CRAN release: 2023-08-17
Fixed bug where sparse data was being coerced to non-sparse format doing
augment()now works for censored regression models.
CRAN release: 2023-04-12
Added several internal functions (to help work with
Survobjects) as a standalone file that can be used in other packages via
usethis::use_standalone("tidymodels/parsnip"). These changes provide tooling for downstream packages to handle inverse probability censoring weights (#893, #897, #937).
An internal method for generating inverse probability of censoring weights (IPCW) of Graf et al (1999) is available via
fit()behave consistently with respect to missingness in the classification setting. Previously,
fit()erroneously raised an error about the class of the outcome when there were no complete cases, and now always passes along complete cases to be handled by the modeling function (#888).
Fixed bug where model fits with
engine = "earth"would fail when the package’s namespace hadn’t been attached (#251).
Fixed bug where model fits with factor predictors and
engine = "kknn"would fail when the package’s namespace hadn’t been attached (#264).
Fixed bug with prediction from a boosted tree model fitted with
"xgboost"using a custom objective function (#875).
Rather than being implemented in each method, the check for the
new_dataargument being mistakenly passed as
multi_predict()now happens in the generic. Packages re-exporting the
multi_predict()generic and implementing now-duplicate checks may see new failures and can remove their own analogous checks. This check already existed in all
predict.model_fit()) and all parsnip
Functions now indicate what class the outcome was if the outcome is the wrong class (#887).
The minimum version for R is now 3.5 (#926).
Various bug fixes and improvements to documentation.
CRAN release: 2023-02-22
For censored regression models, a “reverse Kaplan-Meier” curve is computed for the censoring distribution. This can be used when evaluating this type of model (#855).
CRAN release: 2022-11-11
lightgbmengine (via the bonsai package) is now tunable.
A change in our data checking code resulted in about a 3-fold speed-up in parsnip (#835)
CRAN release: 2022-10-01
A bagged neural network model was added (
bag_mlp()). Engine implementations will live in the baguette package.
Fixed installation failures due to undocumented knitr installation dependency (#785).
fit_xy()now fails when the model mode is unknown.
brulee engine-specific tuning parameters were updated. These changes can be used with dials version > 1.0.0.
Improved prompts related to missing (or not loaded) extension packages as well as better handling of model mode conflicts.
CRAN release: 2022-08-18
- Enabled passing additional engine arguments with the xgboost
boost_tree()engine. To supply engine-specific arguments that are documented in
xgboost::xgb.train()as arguments to be passed via
params, supply the list elements directly as named arguments to
set_engine(). Read more in
CRAN release: 2022-06-16
Enable the use of case weights for models that support them.
show_model_info()now indicates which models can utilize case weights.
Model type functions will now message informatively if a needed parsnip extension package is not loaded (#731).
Refactored internals of model specification printing functions. These changes are non-breaking for extension packages, but the new
print_model_spec()helper is exported for use in extensions if desired (#739).
Fixed a bug where an error would be thrown if arguments to model functions were namespaced (#745).
predict(type = "prob")will now provide an error if the outcome variable has a level called
An inconsistency for probability type predictions for two-class GAM models was fixed (#708)
xgb_train()now allows for case weights
Added a developer function,
.model_param_name_keythat translates names of tuning parameters.
CRAN release: 2022-03-17
Fixed a major bug in spark models induced in the previous version (#671).
Updated the parsnip add-in with new models and engines.
Updated parameter ranges for some
tunable()methods and added a missing engine argument for brulee models.
Added information about how to install the mixOmics package for PLS models (#680)
CRAN release: 2022-03-09
A bug for class predictions of two-class GAM models was fixed (#541)
The list column produced when creating survival probability predictions is now always called
.pred_survivalbeing used inside of the list column).
Fixed outcome type checking affecting a subset of regression models (#625).
When the xy interface is used and the underlying model expects to use a matrix, a better warning is issued when predictors contain non-numeric columns (including dates).
extract_parameter_set_dials()method to extract parameter sets from model specs.
extract_parameter_dials()method to extract a single parameter from model specs.
intervalwas added for prediction: For types
"quantile", estimates for the confidence or prediction interval can be added if available (#615).
varying_args()is soft-deprecated in favor of
parsnip is now more robust working with keras and tensorflow for a larger range of versions (#596).
xgboost engines now use the new
iterationrangeparameter instead of the deprecated
- Models information can be re-registered as long as the information being registered is the same. This is helpful for packages that add new engines and use
CRAN release: 2021-07-21
A model function (
gen_additive_mod()) was added for generalized additive models.
Each model now has a default engine that is used when the model is defined. The default for each model is listed in the help documents. This also adds functionality to declare an engine in the model specification function.
set_engine()is still required if engine-specific arguments need to be added. (#513)
parsnip now checks for a valid combination of engine and mode (#529)
The default engine for
multinom_reg()was changed to
The helper functions
.convert_xy_to_form_new()for converting between formula and matrix interface are now exported for developer use (#508).
New article “Fitting and Predicting with parsnip” which contains examples for various combinations of model type and engine. ( #527)
CRAN release: 2021-05-27
A new linear SVM model
svm_linear()is now available with the
LiblineaRengine (#424) and the
kernlabengine (#438), and the
LiblineaRengine is available for
logistic_reg()as well (#429). These models can use sparse matrices via
fit_xy()(#447) and have a
For models with
- A single value is required for
penalty(either a single numeric value or a value of
- A special argument called
path_valuescan be used to set the
lambdapath as a specific set of numbers (independent of the value of
penalty). A pure ridge regression models (i.e.,
mixture = 1) will generate incorrect values if the path does not include zero. See issue #431 for discussion (#486).
- A single value is required for
The xgboost engine for boosted trees was translating
colsample_bytree. We now map
colsample_bynodesince that is more consistent with how random forest works.
colsample_bytreecan still be optimized by passing it in as an engine argument.
colsample_bynodewas added to xgboost after the
parsnippackage code was written. (#495)
colsample_bytreecan be passed as integer counts or proportions, while
validationshould always be proportions.
xgb_train()now has a new option
FALSE) that states which scale for
colsample_bytreeis being used. (#461)
Re-licensed package from GPL-2 to MIT. See consent from copyright holders here.
Re-organized model documentation:
updatemethods were moved out of the model help files (#479).
- Each model/engine combination has its own help page.
- The model help page has a dynamic bulleted list of the engines with links to the individual help pages.
generics::required_pkgs()was extended for
Prediction functions now give a consistent error when a user uses an unavailable value of
xgboost engines now respect the
event_leveloption for predictions (#460).
CRAN release: 2021-01-19
An RStudio add-in is available that makes writing multiple
parsnipmodel specifications to the source window. It can be accessed via the IDE addin menus or by calling
xgboostmodels, users can now pass
Changes to test for cases when CRAN cannot get
xgboostto work on their Solaris configuration.
There is now an
augument()method for fitted models. See
There is now an
event_levelargument for the
New mode “censored regression” and new prediction types “linear_pred”, “time”, “survival”, “hazard”. (#396)
CRAN release: 2020-10-27
show_engines()will provide information on the current set for a model.
Some added protections were added for function arguments that are dependent on the data dimensions (e.g.,
min_n, etc). (#184)
Infrastructure was improved for running
parsnipmodels in parallel using PSOCK clusters on Windows.
CRAN release: 2020-08-04
CRAN release: 2020-07-03
parsnipnow has options to set specific types of predictor encodings for different models. For example,
rangermodels run using
workflowsdo the same thing by not creating indicator variables. These encodings can be overridden using the
workflows. As a consequence, it is possible to get a different model fit that previous versions of
parsnip. More details about specific encoding changes are below. (#326)
tidyr>= 1.0.0 is now required.
SVM models produced by
kernlabnow use the formula method (see breaking change notice above). This change was due to how
ksvm()made indicator variables for factor predictors (with one-hot encodings). Since the ordinary formula method did not do this, the data are passed as-is to
ksvm()so that the results are closer to what one would get if
ksmv()were called directly.
MARS models produced by
earthnow use the formula method.
xgboost, a one-hot encoding is used when indicator variables are created.
Under-the-hood changes were made so that non-standard data arguments in the modeling packages can be accommodated. (#315)
A new main argument was added to
stop_iterfor early stopping. The
xgb_train()function gained arguments for early stopping and a percentage of data to leave out for a validation set.
fit()is used and the underlying model uses a formula, the actual formula is pass to the model (instead of a placeholder). This makes the model call better.
A function named
repair_call()was added. This can help change the underlying models
callobject to better reflect what they would have obtained if the model function had been used directly (instead of via
parsnip). This is only useful when the user chooses a formula interface and the model uses a formula interface. It will also be of limited use when a recipes is used to construct the feature set in
CRAN release: 2020-05-06
- S3 dispatch for
tidy()was broken on R 4.0.
CRAN release: 2020-01-07
glmnetwas removed as a dependency since the new version depends on 3.6.0 or greater. Keeping it would constrain
parsnipto that same requirement. All
glmnettests are run locally.
A set of internal functions are now exported. These are helpful when creating a new package that registers new model specifications.
- There were some mis-mapped parameters (going between
parsnipand the underlying model function) for
sparkboosted trees and some
kerasmodels. See 897c927.
CRAN release: 2019-11-02
The time elapsed during model fitting is stored in the
$elapsedslot of the parsnip model object, and is printed when the model object is printed.
Some default parameter ranges were updated for SVM, KNN, and MARS models.
udpate()methods gained a
parametersargument for cases when the parameters are contained in a tibble or list.
A bug was fixed standardizing the output column types of
A bug was fixed related to the column names generated by
multi_predict(). The top-level tibble will always have a column named
.predand this list column contains tibbles across sub-models. The column names for these sub-model tibbles will have names consistent with
predict()(which was previously incorrect). See 43c15db.
A bug was fixed standardizing the column names of
nnetclass probability predictions.
CRAN release: 2019-07-31
Unplanned release based on CRAN requirements for Solaris.
The method that
parsnipstores the model information has changed. Any custom models from previous versions will need to use the new method for registering models. The methods are detailed in
?get_model_envand the package vignette for adding models.
The mode needs to be declared for models that can be used for more than one mode prior to fitting and/or translation.
surv_reg(), the engine that uses the
survivalpackage is now called
glmnetmodels, the full regularization path is always fit regardless of the value given to
penalty. Previously, the model was fit with passing
lambdaargument and the model could only make predictions at those specific values. (#195)
add_rowindex()can create a column called
.rowto a data frame.
If a computational engine is not explicitly set, a default will be used. Each default is documented on the corresponding model page. A warning is issued at fit time unless verbosity is zero.
A suite of internal functions were added to help with upcoming model tuning features.
parsnipobject always saved the name(s) of the outcome variable(s) for proper naming of the predicted values.
CRAN release: 2019-03-22
Small release driven by changes in
sample() in the current r-devel.
A “null model” is now available that fits a predictor-free model (using the mean of the outcome for regression or the mode for classification).
fit_xy()can take a single column data frame or matrix for
varying_args()now has a
fullargument to control whether the full set of possible varying arguments is returned (as opposed to only the arguments that are actually varying).
fit_control()not returns an S3 method.
For classification models, an error occurs if the outcome data are not encoded as factors (#115).
The prediction modules (e.g.
predict_numeric, etc) were de-exported. These were internal functions that were not to be used by the users and the users were using them.
An event time data set (
check_times) was included that is the time (in seconds) to run
R CMD checkusing the “r-devel-windows-ix86+x86_64` flavor. Packages that errored are censored.
varying_args()now uses the version from the
genericspackage. This means that the first argument,
x, has been renamed to
objectto align with generics.
find_varying(), the internal function for detecting varying arguments, now returns correct results when a size 0 argument is provided. It can also now detect varying arguments nested deeply into a call (#131, #134).
For multinomial regression, the
.pred_prefix is now only added to prediction column names once (#107).
Confidence and prediction intervals for logistic regression were only computed the intervals for a single level. Both are now computed. (#156)
- The engine, and any associated arguments, are now specified using
set_engine(). There is no
- Arguments to modeling functions are now captured as quosures.
othershas been replaced by
- Data descriptor names have been changed and are now functions. The descriptor definitions for “cols” and “preds” have been switched.
regularizationwas changed to
penaltyin a few models to be consistent with this change.
- If a mode is not chosen in the model specification, it is assigned at the time of fit. 51
- The underlying modeling packages now are loaded by namespace. There will be some exceptions noted in the documentation for each model. For example, in some
earthpackage will need to be attached to be fully operational.
- To be consistent with
newdatawas changed to
predict_rawmethod was added.