R4marketing | канал Алексея Селезнёва | Язык R. Страница 12

library(assertr)

mtcars %>%
      verify(has_all_names("mpg", "vs", "am", "wt")) %>%
      verify(nrow(.) > 10) %>%
      verify(mpg > 0) %>%
      insist(within_n_sds(4), mpg) %>%
      assert(in_set(0,1), am, vs) %>%
      assert_rows(num_row_NAs, within_bounds(0,2), everything()) %>%
      assert_rows(col_concat, is_uniq, mpg, am, wt) %>%
      insist_rows(maha_dist, within_n_mads(10), everything()) %>%
      group_by(cyl) %>%
      summarise(avg.mpg=mean(mpg))

 
sessionInfo()

R version 4.2.0 (2022-04-22)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.4 LTS

 
install.packages('rgoogleads')
install.packages('rfacebookstat')

 
streamHistory <- fromJSON("MyData/StreamingHistory0.json", flatten = TRUE)

#Форматируем дату и добовляем время, день недели
mySpotify <- streamHistory %>% 
  as_tibble() %>% 
  mutate_at("endTime", ymd_hm) %>% 
  mutate_at("endTime", format, tz = "Asia/Yekaterinburg") %>%
  mutate_at("endTime", as_datetime) %>%
  mutate(date = floor_date(endTime, "day") %>% as_date, 
         seconds = msPlayed / 1000, 
         minutes = seconds / 60,
         weekday = weekdays(date, abbreviate = TRUE),
         hour = hour(endTime))

mySpotify %>% 
  mutate(weekday = factor(weekday, levels = rev(c("Пн","Вт","Ср","Чт","Пт","Сб","Вс")))) %>%
  group_by(weekday, hour) %>% 
  summarize(minutesListened = sum(minutes)) %>% 
  ggplot(aes(x = hour, weekday, fill = minutesListened)) + 
  geom_tile(color = "white", width = 0.98, height = 0.98) + 
  scale_y_discrete(expand = c(0,0)) +
  scale_x_continuous(breaks = seq(0,24,1), expand = c(0,0)) + 
  scale_fill_gradient("Минут прослушивания", low = "yellow", high = "red") +
  labs(x = "Время дня", y = "", caption = "Данные: Spotify  Визуализация: Юрий Тукачев") + 
  ggtitle("Когда я чаще всего слушал музыку на Spotify за последний год?", 
          "Еженедельная активность прослушивания по часам дня недели") +
  guides(fill = guide_colorbar(title.position = 'top', title.hjust = .5,
                                barwidth = unit(15, 'lines'), barheight = unit(.5, 'lines'))) +
  theme(plot.title.position = "plot", text = element_text(size = 14, family = "Roboto Mono"),
        plot.caption.position = "plot",
        plot.caption = element_text(color = "gray"),
        plot.subtitle = element_text(color = "gray"),
        legend.position = "top", 
        panel.background = element_blank(),
        plot.margin = margin(25, 25, 10, 25),
        axis.ticks = element_blank())

 
library(assertr)
library(magrittr)
library(data.validator)

report <- data_validation_report()

validate(mtcars, name = "Проверяем таблицу mtcars") %>%
  validate_if(drat < 0, description = "Столбец drat содержит только положительные значения") %>%
  validate_cols(in_set(c(0, 2)), vs, am, description = "Значения столбцов vs и am равны только 0 или 2") %>%
  validate_cols(within_n_sds(1), mpg, description = "Столбец mpg в прделах 1го стандартного отклонения") %>%
  validate_rows(num_row_NAs, within_bounds(0, 2), vs, am, mpg, description = "Количество пропущенных значений в строках не превышает 2") %>%
  validate_rows(maha_dist, within_n_mads(10), everything(), description = "Расстояние Махаланобиса не превышает 10 медианных абсолютных отклонений") %>%
  add_results(report)

between <- function(a, b) {
  function(x) { a <= x && x <= b }
}

validate(iris, name = "Проверяем набор данных iris") %>%
  validate_if(Sepal.Length > 0, description = "Значения столбца Sepal length положительные") %>%
  validate_cols(between(0, 4), Sepal.Width, description = "Значения столбца Sepal находятся в диапазоне от 0 до 4") %>%
  add_results(report)

# выводим результат в консоль
print(report)

# Создаём HTML отчёт
save_report(report, success = FALSE)
browseURL("validation_report.html")

 
library(calendR)
library(tidyverse)
# Data
data <- c(55:154)
weekdays_lbls <- c("Пн", "Вт", "Ср", "Чт", "Пт", "Сб", "Вс")
calendR(year = 2022, title = "100 days of war", text.size = 10, 
        day.size = 3, weeknames = weekdays_lbls,
        lty = 0, papersize = "A5", font.family = "Open Sans", 
        days.col = "gray90", 
        title.size = 28, title.col = "red", 
        text.col = "gray90",
        special.days = data,
        special.col = "black",
        orientation = "portrait")
ggsave("100days.png", dpi = 150, width = 7, height = 9)