update

SD/2.2_dataimport/_demo.qmd

@@ -0,0 +1,154 @@
+---
+title: "Lesson 7"
+format: html
+---
+
+
+```{r}
+require(tidyverse)
+
+
+files <- c(
+  "../../data/01-sales.csv",
+  "../../data/02-sales.csv",
+  "../../data/03-sales.csv"
+)
+
+files <- dir("../../data", pattern = "sales.csv", full.names = TRUE)
+
+
+readr::read_csv(files, id = "file") |>
+  mutate(file = basename(file))
+```
+
+
+```{r}
+require(tidyverse)
+library(tidyverse)
+
+files <- dir(
+  "../../data/gapminder",
+  pattern = "^[12][09][0-9][0-9].xlsx$",
+  full.names = TRUE
+)
+
+
+alldf <- tibble::tibble()
+for (file in files) {
+  alldf <- alldf |>
+    bind_rows(
+      readxl::read_xlsx(file) |>
+        mutate(year = parse_number(basename(file)))
+    )
+}
+
+alldf |>
+  ggplot(aes(x = lifeExp, y = gdpPercap)) +
+  geom_point(aes(color = factor(year))) +
+  geom_smooth(method = "lm", se = FALSE) +
+  scale_y_log10(
+    breaks = scales::trans_breaks("log10", function(x) 10^x),
+    labels = scales::trans_format("log10", scales::math_format(10^.x))
+  ) +
+  facet_wrap(~year, ncol = 4, scale = "fixed")
+```
+
+# slope
+
+```{r}
+file <- files[1]
+
+lm(y ~ x, data)
+
+m <- lm(log10(gdpPercap) ~ lifeExp, readxl::read_xlsx(file))
+
+summary(m)
+
+coef(m)[2]
+
+slopes <- c()
+years <- c()
+for (file in files) {
+  m <- lm(log10(gdpPercap) ~ lifeExp, readxl::read_xlsx(file))
+  years <- c(years, parse_number(basename(file)))
+  slopes <- c(slopes, coef(m)[2])
+}
+years
+slopes
+
+
+plot(years, as.numeric(slopes), type = "b")
+```
+
+# purrr
+
+
+
+```{r}
+require(tidyverse)
+df <- tibble(
+  filename = dir(
+    "../../data/gapminder",
+    pattern = "^[12][09][0-9][0-9].xlsx$",
+    full.names = TRUE
+  )
+) |>
+  dplyr::mutate(
+    data = purrr::map(
+      filename,
+      \(x) readxl::read_xlsx(x)
+    )
+  ) |>
+  mutate(year = parse_number(basename(filename))) |>
+  mutate(
+    m = purrr::map(data, \(xxxx) lm(log10(gdpPercap) ~ lifeExp, data = xxxx))
+  ) |>
+  mutate(slope = purrr::map_dbl(m, \(yyyy) coef(yyyy)[2])) |>
+  unnest(data)
+
+pvalue <- summary(df$m[[1]])$coefficients[2, 4]
+rsq <- summary(df$m[[1]])$r.squared
+
+require(tidymodels)
+generics::tidy(df$m[[1]])
+
+df |>
+  ggplot(aes(x = lifeExp, y = gdpPercap)) +
+  geom_point(aes(color = factor(year))) +
+  geom_smooth(method = "lm", se = FALSE) +
+  scale_y_log10(
+    breaks = scales::trans_breaks("log10", function(x) 10^x),
+    labels = scales::trans_format("log10", scales::math_format(10^.x))
+  ) +
+  facet_wrap(~year, ncol = 4, scale = "fixed")
+
+
+df |>
+  ggplot(aes(x = year, y = slope)) +
+  geom_line() +
+  geom_point()
+
+
+df
+df$slope[[1]]
+
+coef(df$m[[1]])[2]
+
+lm(y ~ x, data)
+
+
+df |>
+  unnest(data) |>
+  ggplot(aes(x = lifeExp, y = gdpPercap)) +
+  geom_point(aes(color = factor(year))) +
+  geom_smooth(method = "lm", se = FALSE) +
+  scale_y_log10(
+    breaks = scales::trans_breaks("log10", function(x) 10^x),
+    labels = scales::trans_format("log10", scales::math_format(10^.x))
+  ) +
+  facet_wrap(~year, ncol = 4, scale = "fixed")
+```
+
+
+
+

SD/2.2_dataimport/index.qmd

@@ -74,7 +74,7 @@ read_csv("../../data/students.csv") |>
 ## 列名不要有空格
 
 ```{r}
-students |> 
+students |>
   rename(
     student_id = `Student ID`,
     full_name = `Full Name`
@@ -158,7 +158,7 @@ x,y,z
 1,2,3"
 
 read_csv(
-  another_csv, 
+  another_csv,
   col_types = cols(.default = col_character())
 )
 read_csv(
@@ -197,9 +197,11 @@ annoying <- tibble(
 ```{r}
 #| message: false
 
-sales_files <- c("../../data/01-sales.csv",
+sales_files <- c(
+  "../../data/01-sales.csv",
   "../../data/02-sales.csv",
-  "../../data/03-sales.csv")
+  "../../data/03-sales.csv"
+)
 read_csv(sales_files, id = "file")
 ```
 
@@ -223,8 +225,11 @@ read_csv(sales_files, id = "file")
 ## 批量读取
 
 ```{r}
-sales_files <- list.files("../../data",
-  pattern = "sales\\.csv$", full.names = TRUE)
+sales_files <- list.files(
+  "../../data",
+  pattern = "sales\\.csv$",
+  full.names = TRUE
+)
 sales_files
 ```
 
@@ -260,7 +265,7 @@ if (FALSE) {
 ```{r}
 if (FALSE) {
   conn <- cctdb::get_dbconn("nationalairquality")
-  metadf <- tbl(conn, "metadf") |> 
+  metadf <- tbl(conn, "metadf") |>
     head(100) |>
     collect()
   DBI::dbDisconnect(conn)
@@ -270,8 +275,8 @@ metadf <- readRDS(file = "../../data/metadfdemo.RDS")
 lang <- "cn"
 metadf |>
   ggplot(aes(lon, lat)) +
-geom_point(aes(fill = Area)) +
-dwfun::theme_sci()
+  geom_point(aes(fill = Area)) +
+  dwfun::theme_sci()
 ```
 
 
@@ -283,8 +288,13 @@ metadf <- readxl::read_xlsx("../../data/airquality.xlsx")
 dir.create("../../data/metacity2/")
 metadf |>
   nest(sitedf = -site) |>
-  mutate(flag = purrr::map2(site, sitedf,
-    ~ writexl::write_xlsx(.y, paste0("../../data/metacity2/", .x, ".xlsx"))))
+  mutate(
+    flag = purrr::map2(
+      site,
+      sitedf,
+      ~ writexl::write_xlsx(.y, paste0("../../data/metacity2/", .x, ".xlsx"))
+    )
+  )
 ```
 
 
@@ -296,7 +306,7 @@ metadf |>
 if (FALSE) {
   require(tidyverse)
   conn <- cctdb::get_dbconn("nationalairquality")
-  metadf <- tbl(conn, "metadf") |> 
+  metadf <- tbl(conn, "metadf") |>
     collect()
   DBI::dbDisconnect(conn)
   metanestdf <- metadf |>
@@ -305,10 +315,16 @@ if (FALSE) {
   writexl::write_xlsx(metanestdf$citydf, path = "../../data/meta_city.xlsx")
   dir.create("../../data/metacity/")
   metanestdf |>
-    mutate(flag = purrr::map2(Area, citydf,
-      ~ writexl::write_xlsx(.y,
-        path = paste0("../../data/metacity/", .x, ".xlsx")
-      )))
+    mutate(
+      flag = purrr::map2(
+        Area,
+        citydf,
+        ~ writexl::write_xlsx(
+          .y,
+          path = paste0("../../data/metacity/", .x, ".xlsx")
+        )
+      )
+    )
 }
 ```