Question

df <- read.csv ('https://raw.githubusercontent.com/ulklc/covid19-timeseries/master/countryReport/raw/rawReport.csv',
                stringsAsFactors = FALSE)

我找到了该代码，该代码找到了报告死亡人数最多并按地区恢复的国家。

    yesterday <- function() Sys.Date() - 1L
    yesterday()
    # [1] "if it doesn't work yesterday()-1  do it" 

    library(tidyverse)
    death_df <- df %>%
    filter(as.Date(day) == yesterday()) %>%
    group_by(region) %>%
    filter(death == max(death)) %>%
    select(Date = day,
         countryName,
         region,
         death,
         recovered)

     recovered_df <- df %>%
     filter(as.Date(day) == yesterday()) %>%
    group_by(region) %>%
    filter(recovered == max(recovered)) %>%
    select(Date = day,
         countryName,
         region,
         death,
         recovered)

    full_df <- bind_rows(death_df, recovered_df)

但是，我需要找到报告死亡人数最多并已康复的国家。

这是我想要的输出：

date            countryName       death        recovered
2020/05/06       united State   **19580**        500
2020/05/06       İran             11500        **98567**

请注意，这些值不是真实的。

数据集每天更新。但是，它可能没有更新1 -2天。让我们注意一下。

Answer 1

这是一种使用dplyr计算每天所需信息的方法。

library(dplyr)
result <- df %>% group_by(day) %>% 
  filter(death == max(death) | recovered == max(recovered)) %>%
  mutate(death = case_when(death == max(death) ~ paste0("**",death[death == max(death)],"**"),
                           TRUE ~ as.character(death)),
         recovered = case_when(recovered == max(recovered) ~ paste0("**",recovered[recovered == max(recovered)],"**"),
                           TRUE ~ as.character(recovered)))
result %>%
  filter(day == "2020/04/06")
# A tibble: 2 x 9
# Groups:   day [1]
  day        countryCode countryName region   lat   lon confirmed recovered death    
  <chr>      <chr>       <chr>       <chr>  <dbl> <dbl>     <int> <chr>     <chr>    
1 2020/04/06 CN          China       Asia    35   105       81708 **77029** 3331     
2 2020/04/06 IT          Italy       Europe  42.8  12.8    132547 22837     **16523**

Answer 2

下面的代码将选择记录中的最大每日death和最大每日recovered以及数据中的最大date。

## call the dplyr library
library(dplyr)
## read the data into R
df <- read.csv ('https://raw.githubusercontent.com/ulklc/covid19-timeseries/master/countryReport/raw/rawReport.csv', stringsAsFactors = FALSE)
## determine the max date contained within the data
max.date <- df[which.max(as.Date(df$day)),"day"]
## copy the data to preserve original
df1 <- df 
## filter the data to only entries from the max day
df1 <- filter(df1, as.Date(date, "%Y/%m/%d") == as.Date(max.date))
## determine the entry with the most deaths
max.deaths <- df1[which.max(df1$death),]
## format the number of deaths as given in the example
max.deaths$death <- paste0("**",max.deaths$death,"**")
## determine the entry with the most recovered
max.recovered <- df1[which.max(df1$recovered),]
## format the number recovered to match the format of the example
max.recovered$recovered <- paste0("**",max.recovered$recovered,"**")
## create a data frame containing our max death and max recovered entries
max.records <- rbind(max.deaths, max.recovered)
## attach a column with the max date which corresponds to the date of the entries selected
max.records$date <- max.date
## organize the data as shown in the example
max.records <- select(max.records, c("day","countryName","death","recovered"))

并且此代码将计算每个国家的死亡总数（或总数）为totalDeaths，并将死亡总数恢复为totalRecovered。然后，它将返回数据中包含最大totalDeath和最大totalRecovered以及最大日期的记录。

## call the dplyr library
library(dplyr)
## read the data into R
df <- read.csv ('https://raw.githubusercontent.com/ulklc/covid19-timeseries/master/countryReport/raw/rawReport.csv', stringsAsFactors = FALSE)
## determine the max date contained within the data
max.date <- df[which.max(as.Date(df$day)),"day"]
## copy the data to preserve the original
df1 <- df 
## group the data by countries
df1 <- group_by(df1, countryName)
## sum the death and recovered of each country
df1 <- summarise(df1, totalDeaths = sum(death), totalRecovered = sum(recovered))
## ungroup your data to avoid errors
df1 <- ungroup(df1)
## determine country with most total deaths reported
max.deaths <- df1[which.max(df1$totalDeaths),]
## format death numbers to match example
max.deaths$totalDeaths <- paste0("**",max.deaths$totalDeaths,"**")
## determine country with most total recovered reported
max.recovered <- df1[which.max(df1$totalRecovered),]
## format recovered numbers to match example
max.recovered$totalRecovered <- paste0("**",max.recovered$totalRecovered,"**")
## create a data frame containing our max entries
max.records <- rbind(max.deaths, max.recovered)
## attach a column with the max date which corresponds to the most current date the data reports
max.records$date <- max.date
## organize the data as shown in the example
max.records <- select(max.records, c("day","countryName","death","recovered"))

注意：两种方法都依赖于dplyr R包。可以通过在R或RStudio中运行dplyr来安装install.packages(dplyr)。

我希望这会有所帮助！

根据最大日期和/或最大值返回数据帧条目（COVID-19数据示例）

2 个答案: