summary.sperroresterror
calculates mean, standard deviation,
median etc. of the calculated error measures at the specified level
(overall, repetition, or fold).
summary.sperrorestreperror
does the same with the pooled error,
at the overall or repetition level.
1 2 3 
object 

level 
Level at which errors are summarized: 0: overall; 1: repetition; 2: fold 
pooled 
If 
na.rm 
Remove 
... 
additional arguments (currently ignored) 
Let's use an example to explain the error.rep
argument.
E.g., assume we are using 100repeated 10fold crossvalidation.
If error.rep = TRUE
(default), the mean and standard deviation calculated
when summarizing at level = 0
are calculated across the error estimates obtained for
each of the 100*10 = 1000
folds.
If error.rep = FALSE
, mean and standard deviation are calculated across
the 100
repetitions, using the weighted average of the foldlevel
errors to calculate an error value for the entire sample.
This will essentially not affect the mean value but of course the
standard deviation of the error. error.rep = FALSE
is not recommended,
it is mainly for testing purposes; when the test sets are small
(as in leaveoneout crossvalidation, in the extreme case),
consider running sperrorest()
with error.rep = TRUE
and
examine only the error.rep
component of its result.
Depending on the level of aggregation, a list
or
data.frame
with mean, and at level 0 also standard deviation,
median and IQR of the error measures.
Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.
Please suggest features or report bugs with the GitHub issue tracker.
All documentation is copyright its authors; we didn't write any of that.