3.3.4 Aggregate Columns and Rows

OREdplyr functions for aggregating columns and rows.

Table 3-5 Aggregating Columns and Rows

Function	Description
count count_	Counts rows by group; similar to tally, but it does the group_by for you.
summarise summarise_	Summarizes columns by using aggregate functions. When an ore.frame object is grouped, the aggregate function is applied group-wise. The resulting ore.frame drops one grouping of the input ore.frame.
tally	Tallies rows by group; a convenient wrapper for summarise that either calls n or sum(n) depending on whether you’re tallying for the first time or re-tallying.

Example 3-74 Aggregating Columns

The following examples use the ore.frame object MTCARS that is created by using the ore.push function on the mtcars data.frame object. They exemplify the use of the aggregation functions count, summarize, and tally. They also use the OREdplyr functions arrange and group_by.

MTCARS <- ore.push(mtcars)
arrange(tally(group_by(MTCARS, cyl)), cyl)
tally(group_by(MTCARS, cyl), sort = TRUE)

# Multiple tallys progressively roll up the groups
cyl_by_gear <- tally(group_by(MTCARS, cyl, gear), sort = TRUE)
tally(cyl_by_gear, sort = TRUE)
tally(tally(cyl_by_gear))

cyl_by_gear <- tally(group_by(MTCARS, cyl, gear), wt = hp, sort = TRUE)
tally(cyl_by_gear, sort = TRUE)
tally(tally(cyl_by_gear))

cyl_by_gear <- count(MTCARS, cyl, gear, wt = hp + mpg, sort = TRUE)
tally(cyl_by_gear, sort = TRUE)
tally(tally(cyl_by_gear))

# Load the magrittr library to use the forward-pipe operator %>%
library(magrittr)
MTCARS %>% group_by(cyl) %>% tally(sort = TRUE)

# count is more succinct and also does the grouping
MTCARS %>% count(cyl) %>% arrange(cyl)
MTCARS %>% count(cyl, wt = hp) %>% arrange(cyl)
MTCARS %>% count_("cyl", wt = hp, sort = TRUE)

Listing for This Example

R> MTCARS <- ore.push(mtcars)
R> arrange(tally(group_by(MTCARS, cyl)), cyl)
  cyl  n
1   4 11
2   6  7
3   8 14
R> tally(group_by(MTCARS, cyl), sort = TRUE)
  cyl  n
1   8 14
2   4 11
3   6  7
R> 
R> # Multiple tallys progressively roll up the groups
R> cyl_by_gear <- tally(group_by(MTCARS, cyl, gear), sort = TRUE)
R> tally(cyl_by_gear, sort = TRUE)
Using n as weighting variable
  cyl  n
1   8 14
2   4 11
3   6  7
R> tally(tally(cyl_by_gear))
Using n as weighting variable
Using n as weighting variable
 n 
32 
R>
R> cyl_by_gear <- tally(group_by(MTCARS, cyl, gear), wt = hp, sort = TRUE)
R> tally(cyl_by_gear, sort = TRUE)
Using n as weighting variable
  cyl    n
1   8 2929
2   4  909
3   6  856
R> tally(tally(cyl_by_gear))
Using n as weighting variable
Using n as weighting variable
   n 
4694 
R> 
R> cyl_by_gear <- count(MTCARS, cyl, gear, wt = hp + mpg, sort = TRUE)
R> tally(cyl_by_gear, sort = TRUE)
Using n as weighting variable
  cyl      n
1   8 3140.4
2   4 1202.3
3   6  994.2
R> tally(tally(cyl_by_gear))
Using n as weighting variable
Using n as weighting variable
     n 
5336.9 
R> 
R> # Load the magrittr library to use the forward-pipe operator %>%
R> library(magrittr)
R> MTCARS %>% group_by(cyl) %>% tally(sort = TRUE)
  cyl  n
1   8 14
2   4 11
3   6  7
R> 
R> # count is more succinct and also does the grouping
R> MTCARS %>% count(cyl) %>% arrange(cyl)
  cyl  n
1   4 11
2   6  7
3   8 14
R> MTCARS %>% count(cyl, wt = hp) %>% arrange(cyl)
  cyl    n
1   4  909
2   6  856
3   8 2929
R> MTCARS %>% count_("cyl", wt = hp, sort = TRUE)
  cyl    n
1   8 2929
2   4  909
3   6  856