Rpackage【dplyr】

原文地址：https://zhuanlan.zhihu.com/p/25784721

使用举例

# 安装并载入dplyr包
install.packages("dplyr")
library(dplyr)
# 创建两个数据集order和price
> set.seed(1)
> ordernum <- sample(1:20, 5)
> itemprice <- sample(100:500, 5)
> order <- data.frame(itemid=1:5, ordernum, class=rep(1:2,2:3))
> order
  itemid ordernum class
1      1        6     1
2      2        8     1
3      3       11     2
4      4       16     2
5      5        4     2
> price <- data.frame(itemid = 2:6, itemprice)
> price
  itemid itemprice
1      2       460
2      3       477
3      4       363
4      5       350
5      6       124

# filter()--返回满足条件的观测

> # filter()
> filter(order, ordernum >= 10)
  itemid ordernum class
1      3       11     2
2      4       16     2
> filter(price, itemid %in% c(3,5))
  itemid itemprice
1      3       477
2      5       350

# select()--选择所需变量

> # select()
> select(order, itemid, ordernum)
  itemid ordernum
1      1        6
2      2        8
3      3       11
4      4       16
5      5        4
> select(price, -itemid)
  itemprice
1       460
2       477
3       363
4       350
5       124

# arrange()--对观测进行排序

> # arrange()
> arrange(order, ordernum)
  itemid ordernum class
1      5        4     2
2      1        6     1
3      2        8     1
4      3       11     2
5      4       16     2
> arrange(price, -itemprice)
  itemid itemprice
1      3       477
2      2       460
3      4       363
4      5       350
5      6       124

# mutate()--添加新变量

> # mutate()
> mutate(price, cost=0.7*itemprice, profit=itemprice-cost)
  itemid itemprice  cost profit
1      2       460 322.0  138.0
2      3       477 333.9  143.1
3      4       363 254.1  108.9
4      5       350 245.0  105.0
5      6       124  86.8   37.2

# rename()--重命名变量

> # rename()
> rename(price, id=itemid)
  id itemprice
1  2       460
2  3       477
3  4       363
4  5       350
5  6       124

# summarise()--数据汇总

> # summarise()
> mygroup <- group_by(order, class)
> summarise(mygroup, count=n(), total=sum(ordernum))
# A tibble: 2 × 3
  class count total
  <int> <int> <int>
1     1     2    14
2     2     3    31

# %>%--管道操作(将上个输出作为下个输入)

> # %>%
> order %>% group_by(class) %>%
+   summarise(count=n(), total=sum(ordernum))
# A tibble: 2 × 3
  class count total
  <int> <int> <int>
1     1     2    14
2     2     3    31

# join()--连接数据集

> # join()
> inner_join(order, price, by="itemid")
  itemid ordernum class itemprice
1      2        8     1       460
2      3       11     2       477
3      4       16     2       363
4      5        4     2       350
> left_join(order, price, by="itemid")
  itemid ordernum class itemprice
1      1        6     1        NA
2      2        8     1       460
3      3       11     2       477
4      4       16     2       363
5      5        4     2       350
> right_join(order, price, by="itemid")
  itemid ordernum class itemprice
1      2        8     1       460
2      3       11     2       477
3      4       16     2       363
4      5        4     2       350
5      6       NA    NA       124
> full_join(order, price, by="itemid")
  itemid ordernum class itemprice
1      1        6     1        NA
2      2        8     1       460
3      3       11     2       477
4      4       16     2       363
5      5        4     2       350
6      6       NA    NA       124
> semi_join(order, price, by="itemid")
  itemid ordernum class
1      2        8     1
2      3       11     2
3      4       16     2
4      5        4     2
> anti_join(order, price, by="itemid")
  itemid ordernum class
1      1        6     1

# src_mysql()--连接MySQL数据库

> # src_mysql()
> library(RMySQL)
> library(DBI)
> src_mysql("mydb",user="root",
password="****") %>% tbl(from = "sample")
Source:   query [?? x 2]
Database: mysql 5.7.17 [root@localhost:/mydb]

  itemid class
   <int> <int>
1      1     1
2      2     1
3      3     1
4      4     2
5      5     2