Changelog
Source:NEWS.md
rsample (development version)
The new
inner_split()
function and its methods for various resamples is for usage in tune to create a inner resample of the analysis set to fit the preprocessor and model on one part and the post-processor on the other part (#483, #488, #489).Bootstrap intervals via
int_pctl()
,int_t()
, andint_bca()
now allow for more flexible grouping (#465).Errors and warnings are now styled via cli (#499, #502). Largely done by @PriKalra (#523, #526, #528, #530, #531, #532), @Dpananos (#516, #517, #529), and @JamesHWade (#518) as part of the tidyverse dev day.
rolling_origin()
is now superseded bysliding_window()
,sliding_index()
, andsliding_period()
which provide more flexibility and control (@nmercadeb, #524).vfold_cv()
andclustering_cv()
now error on implicit leave-one-out cross-validation (@seb09, #527).
Bug fixes
vfold_cv()
now utilizes thebreaks
argument correctly for repeated cross-validation (@ZWael, #471).Grouped resampling functions now work with an explicit
strata = NULL
instead of strata being either a name or missing (#485).
Breaking changes
The class of grouped MC splits is now
group_mc_split
instead ofgrouped_mc_split
, aligning it with the other grouped splits (#478).The
rsplit
objects of anapparent()
split now have the correct class inheritance structure. The order is nowapparent_split
and thenrsplit
rather than the other way around (#477).
Documentation improvements
Improved documentation and formatting: function names are now more easily identifiable through either
()
at the end or being links to the function documentation (@brshallo , #521).Fixed example for
nested_cv()
(@seb09, #520).Formatting improvement: package names are now not in backticks anymore (@agmurray, #525).
Improved documentation for
initial_split()
and friends (@laurabrianna, #519).Removed trailing space in printing of
mc_cv()
objects (@ccani007, #464).
rsample 1.2.1
CRAN release: 2024-03-25
nested_cv()
no longer errors ifoutside
is a long call (#459, #461).The
validation_set
class now has its ownpretty()
method (#456).
rsample 1.2.0
CRAN release: 2023-08-23
The new
initial_validation_split()
, along with variantsinitial_validation_time_split()
andgroup_initial_validation_split()
, generates a three-way split of the data into training, validation, and test sets. With the newvalidation_set()
, this can be turned into anrset
object for tuning (#403, #446).validation_split()
,validation_time_split()
, andgroup_validation_split()
have been soft-deprecated in favor of the new functions implementing a 3-way split (initial_validation_split()
,initial_validation_time_split()
, andgroup_initial_validation_split()
) (#449).Functions which don’t use the ellipsis
...
now enforce empty dots (#429).make_splits()
gained an example in the documentation (@AngelFelizR, #432).training()
,testing()
,analysis()
, andassessment()
are now S3 generics with methods forrsplit
objects. Previously they manually required the input to be anrsplit
object (#384).The
int_*()
functions are now S3 generics and have corresponding methods for classbootstraps
(#435).The underlying mechanics of data splitting were changed so that
Surv
objects maintain their class. This change affects the row names of the resulting objects; they are reindexed from one instead of being a subset of the original row names (#443).
rsample 1.1.1
CRAN release: 2022-12-07
All grouped resampling functions (
group_vfold_cv()
,group_mc_cv()
,group_initial_split()
andgroup_validation_split()
, andgroup_bootstraps()
) now support stratification. Strata must be constant within each group (@mikemahoney218, #317, #360, #363, #364, #365).Added a new function,
clustering_cv()
, for blocked cross-validation in various predictor spaces. This is a very flexible function, taking arguments to bothdistance_function
andcluster_function
, allowing it to be used for spatial clustering as well as potentially phylogenetic and other forms of clustering (@mikemahoney218, #351).bootstraps()
andgroup_bootstraps()
now warn if resampling returns any empty assessment sets. Previously,bootstraps()
was silent whilegroup_bootstraps()
errored (@mikemahoney218, #356, #357).The assessment set of
validation_time_split()
now also contains the lagged observations (#376).The new helper
get_rsplit()
lets you conveniently access thersplit
objects inside anrset
objects (@mikemahoney218, #399).The result of
initial_time_split()
now has its own subclass"initial_time_split"
, in addition to existing classes (#397).The dependency on the ellipsis package has been removed (#393).
Removed an overly strict test in preparation for dplyr 1.1.0 (#380).
rsample 1.1.0
CRAN release: 2022-08-08
rset objects now include all parameters used to create them as attributes (#329).
Objects returned by sliding functions now have an
index
attribute, where appropriate, containing the column name used as an index (#329).Objects returned by
permutations()
now have apermutes
attribute containing the column name used for permutation (#329).Added
breaks
andpool
as attributes to all functions which support stratification (#329).Changed the “strata” attribute on rset objects so that it now is either a character vector identifying the column used to stratify the data, and is not present (set to
NULL
) if stratification was not used. (#329)Added a new function,
reshuffle_rset()
, which takes anrset
object and generates a new version of it using the same arguments but the current random seed. (#79, #329)Added arguments to control how
group_vfold_cv()
combines groups. Usebalance = "groups"
to assign (roughly) the same number of groups to each fold, orbalance = "observations"
to assign (roughly) the same number of observations to each fold.Added a
repeats
argument togroup_vfold_cv()
(#330).Added new functions for grouped resampling:
group_mc_cv()
(#313),group_initial_split()
andgroup_validation_split()
(#315), andgroup_bootstraps()
(#316).Added a new function,
reverse_splits()
, to swap analysis and assessment splits (#319, #284).Improved the error thrown when calling
assessment()
on aperm_split
object created bypermutations()
(#321, #322).
rsample 1.0.0
CRAN release: 2022-06-24
Fixed how
nested_cv()
handles call objects so variables in the environment can be used when specifying resampling schemes (#81).Updated to testthat 3e (#280) and added better checking for
vfold_cv()
(#293).Finally removed the
gather()
method forrset
objects. Usetidyr::pivot_longer()
instead (#280).Changed
initial_split()
to avoid calling tidyselect twice onstrata
(#296). This fix stopsinitial_split()
from generating messages like:
Note: Using an external vector in selections is ambiguous.
i Use `all_of(strata)` instead of `strata` to silence this message.
i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
- Added better printing methods for initial split objects.
rsample 0.1.1
CRAN release: 2021-11-08
Updated documentation on stratified sampling (#245).
Changed
make_splits()
to an S3 generic, with the original functionality a method forlist
and a new method for dataframes that allows users to create a split from existing analysis & assessment sets (@LiamBlake, #246).Added
validation_time_split()
for a single validation sample taking the first samples for training (@mine-cetinkaya-rundel, #256).Escalated the deprecation of the
gather()
method forrset
objects to a hard deprecation. Usetidyr::pivot_longer()
instead (#257).Changed resample “fingerprint” to hash the indices only rather than the entire resample result (including the data object). This is much faster and will still ensure the same resample for the same original data object (#259).
rsample 0.1.0
CRAN release: 2021-05-08
Fixed how
mc_cv()
,initial_split()
, andvalidation_split()
use theprop
argument to first compute the assessment indices, rather than the analysis indices. This is a minor but breaking change in some situations; the previous implementation could cause an inconsistency in the sizes of the generated analysis and assessment sets when compared to howprop
is documented to function (#217, @issactoast).Fixed problem with creation of
apparent()
(#223) andcaret2rsample()
(#232) resamples.Re-licensed package from GPL-2 to MIT. See consent from copyright holders here.
Attempts to stratify on a
Surv
object now error more informatively (#230).Exposed
pool
argument frommake_strata()
in user-facing resampling functions (#229).Deprecated the
gather()
method forrset
objects in favor oftidyr::pivot_longer()
(#233).Fixed bug in
make_strata()
for numeric variables withNA
values (@brian-j-smith, #236).
rsample 0.0.9
CRAN release: 2021-02-17
New
rset_reconstruct()
, a developer tool to ease creation of new rset subclasses (#210).Added
permutations()
, a function for creating permutation resamples by performing column-wise shuffling (@mattwarkentin, #198).Fixed an issue where empty assessment sets couldn’t be created by
make_splits()
(#188).rset
objects now contain a “fingerprint” attribute that can be used to check to see if the same object uses the same resamples.The
reg_intervals()
function is a convenience function forlm()
,glm()
,survreg()
, andcoxph()
models (#206).A few internal functions were exported so that rsample-adjacent packages can use the same underlying code.
Changed the inheritance structure for
rsplit
objects from specific to general and simplified the methods for thecomplement()
generic (#216).
rsample 0.0.8
CRAN release: 2020-09-23
New
manual_rset()
for constructing rset objects manually from custom rsplits (tidymodels/tune#273).Three new time based resampling functions have been added:
sliding_window()
,sliding_index()
, andsliding_period()
, which have more flexibility than the pre-existingrolling_origin()
.Correct
alpha
parameter handling for bootstrap CI functions (#179, #184).
rsample 0.0.7
CRAN release: 2020-06-04
Lower threshold for pooling strata to 10% (from 15%) (#149).
The
print()
methods forrsplit
andval_split
objects were adjusted to show"<Analysis/Assess/Total>"
and<Training/Validation/Total>
, respectively.The
drinks
,attrition
, andtwo_class_dat
data sets were removed. They are in the modeldata package.Compatability with dplyr 1.0.0.
rsample 0.0.6
CRAN release: 2020-03-31
Added
validation_set()
for making a single resample.Correct the tidy method for bootstraps (#115).
Changes for upcoming `tibble release.
Exported constructors for
rset
andsplit
objects (#40)initial_time_split()
androlling_origin()
now have alag
parameter that ensures that previous data are available so that lagged variables can be calculated. (#135, #136)
rsample 0.0.5
CRAN release: 2019-07-12
- Added three functions to compute different bootstrap confidence intervals.
- A new function (
add_resample_id()
) augments a data frame with columns for the resampling identifier. - Updated
initial_split()
,mc_cv()
,vfold_cv()
,bootstraps()
, andgroup_vfold_cv()
to use tidyselect on the stratification variable. - Updated
initial_split()
,mc_cv()
,vfold_cv()
,bootstraps()
with newbreaks
parameter that specifies the number of bins to stratify by for a numeric stratification variable.
rsample 0.0.4
CRAN release: 2019-01-07
Small maintenance release.
Minor improvements and fixes
-
fill()
was removed per the deprecation warning. - Small changes were made for the new version of tibble.
rsample 0.0.3
CRAN release: 2018-11-20
New features
- Added function
initial_time_split()
for ordered initial sampling appropriate for time series data.
Minor improvements and fixes
fill()
has been renamedpopulate()
to avoid a conflict withtidyr::fill()
.Changed the R version requirement to be R >= 3.1 instead of 3.3.3.
The recipes-related
prepper()
function was moved to the recipes package. This makes the rsample install footprint much smaller.rsplit
objects are shown differently inside of a tibble.Moved from the broom package to the generics package.
rsample 0.0.2
CRAN release: 2017-11-12
-
initial_split
,training
, andtesting
were added to do training/testing splits prior to resampling. - Another resampling method,
group_vfold_cv
, was added. -
caret2rsample
andrsample2caret
can convertrset
objects to those used bycaret::trainControl
and vice-versa. - A function called
form_pred
can be used to determine the original names of the predictors in a formula orterms
object. - A vignette and a function (
prepper
) were included to facilitate using the recipes with rsample. - A
gather
method was added forrset
objects. - A
labels
method was added forrsplit
objects. This can help identify which resample is being used even when the wholerset
object is not available. - A variety of dplyr methods were added (e.g.
filter()
,mutate()
, etc) that work without dropping classes or attributes of the rsample objects.