To demonstrate parsnip for classification models, the credit data will be used.
library(tidymodels)
+library(tidymodels)
#> Registered S3 method overwritten by 'xts':
#> method from
#> as.zoo.xts zoo
-#> ── Attaching packages ───────────────────────────────── tidymodels 0.0.2 ──
+#> ── Attaching packages ──────────────────────────────────────────────────────────────────────────────────── tidymodels 0.0.2 ──
#> ✔ broom 0.5.1 ✔ purrr 0.3.2
-#> ✔ dials 0.0.2 ✔ recipes 0.1.4
+#> ✔ dials 0.0.2 ✔ recipes 0.1.5
#> ✔ dplyr 0.8.0.1 ✔ rsample 0.0.4
#> ✔ infer 0.4.0 ✔ yardstick 0.0.2
-#> ── Conflicts ──────────────────────────────────── tidymodels_conflicts() ──
+#> ── Conflicts ─────────────────────────────────────────────────────────────────────────────────────── tidymodels_conflicts() ──
#> ✖ purrr::discard() masks scales::discard()
#> ✖ dplyr::filter() masks stats::filter()
#> ✖ dplyr::lag() masks stats::lag()
#> ✖ recipes::step() masks stats::step()
-data(credit_data)
+data(credit_data)
-set.seed(7075)
+set.seed(7075)
data_split <- initial_split(credit_data, strata = "Status", p = 0.75)
credit_train <- training(data_split)
@@ -136,41 +136,42 @@ Classification Example
test_normalized <- bake(credit_rec, new_data = credit_test, all_predictors())
keras will be used to fit a model with 5 hidden units and uses a 10% dropout rate to regularize the model. At each training iteration (aka epoch) a random 20% of the data will be used to measure the cross-entropy of the model.
-set.seed(57974)
+set.seed(57974)
nnet_fit <-
mlp(epochs = 100, hidden_units = 5, dropout = 0.1) %>%
- # Also set engine-specific arguments:
- set_engine("keras", verbose = 0, validation_split = .20) %>%
- fit(Status ~ ., data = juice(credit_rec))
-
-nnet_fit
-#> parsnip model object
-#>
-#> Model
-#> ___________________________________________________________________________
-#> Layer (type) Output Shape Param #
-#> ===========================================================================
-#> dense_1 (Dense) (None, 5) 115
-#> ___________________________________________________________________________
-#> dense_2 (Dense) (None, 5) 30
-#> ___________________________________________________________________________
-#> dropout_1 (Dropout) (None, 5) 0
-#> ___________________________________________________________________________
-#> dense_3 (Dense) (None, 2) 12
-#> ===========================================================================
-#> Total params: 157
-#> Trainable params: 157
-#> Non-trainable params: 0
-#> ___________________________________________________________________________
+ set_mode("classification") %>%
+ # Also set engine-specific arguments:
+ set_engine("keras", verbose = 0, validation_split = .20) %>%
+ fit(Status ~ ., data = juice(credit_rec))
+
+nnet_fit
+#> parsnip model object
+#>
+#> Model
+#> ___________________________________________________________________________
+#> Layer (type) Output Shape Param #
+#> ===========================================================================
+#> dense_1 (Dense) (None, 5) 115
+#> ___________________________________________________________________________
+#> dense_2 (Dense) (None, 5) 30
+#> ___________________________________________________________________________
+#> dropout_1 (Dropout) (None, 5) 0
+#> ___________________________________________________________________________
+#> dense_3 (Dense) (None, 2) 12
+#> ===========================================================================
+#> Total params: 157
+#> Trainable params: 157
+#> Non-trainable params: 0
+#> ___________________________________________________________________________
In parsnip, the predict function can be used:.
test_results <-
credit_test %>%
select(Status) %>%
as_tibble() %>%
mutate(
- nnet_class = predict(nnet_fit, new_data = test_normalized) %>%
+ nnet_class = predict(nnet_fit, new_data = test_normalized) %>%
pull(.pred_class),
- nnet_prob = predict(nnet_fit, new_data = test_normalized, type = "prob") %>%
+ nnet_prob = predict(nnet_fit, new_data = test_normalized, type = "prob") %>%
pull(.pred_good)
)
@@ -183,12 +184,12 @@ Classification Example
#> # A tibble: 1 x 3
#> .metric .estimator .estimate
#> <chr> <chr> <dbl>
-#> 1 accuracy binary 0.801
+#> 1 accuracy binary 0.805
test_results %>% conf_mat(truth = Status, nnet_class)
#> Truth
#> Prediction bad good
-#> bad 187 95
-#> good 126 705
+#> bad 185 89
+#> good 128 711check_mod_val.RdThese functions are similar to constructors and can be used to validate +that there are no conflicts with the underlying model structures used by the +package.
+ +pred_types + +check_mod_val(model, new = FALSE, existence = FALSE) + +get_model_env() + +check_mode_val(mode) + +check_engine_val(eng) + +check_arg_val(arg) + +check_submodels_val(has_submodel) + +check_func_val(func) + +check_fit_info(fit_obj) + +check_pred_info(pred_obj, type) + +check_pkg_val(pkg) + +set_new_model(model) + +set_model_mode(model, mode) + +set_model_engine(model, mode, eng) + +set_model_arg(model, eng, parsnip, original, func, has_submodel) + +set_dependency(model, eng, pkg) + +get_dependency(model) + +set_fit(model, mode, eng, value) + +get_fit(model) + +set_pred(model, mode, eng, type, value) + +get_pred_type(model, type) + +show_model_info(model) + +get_from_env(items)+ +
| model | +A single character string for the model type (e.g.
+ |
+
|---|---|
| new | +A single logical to check to see if the model that you are check +has not already been registered. |
+
| existence | +A single logical to check to see if the model has already +been registered. |
+
| mode | +A single character string for the model mode (e.g. "regression"). |
+
| eng | +A single character string for the model engine. |
+
| arg | +A single character string for the model argument name. |
+
| has_submodel | +A single logical for whether the argument +can make predictions on mutiple submodels at once. |
+
| func | +A named character vector that describes how to call
+a function. |
+
| fit_obj | +A list with elements |
+
| pred_obj | +A list with elements |
+
| type | +A single character value for the type of prediction. Possible +values are: +'raw', 'numeric', 'class', 'link', 'prob', 'conf_int', 'pred_int', 'quantile' +. |
+
| pkg | +An options character string for a package name. |
+
| parsnip | +A single character string for the "harmonized" argument name
+that |
+
| original | +A single character string for the argument name that +underlying model function uses. |
+
| value | +A list that conforms to the |
+
| items | +A character string of objects in the model environment. |
+
An object of class character of length 8.
These functions are available for users to add their
+own models or engines (in package or otherwise) so that they can
+be accessed using parsnip. This are more thoroughly documented
+on the package web site (see references below).
In short, parsnip stores an environment object that contains
+all of the information and code about how models are used (e.g.
+fitting, predicting, etc). These functions can be used to add
+models to that environment as well as helper functions that can
+be used to makes sure that the model data is in the right
+format.
"Making a parsnip model from scratch" +https://tidymodels.github.io/parsnip/articles/articles/Scratch.html
+ + ++# Show the information about a model: +show_model_info("rand_forest")#> Information for `rand_forest` +#> modes: unknown, classification, regression +#> +#> engines: +#> classification: randomForest, ranger, spark +#> regression: randomForest, ranger, spark +#> +#> arguments: +#> ranger: +#> mtry --> mtry +#> trees --> num.trees +#> min_n --> min.node.size +#> randomForest: +#> mtry --> mtry +#> trees --> ntree +#> min_n --> nodesize +#> spark: +#> mtry --> feature_subset_strategy +#> trees --> num_trees +#> min_n --> min_instances_per_node +#> +#> fit modules: +#> engine mode +#> ranger classification +#> ranger regression +#> randomForest classification +#> randomForest regression +#> spark classification +#> spark regression +#> +#> prediction modules: +#> mode engine methods +#> classification randomForest class, prob, raw +#> classification ranger class, conf_int, prob, raw +#> classification spark class, prob +#> regression randomForest numeric, raw +#> regression ranger conf_int, numeric, raw +#> regression spark numeric +#>#> [1] "boost_tree" "boost_tree_args" +#> [3] "boost_tree_fit" "boost_tree_modes" +#> [5] "boost_tree_pkgs" "boost_tree_predict" +#> [7] "decision_tree" "decision_tree_args" +#> [9] "decision_tree_fit" "decision_tree_modes" +#> [11] "decision_tree_pkgs" "decision_tree_predict" +#> [13] "linear_reg" "linear_reg_args" +#> [15] "linear_reg_fit" "linear_reg_modes" +#> [17] "linear_reg_pkgs" "linear_reg_predict" +#> [19] "logistic_reg" "logistic_reg_args" +#> [21] "logistic_reg_fit" "logistic_reg_modes" +#> [23] "logistic_reg_pkgs" "logistic_reg_predict" +#> [25] "mars" "mars_args" +#> [27] "mars_fit" "mars_modes" +#> [29] "mars_pkgs" "mars_predict" +#> [31] "mlp" "mlp_args" +#> [33] "mlp_fit" "mlp_modes" +#> [35] "mlp_pkgs" "mlp_predict" +#> [37] "models" "modes" +#> [39] "multinom_reg" "multinom_reg_args" +#> [41] "multinom_reg_fit" "multinom_reg_modes" +#> [43] "multinom_reg_pkgs" "multinom_reg_predict" +#> [45] "nearest_neighbor" "nearest_neighbor_args" +#> [47] "nearest_neighbor_fit" "nearest_neighbor_modes" +#> [49] "nearest_neighbor_pkgs" "nearest_neighbor_predict" +#> [51] "null_model" "null_model_args" +#> [53] "null_model_fit" "null_model_modes" +#> [55] "null_model_pkgs" "null_model_predict" +#> [57] "rand_forest" "rand_forest_args" +#> [59] "rand_forest_fit" "rand_forest_modes" +#> [61] "rand_forest_pkgs" "rand_forest_predict" +#> [63] "surv_reg" "surv_reg_args" +#> [65] "surv_reg_fit" "surv_reg_modes" +#> [67] "surv_reg_pkgs" "surv_reg_predict" +#> [69] "svm_poly" "svm_poly_args" +#> [71] "svm_poly_fit" "svm_poly_modes" +#> [73] "svm_poly_pkgs" "svm_poly_predict" +#> [75] "svm_rbf" "svm_rbf_args" +#> [77] "svm_rbf_fit" "svm_rbf_modes" +#> [79] "svm_rbf_pkgs" "svm_rbf_predict"+
#> Classes ‘tbl_df’, ‘tbl’ and 'data.frame': 13626 obs. of 25 variables: +#> Classes ‘tbl_df’, ‘tbl’ and 'data.frame': 13626 obs. of 25 variables: #> $ package : chr "A3" "abbyyR" "abc" "abc.data" ... #> $ authors : int 1 1 1 1 5 3 2 1 4 6 ... #> $ imports : num 0 6 0 0 3 1 0 4 0 7 ... diff --git a/docs/dev/reference/decision_tree.html b/docs/dev/reference/decision_tree.html index bbb7cf34b..22825643e 100644 --- a/docs/dev/reference/decision_tree.html +++ b/docs/dev/reference/decision_tree.html @@ -159,7 +159,7 @@@@ -168,7 +168,7 @@General Interface for Decision Tree Models
time that the model is fit. Other options and argument can be set usingset_engine(). If left to their defaults here (NULL), the values are taken from the underlying model -functions. If parameters need to be modified,update()can be used +functions. If parameters need to be modified,update()can be used in lieu of recreating the object from scratch.General Interface for Decision Tree Models
tree_depth = NULL, min_n = NULL) # S3 method for decision_tree -update(object, cost_complexity = NULL, +update(object, cost_complexity = NULL, tree_depth = NULL, min_n = NULL, fresh = FALSE, ...)Arguments
@@ -205,15 +205,15 @@Arg
... -+ Not used for
update().Not used for
update().Details
-The model can be created using the
fit()function using the +The model can be created using the
fit()function using the following engines:-
- +
R:
"rpart"or"C5.0"(classification only)R:
"rpart"(the default) or"C5.0"(classification only)Spark:
"spark"Note that, for
rpartmodels, butcost_complexityand @@ -226,13 +226,13 @@Note
For models created using the spark engine, there are several differences to consider. First, only the formula -interface to via
@@ -269,7 +269,7 @@fit()is available; usingfit_xy()will +interface to viafit()is available; usingfit_xy()will generate an error. Second, the predictions will always be in a spark table format. The names will be the same as documented but without the dots. Third, there is no equivalent to factor columns in spark tables so class predictions are returned as character columns. Fourth, to retain the model object for a new -R session (viasave()), themodel$fitelement of theparsnip+R session (viasave()), themodel$fitelement of theparsnipobject should be serialized viaml_save(object$fit)and separately saved to disk. In a new session, the object can be reloaded and reattached to theparsnipobject.See also
- +Examples
@@ -288,12 +288,12 @@Examp #> Main Arguments: #> cost_complexity = 10 #> min_n = 3 -#>
#> Decision Tree Model Specification (unknown) +#>#> Decision Tree Model Specification (unknown) #> #> Main Arguments: #> cost_complexity = 1 #> min_n = 3 -#>#> Decision Tree Model Specification (unknown) +#>#> Decision Tree Model Specification (unknown) #> #> Main Arguments: #> cost_complexity = 1 diff --git a/docs/dev/reference/descriptors.html b/docs/dev/reference/descriptors.html index cf0396989..22694b7e7 100644 --- a/docs/dev/reference/descriptors.html +++ b/docs/dev/reference/descriptors.html @@ -134,7 +134,7 @@diff --git a/docs/dev/reference/keras_mlp.html b/docs/dev/reference/keras_mlp.html index 51ffd154c..ed96c15ed 100644 --- a/docs/dev/reference/keras_mlp.html +++ b/docs/dev/reference/keras_mlp.html @@ -139,7 +139,7 @@Data Set Characteristics Available when Fitting Models
-When using the
fit()functions there are some +When using the
fit()functions there are some variables that will be available for use in arguments. For example, if the user would like to choose an argument value based on the current number of rows in a data set, the.obs()@@ -174,7 +174,7 @@Details
.y(): The known outcomes returned in the format given. Either a vector, matrix, or data frame.
.dat(): A data frame containing all of the predictors and the -outcomes. Iffit_xy()was used, the outcomes are attached as the +outcomes. Iffit_xy()was used, the outcomes are attached as the column,..y.For example, if you use the model formula
Sepal.Width ~ .with theiris@@ -200,7 +200,7 @@Details
To use these in a model fit, pass them to a model specification. The evaluation is delayed until the time when the -model is run via
fit()(and the variables listed above are available). +model is run viafit()(and the variables listed above are available). For example:data("lending_club") diff --git a/docs/dev/reference/fit.html b/docs/dev/reference/fit.html index 59c76b4e1..8bd1bbf94 100644 --- a/docs/dev/reference/fit.html +++ b/docs/dev/reference/fit.html @@ -132,18 +132,18 @@Fit a Model Specification to a Dataset
-
fit()andfit_xy()take a model specification, translate the required +
fit()andfit_xy()take a model specification, translate the required code by substituting arguments, and execute the model fit routine.# S3 method for model_spec -fit(object, formula = NULL, data = NULL, +fit(object, formula = NULL, data = NULL, control = fit_control(), ...) # S3 method for model_spec -fit_xy(object, x = NULL, y = NULL, +fit_xy(object, x = NULL, y = NULL, control = fit_control(), ...)Arguments
@@ -206,21 +206,24 @@Value
Details
-
fit()andfit_xy()substitute the current arguments in the model +
fit()andfit_xy()substitute the current arguments in the model specification into the computational engine's code, checks them for validity, then fits the model using the data and the engine-specific code. Different model functions have different interfaces (e.g. formula orx/y) and these functions translate -between the interface used whenfit()orfit_xy()were invoked and the one +between the interface used whenfit()orfit_xy()were invoked and the one required by the underlying model.When possible, these functions attempt to avoid making copies of the data. For example, if the underlying model uses a formula and -
+fit()is invoked, the original data are references +fit()is invoked, the original data are references when the model is fit. However, if the underlying model uses something else, such asx/y, the formula is evaluated and the data are converted to the required format. In this case, any calls in the resulting model objects reference the temporary objects used to fit the model.If the model engine has not been set, the model's default engine will be used +(as discussed on each model page). If the
verbosityoption of +fit_control()is greater than zero, a warning will be produced.See also
@@ -231,28 +234,26 @@Examp
# Although `glm()` only has a formula interface, different # methods for specifying the model can be used -library(dplyr)#> +library(dplyr)#> #>#> #> #>#> #> #>#> #> -#>data("lending_club") lr_mod <- logistic_reg() using_formula <- lr_mod %>% set_engine("glm") %>% - fit(Class ~ funded_amnt + int_rate, data = lending_club) + fit(Class ~ funded_amnt + int_rate, data = lending_club) using_xy <- lr_mod %>% set_engine("glm") %>% - fit_xy(x = lending_club[, c("funded_amnt", "int_rate")], + fit_xy(x = lending_club[, c("funded_amnt", "int_rate")], y = lending_club$Class) using_formula#> parsnip model object diff --git a/docs/dev/reference/fit_control.html b/docs/dev/reference/fit_control.html index aa73f18a4..d93ee2181 100644 --- a/docs/dev/reference/fit_control.html +++ b/docs/dev/reference/fit_control.html @@ -131,7 +131,7 @@@@ -339,6 +359,7 @@Control the fit function
-@@ -154,7 +154,7 @@Options can be passed to the
fit()function that control the output and +Options can be passed to the
fit()function that control the output and computationsArg
diff --git a/docs/dev/reference/get_model_env.html b/docs/dev/reference/get_model_env.html new file mode 100644 index 000000000..0c2b8d9bb --- /dev/null +++ b/docs/dev/reference/get_model_env.html @@ -0,0 +1,191 @@ + + + + + + + + + catch A logical where a value of
TRUEwill evaluate -the model inside oftry(, silent = TRUE). If the model fails, +the model inside oftry(, silent = TRUE). If the model fails, an object is still returned (without an error) that inherits the class "try-error".Tools to Register Models — get_model_env • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +++ + + + + + diff --git a/docs/dev/reference/index.html b/docs/dev/reference/index.html index 3252d3df9..0540feb14 100644 --- a/docs/dev/reference/index.html +++ b/docs/dev/reference/index.html @@ -226,6 +226,12 @@+ + + + + +++ + +++ +++ +Tools to Register Models
+ ++get_model_env.Rd+ ++ +Tools to Register Models
+ +get_model_env() + +set_new_model(model) + +set_model_mode(model, mode) + +set_model_engine(model, mode, eng) + +set_model_arg(model, eng, parsnip, original, func, has_submodel) + +set_dependency(model, eng, pkg) + +get_dependency(model) + +set_fit(model, mode, eng, value) + +get_fit(model) + +set_pred(model, mode, eng, type, value) + +get_pred_type(model, type) + +show_model_info(model) + +get_from_env(items)+ + +add_rowindex() + +
+ Add a column of row numbers to a data frame
+ + +@@ -329,6 +335,20 @@ <
Execution Time Data
+ ++ +Developer Tools
+ ++ + + ++
pred_typescheck_mod_val()get_model_env()check_mode_val()check_engine_val()check_arg_val()check_submodels_val()check_func_val()check_fit_info()check_pred_info()check_pkg_val()set_new_model()set_model_mode()set_model_engine()set_model_arg()set_dependency()get_dependency()set_fit()get_fit()set_pred()get_pred_type()show_model_info()get_from_env()+ Tools to Register Models
Contents
Models Infrastructure Data +Developer Tools Simple interface to MLP models via keras
keras_mlp(x, y, hidden_units = 5, decay = 0, dropout = 0, - epochs = 20, act = "softmax", seeds = sample.int(10^5, size = 3), + epochs = 20, act = "softmax", seeds = sample.int(10^5, size = 3), ...)
#> Classes ‘tbl_df’, ‘tbl’ and 'data.frame': 9857 obs. of 23 variables: +#> Classes ‘tbl_df’, ‘tbl’ and 'data.frame': 9857 obs. of 23 variables: #> $ funded_amnt : int 16100 32000 10000 16800 3500 10000 11000 15000 6000 20000 ... #> $ term : Factor w/ 2 levels "term_36","term_60": 1 2 1 2 1 1 1 1 1 2 ... #> $ int_rate : num 13.99 11.99 16.29 13.67 7.39 ... diff --git a/docs/dev/reference/linear_reg.html b/docs/dev/reference/linear_reg.html index 35a3375f8..31dfc2fbf 100644 --- a/docs/dev/reference/linear_reg.html +++ b/docs/dev/reference/linear_reg.html @@ -155,7 +155,7 @@@@ -163,7 +163,7 @@General Interface for Linear Regression Models
time that the model is fit. Other options and argument can be set usingset_engine(). If left to their defaults here (NULL), the values are taken from the underlying model -functions. If parameters need to be modified,update()can be used +functions. If parameters need to be modified,update()can be used in lieu of recreating the object from scratch.General Interface for Linear Regression Models
linear_reg(mode = "regression", penalty = NULL, mixture = NULL) # S3 method for linear_reg -update(object, penalty = NULL, mixture = NULL, +update(object, penalty = NULL, mixture = NULL, fresh = FALSE, ...)Arguments
@@ -200,7 +200,7 @@Arg
@@ -209,9 +209,9 @@ ... -+ Not used for
update().Not used for
update().Details
The data given to the function are not saved and are only used to determine the mode of the model. For
-linear_reg(), the mode will always be "regression".The model can be created using the
fit()function using the +The model can be created using the
fit()function using the following engines:-
- +
R:
"lm"or"glmnet"R:
"lm"(the default) or"glmnet"Stan:
"stan"Spark:
"spark"- @@ -221,13 +221,13 @@
keras:
"keras"Note
For models created using the spark engine, there are several differences to consider. First, only the formula -interface to via
@@ -265,8 +265,8 @@fit()is available; usingfit_xy()will +interface to viafit()is available; usingfit_xy()will generate an error. Second, the predictions will always be in a spark table format. The names will be the same as documented but without the dots. Third, there is no equivalent to factor columns in spark tables so class predictions are returned as character columns. Fourth, to retain the model object for a new -R session (viasave()), themodel$fitelement of theparsnip+R session (viasave()), themodel$fitelement of theparsnipobject should be serialized viaml_save(object$fit)and separately saved to disk. In a new session, the object can be reloaded and reattached to theparsnipobject.See also
- +Examples
@@ -296,12 +296,12 @@Examp #> Main Arguments: #> penalty = 10 #> mixture = 0.1 -#>
#> Linear Regression Model Specification (regression) +#>#> Linear Regression Model Specification (regression) #> #> Main Arguments: #> penalty = 1 #> mixture = 0.1 -#>#> Linear Regression Model Specification (regression) +#>#> Linear Regression Model Specification (regression) #> #> Main Arguments: #> penalty = 1 diff --git a/docs/dev/reference/logistic_reg.html b/docs/dev/reference/logistic_reg.html index 7dde682bf..ca4332916 100644 --- a/docs/dev/reference/logistic_reg.html +++ b/docs/dev/reference/logistic_reg.html @@ -155,7 +155,7 @@@@ -163,7 +163,7 @@General Interface for Logistic Regression Models
time that the model is fit. Other options and argument can be set usingset_engine(). If left to their defaults here (NULL), the values are taken from the underlying model -functions. If parameters need to be modified,update()can be used +functions. If parameters need to be modified,update()can be used in lieu of recreating the object from scratch.General Interface for Logistic Regression Models
logistic_reg(mode = "classification", penalty = NULL, mixture = NULL) # S3 method for logistic_reg -update(object, penalty = NULL, mixture = NULL, +update(object, penalty = NULL, mixture = NULL, fresh = FALSE, ...)Arguments
@@ -200,16 +200,16 @@Arg
... -+ Not used for
update().Not used for
update().Details
For
-logistic_reg(), the mode will always be "classification".The model can be created using the
fit()function using the +The model can be created using the
fit()function using the following engines:
R: "glm" or "glmnet"
R: "glm" (the default) or "glmnet"
Stan: "stan"
Spark: "spark"
keras: "keras"
For models created using the spark engine, there are
several differences to consider. First, only the formula
-interface to via fit() is available; using fit_xy() will
+interface to via fit() is available; using fit_xy() will
generate an error. Second, the predictions will always be in a
spark table format. The names will be the same as documented but
without the dots. Third, there is no equivalent to factor
columns in spark tables so class predictions are returned as
character columns. Fourth, to retain the model object for a new
-R session (via save()), the model$fit element of the parsnip
+R session (via save()), the model$fit element of the parsnip
object should be serialized via ml_save(object$fit) and
separately saved to disk. In a new session, the object can be
reloaded and reattached to the parsnip object.
set_engine(). If left to their defaults
here (NULL), the values are taken from the underlying model
-functions. If parameters need to be modified, update() can be used
+functions. If parameters need to be modified, update() can be used
in lieu of recreating the object from scratch.
Not used for update().
Not used for update().
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "earth"
R: "earth" (the default)
Not used for update().
Not used for update().
set_engine(). If left to their defaults
here (see above), the values are taken from the underlying model
-functions. One exception is hidden_units when nnet::nnet is used; that
+functions. One exception is hidden_units when nnet::nnet is used; that
function's size argument has no default so a value of 5 units will be
used. Also, unless otherwise specified, the linout argument to
-nnet::nnet() will be set to TRUE when a regression model is created.
-If parameters need to be modified, update() can be used
+nnet::nnet() will be set to TRUE when a regression model is created.
+If parameters need to be modified, update() can be used
in lieu of recreating the object from scratch.
-The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "nnet"
R: "nnet" (the default)
keras: "keras"
An error is thrown if both penalty and dropout are specified for
@@ -271,7 +271,7 @@
set_engine(). If left to their defaults
here (NULL), the values are taken from the underlying model
-functions. If parameters need to be modified, update() can be used
+functions. If parameters need to be modified, update() can be used
in lieu of recreating the object from scratch.
A single integer for the number of neighbors
-to consider (often called k).
k). For kknn, a value of 5
+is used if neighbors is not specified.
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "kknn"
R: "kknn" (the default)
kknn::train.kknn(formula = missing_arg(), data = missing_arg(), - kmax = missing_arg()) + ks = 5)
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "parsnip"
| type | A single character value or |
|
|---|---|---|
| ... | -Not used for |
+ Not used for |
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "ranger" or "randomForest"
R: "ranger" (the default) or "randomForest"
Spark: "spark"
For models created using the spark engine, there are
several differences to consider. First, only the formula
-interface to via fit() is available; using fit_xy() will
+interface to via fit() is available; using fit_xy() will
generate an error. Second, the predictions will always be in a
spark table format. The names will be the same as documented but
without the dots. Third, there is no equivalent to factor
@@ -274,7 +274,7 @@
These objects are imported from other packages. Follow the links below to see their documentation.
%>%
surv_reg(mode = "regression", dist = NULL) # S3 method for surv_reg -update(object, dist = NULL, fresh = FALSE, ...)+update(object, dist = NULL, fresh = FALSE, ...)
| ... | -Not used for |
+ Not used for |
|---|
surv_reg(),the
mode will always be "regression".
Since survival models typically involve censoring (and require the use of
-survival::Surv() objects), the fit() function will require that the
+survival::Surv() objects), the fit() function will require that the
survival model be specified via the formula interface.
Also, for the flexsurv::flexsurvfit engine, the typical
strata function cannot be used. To achieve the same effect,
the extra parameter roles can be used (as described above).
For surv_reg(), the mode will always be "regression".
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "flexsurv", "survreg"
R: "flexsurv", "survival" (the default)
survival
survival::survreg(formula = missing_arg(), data = missing_arg(),
weights = missing_arg(), model = TRUE)
@@ -233,7 +233,7 @@ R
See also
-
+
Examples
@@ -249,7 +249,7 @@ Examp
#>
#> Main Arguments:
#> dist = weibull
-#>
set_engine(). If left to their defaults
here (NULL), the values are taken from the underlying model
-functions. If parameters need to be modified, update() can be used
+functions. If parameters need to be modified, update() can be used
in lieu of recreating the object from scratch.
Not used for update().
Not used for update().
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "kernlab"
R: "kernlab" (the default)
set_engine(). If left to their defaults
here (NULL), the values are taken from the underlying model
-functions. If parameters need to be modified, update() can be used
+functions. If parameters need to be modified, update() can be used
in lieu of recreating the object from scratch.
Not used for update().
Not used for update().
The model can be created using the fit() function using the
+
The model can be created using the fit() function using the
following engines:
R: "kernlab"
R: "kernlab" (the default)
translate() produces a template call that lacks the specific
argument values (such as data, etc). These are filled in once
-fit() is called with the specifics of the data for the model.
+fit() is called with the specifics of the data for the model.
The call may also include varying arguments if these are in
the specification.
It does contain the resolved argument names that are specific to diff --git a/docs/dev/reference/varying_args.model_spec.html b/docs/dev/reference/varying_args.model_spec.html index 22a240d23..526546ac9 100644 --- a/docs/dev/reference/varying_args.model_spec.html +++ b/docs/dev/reference/varying_args.model_spec.html @@ -132,20 +132,20 @@
varying_args() takes a model specification or a recipe and returns a tibble
+
varying_args() takes a model specification or a recipe and returns a tibble
of information on all possible varying arguments and whether or not they
are actually varying.
# S3 method for model_spec -varying_args(object, full = TRUE, ...) +varying_args(object, full = TRUE, ...) # S3 method for recipe -varying_args(object, full = TRUE, ...) +varying_args(object, full = TRUE, ...) # S3 method for step -varying_args(object, full = TRUE, ...)+varying_args(object, full = TRUE, ...)