替代R中的数据帧

Question

What's the right type of object to encapsulate objects of different type in R? Data frames don't seem to be the right type of object for this? Data frames have a number of entries each having the same sub-fields.

What I need is a single object, with different sub-objects some of which might be arrays.

Example:

score<-0.95
confidence_in_score<-0.5
confidence_interval<-c(0,1)
token<-"foobar"
object_to_return<-data.frame(score,confidence_in_score,confidence_interval,token)


  score confidence_in_score confidence_interval token
1  0.05                 0.5                   0   ggg
2  0.05                 0.5                   1   ggg

What I really want is a container where one element will be an array confidence_interval with just two elements.

Motivation: To pass back to the calling program a single object instead of several individual sub-objects.

Answer 1

Try this:

# next 4 lines are from question 
score<-0.95
confidence_in_score<-0.5
confidence_interval<-c(0,1)
token<-"foobar"

list(score = score, 
     confidence_in_score = confidence_in_score, 
     confidence_interval = confidence_interval, 
     token = token)

As an example of returning a list from R itself look at the last line of source code of eigen . It returns a list with two components named values and vectors . (Just type eigen on a line by itself in R to see its source.)

Try ?list from within R for more info and examples.

These links may also be helpful:

https://cran.r-project.org/doc/manuals/r-release/R-intro.html#Lists-and-data-frames

http://www.homogenisation.org/admin/docs/Lists&DataFrames.pdf

Answer 2

The object you look for is a list . The sub-elements of a list can contain any kind of R object. For example:

str(list(mtcars, summary(mtcars), c('bla', 'spam', 'ham'), array(runif(1000))))
List of 4
 $ :'data.frame':   32 obs. of  11 variables:
  ..$ mpg : num [1:32] 21 21 22.8 21.4 18.7 18.1 14.3 24.4 22.8 19.2 ...
  ..$ cyl : num [1:32] 6 6 4 6 8 6 8 4 4 6 ...
  ..$ disp: num [1:32] 160 160 108 258 360 ...
  ..$ hp  : num [1:32] 110 110 93 110 175 105 245 62 95 123 ...
  ..$ drat: num [1:32] 3.9 3.9 3.85 3.08 3.15 2.76 3.21 3.69 3.92 3.92 ...
  ..$ wt  : num [1:32] 2.62 2.88 2.32 3.21 3.44 ...
  ..$ qsec: num [1:32] 16.5 17 18.6 19.4 17 ...
  ..$ vs  : num [1:32] 0 0 1 1 0 1 0 1 1 1 ...
  ..$ am  : num [1:32] 1 1 1 0 0 0 0 0 0 0 ...
  ..$ gear: num [1:32] 4 4 4 3 3 3 3 4 4 4 ...
  ..$ carb: num [1:32] 4 4 1 1 2 1 4 2 2 4 ...
 $ : 'table' chr [1:6, 1:11] "Min.   :10.40  " "1st Qu.:15.43  " "Median :19.20  " "Mean   :20.09  " ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr [1:6] "" "" "" "" ...
  .. ..$ : chr [1:11] "     mpg" "     cyl" "     disp" "      hp" ...
 $ : chr [1:3] "bla" "spam" "ham"
 $ : num [1:1000(1d)] 0.5061 0.0806 0.4081 0.5038 0.8896 ...

The following tutorial I wrote could be of benefit to you.

Answer 3

I would NOT recommend a list. Every set of data can be represented in a relational database, which is much easier to work with. In your case, you have two options:

data = data.frame(score = 0.95,
                  confidence_in_score = 0.5,
                  confidence_interval_lower_limit = 0,
                  confidence_interval_upper_limit = 1,
                  token = "foobar")

Or

interval = data.frame(score = 0.95,
                      confidence_in_score = 0.5,
                      token = "foobar")

limit = data.frame(token = c("foobar", "foobar"),
                   type = c("lower_limit", "upper_limit"),
                   value= c(0, 1) )

You can use dplyr and tidyr to easily manipulate this data. For example,

library(dplyr)
library(tidyr)

data %>%
  select(token, 
         confidence_interval_lower_limit,
         confidence_interval_upper_limit) %>%
  gather(type, value, -token)

or

interval %>%
  spread(type, value) %>%
  full_join(limit)

Two rules of relational data are:

Never repeat information anywhere except for ID columns

No cell should contain more than one piece of information