyr cheatsheet](../../image/cheatsheet/data-transformation.pdf)
+
+
+```{r}
+#| echo: false
+dwfun::ggsavep("../../image/cheatsheet/data-transformation.svg", loadit = TRUE)
+```
+
+
+
+## 查看数据
+
+```{r}
+flights
+```
+
+
+## 选择列
+
+```{r}
+#| results: false
+
+flights |>
+  select(year, month, day)
+```
+
+## 选择列
+
+```{r}
+#| results: false
+
+flights |>
+  select(year:day)
+```
+
+
+## 选择列
+
+```{r}
+flights |>
+  select(3:5)
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(!year:day)
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(-(year:day))
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(where(is.character))
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(!where(is.character)) |>
+  select(contains("_"))
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(tail_num = tailnum)
+```
+
+## 选择列
+
+```{r}
+flights |>
+  select(air_time, everything())
+```
+
+
+## 重命名
+
+```{r}
+flights |>
+  rename(tail_num = tailnum)
+```
+
+## 重命名
+
+```{r}
+flights |>
+  rename(年份 = 1) |>
+  rename(月份 = 2)
+```
+
+## 重命名
+
+```{r}
+flights |> select(1:4) |> head(n = 3)
+# 重命名
+flights |> select(1:4) |> head(n = 3) |> rename_all(~ c("c1", "c2", "c3", "c4"))
+```
+
+## 重命名
+
+```{r}
+flights |> select(1:4) |> head(n = 3)
+# 重命名
+flights |> select(1:4) |> head(n = 3) |> rename_all(toupper)
+```
+
+
+## 重命名
+
+```{r}
+flights |> select(1:4) |> head(n = 3)
+# 重命名
+flights |>
+  select(1:4) |>
+  head(n = 3) |>
+  rename_all(~ paste0(toupper(.), "_NEW"))
+```
+
+
+## 练习
+
+将含有下划线的列名中的下划线去掉。
+
+
+```{r}
+flights |> select(1:4) |> head(n = 3)
+```
+
+
+## 练习
+
+将`airqualitydf`中列名的单位信息去除（前5列）。
+
+
+```{r}
+airqualitydf <- readxl::read_xlsx("../../data/airquality.xlsx", sheet = 2)
+airqualitydf |> select(1:5)
+```
+
+
+
+## `filter`
+
+```{r}
+flights |>
+  filter(dep_delay > 120)
+```
+
+## filter 练习
+
+Flights that departed on January 1.
+
+```{r}
+#| echo: false
+flights |>
+  filter(month == 1 & day == 1)
+```
+
+## filter 练习
+
+
+Select flights that departed in January or February
+
+```{r}
+#| echo: false
+flights |>
+  filter(month %in% c(1, 2))
+```
+
+## filter 练习
+
+```{r}
+jan1 <- flights |>
+  filter(month == 1 & day == 1)
+```
+
+## filter
+
+```{r}
+#| error: true
+#| eval: false
+
+flights |>
+  filter(month = 1)
+```
+
+## filter
+
+```{r}
+flights |>
+  filter(month == 1 | 2)
+```
+
+## 排序
+
+```{r}
+flights |>
+  arrange(year, month, day, dep_time)
+```
+
+## 排序
+
+```{r}
+flights |>
+  arrange(desc(dep_delay))
+```
+
+## slice
+
+```{r}
+flights |> head(n = 5)
+flights |> slice(1:5)
+```
+
+## slice
+
+```{r}
+flights |>
+  slice_max(dep_delay, n = 5)
+```
+
+## slice
+
+```{r}
+flights |>
+  slice_min(dep_delay, prop = 0.005)
+```
+
+## 排序练习
+
+根据`origin`、`dest`、`air_time`倒序排序。
+
+```{r}
+#| echo: false
+flights |>
+  arrange(origin, dest, desc(air_time)) |>
+  select(origin, dest, air_time, everything())
+```
+
+
+## 去重
+
+```{r}
+# Remove duplicate rows, if any
+flights |>
+  distinct()
+```
+
+## 去重
+
+```{r}
+# Find all unique origin and destination pairs
+flights |>
+  distinct(origin, dest)
+```
+
+## 去重
+
+```{r}
+flights |>
+  distinct(origin, dest, .keep_all = TRUE)
+```
+
+
+
+## 欢迎讨论！{.center}
+
+```{r}
+#| results: 'asis'
+rmdify::slideend(
+  wechat = FALSE,
+  type = "public",
+  tel = FALSE,
+  thislink = "../"
+)
+```
+