diff --git a/DESCRIPTION b/DESCRIPTION index 4d305e962..08001baaa 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -1,5 +1,5 @@ Package: parsnip -Version: 0.0.2.9000 +Version: 0.0.3 Title: A Common API to Modeling and Analysis Functions Description: A common interface is provided to allow users to specify a model without having to remember the different argument names across different functions or computational engines (e.g. 'R', 'Spark', 'Stan', etc). Authors@R: c( diff --git a/NAMESPACE b/NAMESPACE index 37745ade0..c4014df9a 100644 --- a/NAMESPACE +++ b/NAMESPACE @@ -5,6 +5,12 @@ S3method(fit_xy,model_spec) S3method(has_multi_predict,default) S3method(has_multi_predict,model_fit) S3method(has_multi_predict,workflow) +S3method(min_grid,boost_tree) +S3method(min_grid,linear_reg) +S3method(min_grid,logistic_reg) +S3method(min_grid,mars) +S3method(min_grid,multinom_reg) +S3method(min_grid,nearest_neighbor) S3method(multi_predict,"_C5.0") S3method(multi_predict,"_earth") S3method(multi_predict,"_elnet") @@ -50,8 +56,11 @@ S3method(print,svm_rbf) S3method(translate,boost_tree) S3method(translate,decision_tree) S3method(translate,default) +S3method(translate,linear_reg) +S3method(translate,logistic_reg) S3method(translate,mars) S3method(translate,mlp) +S3method(translate,multinom_reg) S3method(translate,nearest_neighbor) S3method(translate,rand_forest) S3method(translate,surv_reg) @@ -104,6 +113,13 @@ export(linear_reg) export(logistic_reg) export(make_classes) export(mars) +export(min_grid) +export(min_grid.boost_tree) +export(min_grid.linear_reg) +export(min_grid.logistic_reg) +export(min_grid.mars) +export(min_grid.multinom_reg) +export(min_grid.nearest_neighbor) export(mlp) export(model_printer) export(multi_predict) diff --git a/NEWS.md b/NEWS.md index 53875047a..dad6ada58 100644 --- a/NEWS.md +++ b/NEWS.md @@ -1,17 +1,26 @@ -# parsnip 0.0.2.9000 +# parsnip 0.0.3 + +Unplanned release based on CRAN requirements for Solaris. ## Breaking Changes - * The method that `parsnip` stores 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_env()` and the [package vignette for adding models](https://tidymodels.github.io/parsnip/articles/articles/Scratch.html). - * The mode need to be declared for models that can be used for more than one mode prior to fitting and/or translation). + * The method that `parsnip` stores 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_env` and the [package vignette for adding models](https://tidymodels.github.io/parsnip/articles/articles/Scratch.html). + + * The mode needs to be declared for models that can be used for more than one mode prior to fitting and/or translation. + * For `surv_reg()`, the engine that uses the `survival` package is now called `survival` instead of `survreg`. + * For `glmnet` models, the full regularization path is always fit regardless of the value given to `penalty`. Previously, the model was fit with passing `penalty` to `glmnet`'s `lambda` argument and the model could only make predictions at those specific values. [(#195)](https://github.com/tidymodels/parsnip/issues/195) + ## New Features * `add_rowindex()` can create a column called `.row` to 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. - * `nearest_neighbor` gained a `multi_predict` method. The `multi_predict()` documentation is a little better organized. + + * `nearest_neighbor()` gained a `multi_predict` method. The `multi_predict()` documentation is a little better organized. + + * A suite of internal functions were added to help with upcoming model tuning features. # parsnip 0.0.2 diff --git a/R/aaa.R b/R/aaa.R index a49415299..303579c25 100644 --- a/R/aaa.R +++ b/R/aaa.R @@ -19,10 +19,102 @@ convert_stan_interval <- function(x, level = 0.95, lower = TRUE) { } # ------------------------------------------------------------------------------ +# min_grid generic - put here so that the generic shows up first in the man file + +#' Determine the minimum set of model fits +#' +#' `min_grid` determines exactly what models should be fit in order to +#' evaluate the entire set of tuning parameter combinations. This is for +#' internal use only and the API may change in the near future. +#' @param x A model specification. +#' @param grid A tibble with tuning parameter combinations. +#' @param ... Not currently used. +#' @return A tibble with the minimum tuning parameters to fit and an additional +#' list column with the parameter combinations used for prediction. +#' @keywords internal +#' @export +min_grid <- function(x, grid, ...) { + # x is a `model_spec` object from parsnip + # grid is a tibble of tuning parameter values with names + # matching the parameter names. + UseMethod("min_grid") +} + +# As an example, if we fit a boosted tree model and tune over +# trees = 1:20 and min_n = c(20, 30) +# we should only have to fit two models: +# +# trees = 20 & min_n = 20 +# trees = 20 & min_n = 30 +# +# The logic related to how this "mini grid" gets made is model-specific. +# +# To get the full set of predictions, we need to know, for each of these two +# models, what values of num_terms to give to the multi_predict() function. +# +# The current idea is to have a list column of the extra models for prediction. +# For the example above: +# +# # A tibble: 2 x 3 +# trees min_n .submodels +# +# 1 20 20 +# 2 20 30 +# +# and the .submodels would both be +# +# list(trees = 1:19) +# +# There are a lot of other things to consider in future versions like grids +# where there are multiple columns with the same name (maybe the results of +# a recipe) and so on. + +# ------------------------------------------------------------------------------ +# helper functions + +# Template for model results that do no have the sub-model feature +blank_submodels <- function(grid) { + grid %>% + dplyr::mutate(.submodels = map(1:nrow(grid), ~ list())) +} + +get_fixed_args <- function(info) { + # Get non-sub-model columns to iterate over + fixed_args <- info$name[!info$has_submodel] +} + +get_submodel_info <- function(spec, grid) { + param_info <- + get_from_env(paste0(class(spec)[1], "_args")) %>% + dplyr::filter(engine == spec$engine) %>% + dplyr::select(name = parsnip, has_submodel) + + # In case a recipe or other activity has grid parameter columns, + # add those to the results + grid_names <- names(grid) + is_mod_param <- grid_names %in% param_info$name + if (any(!is_mod_param)) { + param_info <- + param_info %>% + dplyr::bind_rows( + tibble::tibble(name = grid_names[!is_mod_param], + has_submodel = FALSE) + ) + } + param_info %>% dplyr::filter(name %in% grid_names) +} + + +# ------------------------------------------------------------------------------ +# nocov #' @importFrom utils globalVariables utils::globalVariables( c('.', '.label', '.pred', '.row', 'data', 'engine', 'engine2', 'group', 'lab', 'original', 'predicted_label', 'prediction', 'value', 'type', - "neighbors") + "neighbors", ".submodels", "has_submodel", "max_neighbor", "max_penalty", + "max_terms", "max_tree", "name", "num_terms", "penalty", "trees", + "sub_neighbors") ) + +# nocov end diff --git a/R/boost_tree.R b/R/boost_tree.R index 2963de1a4..ad9be1f2d 100644 --- a/R/boost_tree.R +++ b/R/boost_tree.R @@ -514,3 +514,41 @@ C50_by_tree <- function(tree, object, new_data, type, ...) { pred[, c(".row", "trees", nms)] } +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.boost_tree +#' @rdname min_grid +min_grid.boost_tree <- function(x, grid, ...) { + grid_names <- names(grid) + param_info <- get_submodel_info(x, grid) + + # No ability to do submodels? Finish here: + if (!any(param_info$has_submodel)) { + return(blank_submodels(grid)) + } + + fixed_args <- get_fixed_args(param_info) + + # For boosted trees, fit the model with the most trees (conditional on the + # other parameters) so that you can do predictions on the smaller models. + fit_only <- + grid %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(trees = max(trees, na.rm = TRUE)) %>% + dplyr::ungroup() + + # Add a column .submodels that is a list with what should be predicted + # by `multi_predict()` (assuming `predict()` has already been executed + # on the original value of 'trees') + min_grid_df <- + dplyr::full_join(fit_only %>% rename(max_tree = trees), grid, by = fixed_args) %>% + dplyr::filter(trees != max_tree) %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(.submodels = list(list(trees = trees))) %>% + dplyr::ungroup() %>% + dplyr::full_join(fit_only, grid, by = fixed_args) + + min_grid_df %>% dplyr::select(dplyr::one_of(grid_names), .submodels) +} + diff --git a/R/linear_reg.R b/R/linear_reg.R index f9e5a8e74..28ebc4935 100644 --- a/R/linear_reg.R +++ b/R/linear_reg.R @@ -68,9 +68,9 @@ #' #' \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::linear_reg(), "keras")} #' -#' When using `glmnet` models, there is the option to pass -#' multiple values (or no values) to the `penalty` argument. This -#' can have an effect on the model object results. When using the +#' For `glmnet` models, the full regularization path is always fit regardless +#' of the value given to `penalty`. Also, there is the option to pass +#' multiple values (or no values) to the `penalty` argument. When using the #' `predict()` method in these cases, the return value depends on #' the value of `penalty`. When using `predict()`, only a single #' value of the penalty can be used. When predicting on multiple @@ -138,6 +138,23 @@ print.linear_reg <- function(x, ...) { invisible(x) } + +#' @export +translate.linear_reg <- function(x, engine = x$engine, ...) { + x <- translate.default(x, engine, ...) + + if (engine == "glmnet") { + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + x$method$fit$args$lambda <- NULL + # Since the `fit` infomration is gone for the penalty, we need to have an + # evaludated value for the parameter. + x$args$penalty <- rlang::eval_tidy(x$args$penalty) + } + + x +} + + # ------------------------------------------------------------------------------ #' @inheritParams update.boost_tree @@ -274,6 +291,11 @@ predict._elnet <- if (any(names(enquos(...)) == "newdata")) stop("Did you mean to use `new_data` instead of `newdata`?", call. = FALSE) + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (is.null(penalty) & !is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } + object$spec$args$penalty <- check_penalty(penalty, object, multi) object$spec <- eval_args(object$spec) @@ -314,7 +336,12 @@ multi_predict._elnet <- object$spec <- eval_args(object$spec) if (is.null(penalty)) { - penalty <- object$fit$lambda + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (!is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } else { + penalty <- object$fit$lambda + } } pred <- predict._elnet(object, new_data = new_data, type = "raw", @@ -332,3 +359,37 @@ multi_predict._elnet <- names(pred) <- NULL tibble(.pred = pred) } + + +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.linear_reg +#' @rdname min_grid +min_grid.linear_reg <- function(x, grid, ...) { + + grid_names <- names(grid) + param_info <- get_submodel_info(x, grid) + + if (!any(param_info$has_submodel)) { + return(blank_submodels(grid)) + } + + fixed_args <- get_fixed_args(param_info) + + fit_only <- + grid %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(penalty = max(penalty, na.rm = TRUE)) %>% + dplyr::ungroup() + + min_grid_df <- + dplyr::full_join(fit_only %>% rename(max_penalty = penalty), grid, by = fixed_args) %>% + dplyr::filter(penalty != max_penalty) %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(.submodels = list(list(penalty = penalty))) %>% + dplyr::ungroup() %>% + dplyr::full_join(fit_only, grid, by = fixed_args) + + min_grid_df %>% dplyr::select(dplyr::one_of(grid_names), .submodels) +} diff --git a/R/logistic_reg.R b/R/logistic_reg.R index 7264680aa..d9d3c87a5 100644 --- a/R/logistic_reg.R +++ b/R/logistic_reg.R @@ -66,9 +66,9 @@ #' #' \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::logistic_reg(), "keras")} #' -#' When using `glmnet` models, there is the option to pass -#' multiple values (or no values) to the `penalty` argument. This -#' can have an effect on the model object results. When using the +#' For `glmnet` models, the full regularization path is always fit regardless +#' of the value given to `penalty`. Also, there is the option to pass +#' multiple values (or no values) to the `penalty` argument. When using the #' `predict()` method in these cases, the return value depends on #' the value of `penalty`. When using `predict()`, only a single #' value of the penalty can be used. When predicting on multiple @@ -137,6 +137,9 @@ print.logistic_reg <- function(x, ...) { invisible(x) } +#' @export +translate.logistic_reg <- translate.linear_reg + # ------------------------------------------------------------------------------ #' @inheritParams update.boost_tree @@ -235,7 +238,7 @@ organize_glmnet_prob <- function(x, object) { } # ------------------------------------------------------------------------------ -# glmnet call stack for linear regression using `predict` when object has +# glmnet call stack for logistic regression using `predict` when object has # classes "_lognet" and "model_fit" (for class predictions): # # predict() @@ -247,7 +250,7 @@ organize_glmnet_prob <- function(x, object) { # predict.lognet() -# glmnet call stack for linear regression using `multi_predict` when object has +# glmnet call stack for logistic regression using `multi_predict` when object has # classes "_lognet" and "model_fit" (for class predictions): # # multi_predict() @@ -262,10 +265,15 @@ organize_glmnet_prob <- function(x, object) { # ------------------------------------------------------------------------------ #' @export -predict._lognet <- function (object, new_data, type = NULL, opts = list(), penalty = NULL, multi = FALSE, ...) { +predict._lognet <- function(object, new_data, type = NULL, opts = list(), penalty = NULL, multi = FALSE, ...) { if (any(names(enquos(...)) == "newdata")) stop("Did you mean to use `new_data` instead of `newdata`?", call. = FALSE) + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (is.null(penalty) & !is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } + object$spec$args$penalty <- check_penalty(penalty, object, multi) object$spec <- eval_args(object$spec) @@ -286,8 +294,16 @@ multi_predict._lognet <- penalty <- eval_tidy(penalty) dots <- list(...) - if (is.null(penalty)) - penalty <- eval_tidy(object$fit$lambda) + + if (is.null(penalty)) { + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (!is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } else { + penalty <- object$fit$lambda + } + } + dots$s <- penalty if (is.null(type)) @@ -330,7 +346,7 @@ multi_predict._lognet <- #' @export -predict_class._lognet <- function (object, new_data, ...) { +predict_class._lognet <- function(object, new_data, ...) { if (any(names(enquos(...)) == "newdata")) stop("Did you mean to use `new_data` instead of `newdata`?", call. = FALSE) @@ -339,7 +355,7 @@ predict_class._lognet <- function (object, new_data, ...) { } #' @export -predict_classprob._lognet <- function (object, new_data, ...) { +predict_classprob._lognet <- function(object, new_data, ...) { if (any(names(enquos(...)) == "newdata")) stop("Did you mean to use `new_data` instead of `newdata`?", call. = FALSE) @@ -348,7 +364,7 @@ predict_classprob._lognet <- function (object, new_data, ...) { } #' @export -predict_raw._lognet <- function (object, new_data, opts = list(), ...) { +predict_raw._lognet <- function(object, new_data, opts = list(), ...) { if (any(names(enquos(...)) == "newdata")) stop("Did you mean to use `new_data` instead of `newdata`?", call. = FALSE) @@ -356,3 +372,10 @@ predict_raw._lognet <- function (object, new_data, opts = list(), ...) { predict_raw.model_fit(object, new_data = new_data, opts = opts, ...) } + +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.logistic_reg +#' @rdname min_grid +min_grid.logistic_reg <- min_grid.linear_reg diff --git a/R/logistic_reg_data.R b/R/logistic_reg_data.R index d04f0100e..02b13f398 100644 --- a/R/logistic_reg_data.R +++ b/R/logistic_reg_data.R @@ -218,7 +218,7 @@ set_model_arg( parsnip = "penalty", original = "reg_param", func = list(pkg = "dials", fun = "penalty"), - has_submodel = TRUE + has_submodel = FALSE ) set_model_arg( diff --git a/R/mars.R b/R/mars.R index f83e56e29..22884852d 100644 --- a/R/mars.R +++ b/R/mars.R @@ -262,3 +262,36 @@ earth_by_terms <- function(num_terms, object, new_data, type, ...) { pred[[".row"]] <- 1:nrow(new_data) pred[, c(".row", "num_terms", nms)] } + +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.mars +#' @rdname min_grid +min_grid.mars <- function(x, grid, ...) { + + grid_names <- names(grid) + param_info <- get_submodel_info(x, grid) + + if (!any(param_info$has_submodel)) { + return(blank_submodels(grid)) + } + + fixed_args <- get_fixed_args(param_info) + + fit_only <- + grid %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(num_terms = max(num_terms, na.rm = TRUE)) %>% + dplyr::ungroup() + + min_grid_df <- + dplyr::full_join(fit_only %>% rename(max_terms = num_terms), grid, by = fixed_args) %>% + dplyr::filter(num_terms != max_terms) %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(.submodels = list(list(num_terms = num_terms))) %>% + dplyr::ungroup() %>% + dplyr::full_join(fit_only, grid, by = fixed_args) + + min_grid_df %>% dplyr::select(dplyr::one_of(grid_names), .submodels) +} diff --git a/R/multinom_reg.R b/R/multinom_reg.R index b8bc0a479..9b6a6e768 100644 --- a/R/multinom_reg.R +++ b/R/multinom_reg.R @@ -57,9 +57,9 @@ #' #' \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::multinom_reg(), "keras")} #' -#' When using `glmnet` models, there is the option to pass -#' multiple values (or no values) to the `penalty` argument. This -#' can have an effect on the model object results. When using the +#' For `glmnet` models, the full regularization path is always fit regardless +#' of the value given to `penalty`. Also, there is the option to pass +#' multiple values (or no values) to the `penalty` argument. When using the #' `predict()` method in these cases, the return value depends on #' the value of `penalty`. When using `predict()`, only a single #' value of the penalty can be used. When predicting on multiple @@ -112,7 +112,7 @@ print.multinom_reg <- function(x, ...) { cat("Multinomial Regression Model Specification (", x$mode, ")\n\n", sep = "") model_printer(x, ...) - if(!is.null(x$method$fit$args)) { + if (!is.null(x$method$fit$args)) { cat("Model fit template:\n") print(show_call(x)) } @@ -120,6 +120,9 @@ print.multinom_reg <- function(x, ...) { invisible(x) } +#' @export +translate.multinom_reg <- translate.linear_reg + # ------------------------------------------------------------------------------ #' @inheritParams update.boost_tree @@ -188,7 +191,7 @@ organize_multnet_prob <- function(x, object) { } # ------------------------------------------------------------------------------ -# glmnet call stack for linear regression using `predict` when object has +# glmnet call stack for multinomial regression using `predict` when object has # classes "_multnet" and "model_fit" (for class predictions): # # predict() @@ -199,7 +202,7 @@ organize_multnet_prob <- function(x, object) { # predict.multnet() -# glmnet call stack for linear regression using `multi_predict` when object has +# glmnet call stack for multinomial regression using `multi_predict` when object has # classes "_multnet" and "model_fit" (for class predictions): # # multi_predict() @@ -217,6 +220,11 @@ organize_multnet_prob <- function(x, object) { predict._multnet <- function(object, new_data, type = NULL, opts = list(), penalty = NULL, multi = FALSE, ...) { + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (is.null(penalty) & !is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } + object$spec$args$penalty <- check_penalty(penalty, object, multi) object$spec <- eval_args(object$spec) @@ -242,14 +250,20 @@ multi_predict._multnet <- penalty <- eval_tidy(penalty) dots <- list(...) - if (is.null(penalty)) - penalty <- eval_tidy(object$fit$lambda) + if (is.null(penalty)) { + # See discussion in https://github.com/tidymodels/parsnip/issues/195 + if (!is.null(object$spec$args$penalty)) { + penalty <- object$spec$args$penalty + } else { + penalty <- object$fit$lambda + } + } dots$s <- penalty if (is.null(type)) type <- "class" if (!(type %in% c("class", "prob", "link", "raw"))) { - stop ("`type` should be either 'class', 'link', 'raw', or 'prob'.", call. = FALSE) + stop("`type` should be either 'class', 'link', 'raw', or 'prob'.", call. = FALSE) } if (type == "prob") dots$type <- "response" @@ -290,19 +304,19 @@ multi_predict._multnet <- } #' @export -predict_class._multnet <- function (object, new_data, ...) { +predict_class._multnet <- function(object, new_data, ...) { object$spec <- eval_args(object$spec) predict_class.model_fit(object, new_data = new_data, ...) } #' @export -predict_classprob._multnet <- function (object, new_data, ...) { +predict_classprob._multnet <- function(object, new_data, ...) { object$spec <- eval_args(object$spec) predict_classprob.model_fit(object, new_data = new_data, ...) } #' @export -predict_raw._multnet <- function (object, new_data, opts = list(), ...) { +predict_raw._multnet <- function(object, new_data, opts = list(), ...) { object$spec <- eval_args(object$spec) predict_raw.model_fit(object, new_data = new_data, opts = opts, ...) } @@ -323,3 +337,10 @@ check_glmnet_lambda <- function(dat, object) { dat } + +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.multinom_reg +#' @rdname min_grid +min_grid.multinom_reg <- min_grid.linear_reg diff --git a/R/multinom_reg_data.R b/R/multinom_reg_data.R index 7b7b16b3b..777126a01 100644 --- a/R/multinom_reg_data.R +++ b/R/multinom_reg_data.R @@ -44,7 +44,7 @@ set_pred( mode = "classification", type = "class", value = list( - pre = check_glmnet_lambda, + pre = NULL, post = organize_multnet_class, func = c(fun = "predict"), args = @@ -104,7 +104,7 @@ set_model_arg( parsnip = "penalty", original = "reg_param", func = list(pkg = "dials", fun = "penalty"), - has_submodel = TRUE + has_submodel = FALSE ) set_model_arg( diff --git a/R/nearest_neighbor.R b/R/nearest_neighbor.R index b28a3df46..6e29cf45d 100644 --- a/R/nearest_neighbor.R +++ b/R/nearest_neighbor.R @@ -213,8 +213,43 @@ multi_predict._train.kknn <- } knn_by_k <- function(k, object, new_data, type, ...) { - object$fit$call$ks <- k + object$fit$best.parameters$k <- k + predict(object, new_data = new_data, type = type, ...) %>% dplyr::mutate(neighbors = k, .row = dplyr::row_number()) %>% dplyr::select(.row, neighbors, dplyr::starts_with(".pred")) } + +# ------------------------------------------------------------------------------ + +#' @export +#' @export min_grid.nearest_neighbor +#' @rdname min_grid +min_grid.nearest_neighbor <- function(x, grid, ...) { + + grid_names <- names(grid) + param_info <- get_submodel_info(x, grid) + + if (!any(param_info$has_submodel)) { + return(blank_submodels(grid)) + } + + fixed_args <- get_fixed_args(param_info) + + fit_only <- + grid %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(neighbors = max(neighbors, na.rm = TRUE)) %>% + dplyr::ungroup() + + min_grid_df <- + dplyr::full_join(fit_only %>% rename(max_neighbor = neighbors), grid, by = fixed_args) %>% + dplyr::filter(neighbors != max_neighbor) %>% + dplyr::rename(sub_neighbors = neighbors, neighbors = max_neighbor) %>% + dplyr::group_by(!!!rlang::syms(fixed_args)) %>% + dplyr::summarize(.submodels = list(list(neighbors = sub_neighbors))) %>% + dplyr::ungroup() %>% + dplyr::full_join(fit_only, grid, by = fixed_args) + + min_grid_df %>% dplyr::select(dplyr::one_of(grid_names), .submodels) +} diff --git a/_pkgdown.yml b/_pkgdown.yml index 644889998..69092b9e4 100644 --- a/_pkgdown.yml +++ b/_pkgdown.yml @@ -81,6 +81,8 @@ navbar: href: articles/articles/Classification.html - text: Making a parsnip model from scratch href: articles/articles/Scratch.html + - text: Evaluating submodels with the same model object + href: articles/articles/Submodels.html - text: News href: news/index.html - text: Reference diff --git a/docs/404.html b/docs/404.html index b120a1d2f..5ec6cfca1 100644 --- a/docs/404.html +++ b/docs/404.html @@ -52,6 +52,7 @@ gtag('config', 'UA-115082821-1'); + @@ -71,7 +72,7 @@ parsnip
part of tidymodels - 0.0.2 + 0.0.3
@@ -99,6 +100,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -115,6 +119,7 @@ +
    @@ -130,6 +135,7 @@

    Page not found (404)

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -141,11 +147,14 @@

    Page not found (404)

    Site built by pkgdown.

    +
    + + diff --git a/docs/articles/articles/Classification.html b/docs/articles/articles/Classification.html index f5e50237a..2ec6a523f 100644 --- a/docs/articles/articles/Classification.html +++ b/docs/articles/articles/Classification.html @@ -40,7 +40,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -68,6 +68,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -86,6 +89,7 @@ +
    @@ -100,28 +104,26 @@

    Classification Example

    To demonstrate parsnip for classification models, the credit data will be used.

    -
    library(tidymodels)
-#> Registered S3 method overwritten by 'xts':
-#>   method     from
-#>   as.zoo.xts zoo
-#> ── Attaching packages ───────────────────────────────── tidymodels 0.0.2 ──
-#> ✔ broom     0.5.1       ✔ purrr     0.3.2  
-#> ✔ dials     0.0.2       ✔ recipes   0.1.4  
-#> ✔ dplyr     0.8.0.1     ✔ rsample   0.0.4  
-#> ✔ infer     0.4.0       ✔ yardstick 0.0.2
-#> ── Conflicts ──────────────────────────────────── tidymodels_conflicts() ──
-#> ✖ purrr::discard() masks scales::discard()
-#> ✖ dplyr::filter()  masks stats::filter()
-#> ✖ dplyr::lag()     masks stats::lag()
-#> ✖ recipes::step()  masks stats::step()
+
+set.seed(7075)
+data_split <- initial_split(credit_data, strata = "Status", p = 0.75)
+
+credit_train <- training(data_split)
+credit_test  <- testing(data_split)

    A single hidden layer neural network will be used to predict a person’s credit status. To do so, the columns of the predictor matrix should be numeric and on a common scale. recipes will be used to do so.

    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_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:.

    +#> bad 191 96 +#> good 122 704
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -206,10 +210,12 @@

    Classification Example

    Site built by pkgdown.

    + + diff --git a/docs/articles/articles/Models.html b/docs/articles/articles/Models.html index 0588e29fa..0e60299a4 100644 --- a/docs/articles/articles/Models.html +++ b/docs/articles/articles/Models.html @@ -40,7 +40,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -68,6 +68,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -86,6 +89,7 @@ +
    @@ -107,16 +111,16 @@

    List of Models

    + - - - - + + - - - - - - @@ -210,21 +200,18 @@

    List of Models

    × - - + + + + - - - - - + + - - - @@ -366,21 +354,21 @@

    List of Models

    ✔ + + - - - + + + + + + - + - + + - + + + + + - + - + + + + + + + + + + + + + @@ -809,6 +868,64 @@

    List of Models

    × + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + @@ -838,23 +1016,113 @@

    List of Models

    × + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + @@ -991,6 +1512,7 @@

    List of Models

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -1001,10 +1523,12 @@

    List of Models

    Site built by pkgdown.

    +
    + diff --git a/docs/articles/articles/Regression.html b/docs/articles/articles/Regression.html index 518585c0e..f11bf85da 100644 --- a/docs/articles/articles/Regression.html +++ b/docs/articles/articles/Regression.html @@ -40,7 +40,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -68,6 +68,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -86,6 +89,7 @@ +
    @@ -100,51 +104,48 @@

    Regression Example

    The Ames housing data will be used to to demonstrate how regression models can be made using parsnip. We’ll create the data set and create a simple training/test set split:

    -
    library(AmesHousing)
-ames <- make_ames()
+
    library(AmesHousing)
+ames <- make_ames()
 
-library(tidymodels)
-#> Registered S3 method overwritten by 'xts':
-#>   method     from
-#>   as.zoo.xts zoo
-#> ── Attaching packages ───────────────────────────────── tidymodels 0.0.2 ──
-#> ✔ broom     0.5.1       ✔ recipes   0.1.4  
-#> ✔ dials     0.0.2       ✔ rsample   0.0.4  
-#> ✔ dplyr     0.8.0.1     ✔ tibble    2.1.1  
-#> ✔ infer     0.4.0       ✔ yardstick 0.0.2  
-#> ✔ purrr     0.3.2
-#> ── Conflicts ──────────────────────────────────── tidymodels_conflicts() ──
-#> ✖ dplyr::combine() masks randomForest::combine()
-#> ✖ purrr::discard() masks scales::discard()
-#> ✖ dplyr::filter()  masks stats::filter()
-#> ✖ dplyr::lag()     masks stats::lag()
-#> ✖ dials::margin()  masks ggplot2::margin(), randomForest::margin()
-#> ✖ recipes::step()  masks stats::step()
+library(tidymodels)
+#> ── Attaching packages ─────────────────────────────────────── tidymodels 0.0.2 ──
+#> ✔ broom     0.5.1          ✔ recipes   0.1.6     
+#> ✔ dials     0.0.2.9000     ✔ rsample   0.0.5     
+#> ✔ dplyr     0.8.3          ✔ tibble    2.1.3     
+#> ✔ infer     0.4.0          ✔ yardstick 0.0.3.9000
+#> ✔ purrr     0.3.2
+#> ── Conflicts ────────────────────────────────────────── tidymodels_conflicts() ──
+#> ✖ dplyr::combine() masks randomForest::combine()
+#> ✖ purrr::discard() masks scales::discard()
+#> ✖ dplyr::filter()  masks stats::filter()
+#> ✖ dplyr::lag()     masks stats::lag()
+#> ✖ dials::margin()  masks ggplot2::margin(), randomForest::margin()
+#> ✖ recipes::step()  masks stats::step()
+
+set.seed(4595)
+data_split <- initial_split(ames, strata = "Sale_Price", p = 0.75)
 
-set.seed(4595)
-data_split <- initial_split(ames, strata = "Sale_Price", p = 0.75)
-
-ames_train <- training(data_split)
-ames_test  <- testing(data_split)
    
+ames_train <- training(data_split)
+ames_test  <- testing(data_split)

    Random Forests

    We’ll start by fitting a random forest model to a small set of parameters. Let’s say that the model would include predictors: Longitude, Latitude, Lot_Area, Neighborhood, and Year_Sold. A simple random forest model can be specified via

    -
    library(parsnip)
+
-
    The model will be fit with the ranger package. Since we didn’t add any extra arguments to fit, many of the arguments will be set to their defaults from the specific function that is used by ranger::ranger. The help pages for the model function describes the changes to the default parameters that are made and the translate function can also be used.
    
+
    The model will be fit with the ranger package. Since we didn’t add any extra arguments to fit, many of the arguments will be set to their defaults from the specific function that is used by ranger::ranger. The help pages for the model function describes the changes to the default parameters that are made and the translate function can also be used.
    
 
    parsnip gives two different interfaces to the models: the formula and non-formula interfaces. Let’s start with the non-formula interface:
    
-
 
    The non-formula interface doesn’t do anything to the predictors before giving it to the underlying model function. This particular model does not require indicator variables to be create prior to the model (note that the output shows “Number of independent variables: 5”).
    
 
    For regression models, the basic predict method can be used and returns a tibble with a column named .pred:
    
 
+#> 1 rmse    standard      0.0916
+#> 2 rsq     standard      0.715 
+#> 3 mae     standard      0.0663

    Note that:

    Suppose that there was some feature in the randomForest package that we’d like to evaluate. To do so, the only part of the syntaxt that needs to change is the set_engine argument:

    rand_forest(mode = "regression", mtry = 3, trees = 1000) %>%
   set_engine("randomForest") %>%
-  fit(
-    log10(Sale_Price) ~ Longitude + Latitude + Lot_Area + Neighborhood + Year_Sold,
+  fit(
+    log10(Sale_Price) ~ Longitude + Latitude + Lot_Area + Neighborhood + Year_Sold,
     data = ames_train
   )
 #> parsnip model object
@@ -241,10 +242,10 @@ 
    
 #>                      Number of trees: 1000
 #> No. of variables tried at each split: 3
 #> 
-#>           Mean of squared residuals: 0.0127
-#>                     % Var explained: 58.6
    +#> Mean of squared residuals: 0.013 +#> % Var explained: 59.4

    Look at the formula code that was printed out, one function uses the argument name ntree and the other uses num.trees. parsnip doesn’t require you to know the specific names of the main arguments.

    -

    Now suppose that we want to modify the value of mtry based on the number of predictors in the data. Usually, the default value would be floor(sqrt(num_predictors)). To use a pure bagging model would require an mtry value equal to the total number of parameters. There may be cases where you may not know how many predictors are going to be present (perhaps due to the generation of indicator variables or a variable filter) so that might be difficult to know exactly.

    +

    Now suppose that we want to modify the value of mtry based on the number of predictors in the data. Usually, the default value would be floor(sqrt(num_predictors)). To use a pure bagging model would require an mtry value equal to the total number of parameters. There may be cases where you may not know how many predictors are going to be present (perhaps due to the generation of indicator variables or a variable filter) so that might be difficult to know exactly.

    When the model it being fit by parsnip, data descriptors are made available. These attempt to let you know what you will have available when the model is fit. When a model object is created (say using rand_forest), the values of the arguments that you give it are immediately evaluated… unless you delay them. To delay the evaluation of any argument, you can used rlang::expr to make an expression.

    Two relevant descriptors for what we are about to do are:

    +#> Df %Dev Lambda +#> [1,] 0 0.0000 0.137000 +#> [2,] 1 0.0188 0.125000 +#> [3,] 1 0.0355 0.114000 +#> [4,] 1 0.0504 0.104000 +#> [5,] 2 0.0680 0.094600 +#> [6,] 4 0.0933 0.086200 +#> [7,] 5 0.1250 0.078500 +#> [8,] 5 0.1530 0.071600 +#> [9,] 7 0.1840 0.065200 +#> [10,] 7 0.2140 0.059400 +#> [11,] 7 0.2400 0.054100 +#> [12,] 8 0.2620 0.049300 +#> [13,] 8 0.2860 0.044900 +#> [14,] 8 0.3060 0.040900 +#> [15,] 8 0.3230 0.037300 +#> [16,] 8 0.3380 0.034000 +#> [17,] 8 0.3500 0.031000 +#> [18,] 8 0.3610 0.028200 +#> [19,] 9 0.3700 0.025700 +#> [20,] 9 0.3790 0.023400 +#> [21,] 9 0.3860 0.021300 +#> [22,] 9 0.3920 0.019500 +#> [23,] 9 0.3970 0.017700 +#> [24,] 9 0.4010 0.016100 +#> [25,] 9 0.4050 0.014700 +#> [26,] 9 0.4080 0.013400 +#> [27,] 10 0.4100 0.012200 +#> [28,] 11 0.4130 0.011100 +#> [29,] 11 0.4150 0.010100 +#> [30,] 11 0.4170 0.009240 +#> [31,] 12 0.4180 0.008420 +#> [32,] 12 0.4200 0.007670 +#> [33,] 12 0.4210 0.006990 +#> [34,] 12 0.4220 0.006370 +#> [35,] 12 0.4230 0.005800 +#> [36,] 12 0.4230 0.005290 +#> [37,] 12 0.4240 0.004820 +#> [38,] 12 0.4250 0.004390 +#> [39,] 12 0.4250 0.004000 +#> [40,] 12 0.4250 0.003650 +#> [41,] 12 0.4260 0.003320 +#> [42,] 12 0.4260 0.003030 +#> [43,] 12 0.4260 0.002760 +#> [44,] 12 0.4260 0.002510 +#> [45,] 12 0.4260 0.002290 +#> [46,] 12 0.4260 0.002090 +#> [47,] 12 0.4270 0.001900 +#> [48,] 12 0.4270 0.001730 +#> [49,] 12 0.4270 0.001580 +#> [50,] 12 0.4270 0.001440 +#> [51,] 12 0.4270 0.001310 +#> [52,] 12 0.4270 0.001190 +#> [53,] 12 0.4270 0.001090 +#> [54,] 12 0.4270 0.000991 +#> [55,] 12 0.4270 0.000903 +#> [56,] 12 0.4270 0.000823 +#> [57,] 12 0.4270 0.000750 +#> [58,] 12 0.4270 0.000683 +#> [59,] 12 0.4270 0.000622 +#> [60,] 12 0.4270 0.000567 +#> [61,] 12 0.4270 0.000517 +#> [62,] 12 0.4270 0.000471 +#> [63,] 12 0.4270 0.000429 +#> [64,] 12 0.4270 0.000391 +#> [65,] 12 0.4270 0.000356

    If penalty were not specified, all of the lambda values would be computed.

    To get the predictions for this specific value of lambda (aka penalty):

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -382,10 +448,12 @@

    Site built by pkgdown.

    +

    + diff --git a/docs/articles/articles/Regression_files/figure-html/glmn-pred-1.png b/docs/articles/articles/Regression_files/figure-html/glmn-pred-1.png index 12f0f8e85..6d6cfccf9 100644 Binary files a/docs/articles/articles/Regression_files/figure-html/glmn-pred-1.png and b/docs/articles/articles/Regression_files/figure-html/glmn-pred-1.png differ diff --git a/docs/articles/articles/Scratch.html b/docs/articles/articles/Scratch.html index f6a9391fc..8dd8fd039 100644 --- a/docs/articles/articles/Scratch.html +++ b/docs/articles/articles/Scratch.html @@ -5,12 +5,12 @@ -Making a <code>parsnip</code> model from scratch • parsnip +Making a parsnip model from scratch • parsnip - + +
    -

    Making a parsnip model from scratch

    +

    Making a parsnip model from scratch

    @@ -104,392 +108,433 @@

    Making a parsnip model from scratch

  • It eliminates a lot of duplicate code.
  • Since the expressions are not evaluated until fitting, it eliminates a large amount of package dependencies.
    • -

    A parsnip model function is itself very general. For example, the logistic_reg function itself doesn’t have any model code within it. Instead, each model function is associated with one or more computational engines. These might be different R packages or some function in another language (that can be evaluated by R).

    +

    A parsnip model function is itself very general. For example, the logistic_reg() function itself doesn’t have any model code within it. Instead, each model function is associated with one or more computational engines. These might be different R packages or some function in another language (that can be evaluated by R).

    This vignette describes the process of creating a new model function. Before proceeding, take a minute and read our guidelines on creating modeling packages to get the general themes and conventions that we use.

    -

    As an example, we’ll create a function for mixture discriminant analysis. There are a few packages that do this but we’ll focus on mda::mda:

    -
    str(mda::mda)
+
    As an example, we’ll create a function for mixture discriminant analysis. There are a few packages that do this but we’ll focus on mda::mda:
    
+
 
    The main hyper-parameter is the number of subclasses. We’ll name our function mixture_da.
    
-
    
+
    
+
    
+Aspects of Models
    
+
    Before proceeding, it helps to to review how parsnip categorizes models:
    
+
      
+
    • The model type is related to the structural aspect of the model. For example, the model type linear_reg represents linear models (slopes and intercepts) that model a numeric outcome. Other model types in the package are neighest_nighbors, decision_tree`, and so on.
      • 
+
    • Within a model type is the mode. This relates to the modeling goal. Currently the two modes in the package are “regression” and “classification”. Some models have methods for both models (e.g. nearest neighbors) while others are specific to a single mode (e.g. logistic regression).
      • 
+
    • The computation engine is a combination of the estimation method and the implementation. For example, for linear regression, one model is "lm" and this uses ordinal least squares analysis using the lm package. Another engine is "stan" which uses the Stan infrastructure to estimate parameters using Bayes rule.
      • 
+
    
+
    When adding a model into parsnip, the user has to specific which modes and engines are used. The package also enables users to add a new mode or engine to an existing model.
    
+
    
+
    
 
    
-Step 1. Make the objects for the general method
    
-
    There are three objects that define the parameters and other characteristics of the model function.
    
-
    First, is the object that describes the model’s mode(s). The modes are the type of model and the two main values are “classification” and “regression”. A third mode, “unknown”, is used for initializing objects but models will fail if it is used further.
    
-
    The convention in parsnip is to use the name {model name}_modes. In our case, we have:
    
-
    mixture_da_modes <- c("classification", "unknown")
    
-
    Next, we define the engines used by the model and the associated mode. Here, the columns correspond to the engine names and rows are the modes (via row names). We have two engines and one effective mode, so our object will have the suffix _engines:
    
-
-
    A row for “unknown” modes is not needed in this object.
    
-
    Now, we enumerate the main arguments for each engine. parsnip standardizes the names of arguments across different models and engines. For example, random forest and boosting use multiple trees to create the ensemble. Instead of using different argument names, parsnip standardizes on trees and the underlying code translates to the actual arguments used by the different functions.
    
-
    In our case, the MDA argument name will be “sub_classes”.
    
-
    Here, the object name will have the suffix _arg_key and will have columns for the engines and rows for the arguments. The entries for the data frame are the actual arguments for each engine (and is NA when an engine doesn’t have that argument). Ours:
    
-
-
    As an example of a model with multiple engines, here is the object for logistic regression:
    
-
-
    The internals of parsnip will use these objects during the creation of the model code.
    
+The General Process
+
    parsnip stores information about the models in an internal environment object. The environment can be accessed via the function get_model_env(). The package includes a variety of functions that can get or set the different aspects of the models.
    
+
    If you are adding a new model form your own package, you can use these functions to add new entries into the model environment.
    
+
    
+
    
+
    
+Step 1. Register the Model, Modes, and Arguments.
    
+
    We will add the MDA model using the model type mixture_da. Since this is a classification method, we only have to register a single mode:
    
+
    library(parsnip)
+set_new_model("mixture_da")
+set_model_mode(model = "mixture_da", mode = "classification")
+set_model_engine(
+  "mixture_da", 
+  mode = "classification", 
+  eng = "mda"
+)
    
+
    These functions should silently finish. There is also a function that can be used to show what aspects of the model have been added to parsnip:
    
+
+
    The next step would be the declare the main arguments to the model. These are declared indpendnent of the mode. To specifiy the argument, there are a few slots to fill in:
    
+
      
+
    • The name that parsnip uses for the argument. In general, we try to use non-jargony names for arguments (e.g. “penalty” instead of “lambda” for regularised regression). We recommend consulting this page to see if an existing argument name can be used before creating a new one.
      • 
+
    • The argument name that is used by the underlying modeling function.
      • 
+
    • A function reference for a contructor that will be used to generate tuning parameter values. This should be a character vector that has a named element called fun that is the contructor function. There is an optional element pkg that can be used to call the function using its namespace.
      • 
+
    • A logical value for wether the argument can be used to generate mutiple predictions for a single R object. For example, for boosted trees, if a model is fit with 10 boosting iterations, many modeling packages allow the model object to make predictions for any iterations less than the one used to fit the model. In general this is not the case so one would use has_submodels = FALSE.
      • 
+
    
+
    For mda::mda(), the main tuning parameter is subclasses which we will rewrite as sub_classes.
    
+
 
    
-
    
+
    
 
    
-Step 2. Create the model function
    
+Step 3. Create the model function
 
    This is a fairly simple function that can follow a basic template. The main arguments to our function will be:
    
 
      
-
    • The mode. If the model can do more than one mode, you might default this to “unknown”. In our case, since it is only a classification model, it makes sense to default it to that mode.
      • 
-
    • The argument names (sub_classes here). These should be defaulted to NULL.
      • 
-
    • 
-... are not used in the main model function.
      • 
+
    • The mode. If the model can do more than one mode, you might default this to “unknown”. In our case, since it is only a classification model, it makes sense to default it to that mode so that the users won’t have to specify it.
      • 
+
    • The argument names (sub_classes here). These should be defaulted to NULL.
      • 
 
    
 
    A basic version of the function is:
    
-
+
 
    This is pretty simple since the data are not exposed to this function.
    
 
    
-
    
+
    
 
    
-Step 3. Make the model object
    
-
    This is where the details of the models are specified. This will be a list that has a few different elements:
    
+Step 3. Add a Fit Module
+
    Now that parsnip knows about the model, mode, and engine, we can give it the infomration on fitting the model for our engine. In infomration needed to fit the model is contained in another list. The elements are:
    
 
      
-
    • 
-libs is a character string that has any package names that will be required for the model fit.
      • 
-
    • 
-fit has details for the model fit function.
      • 
-
    • 
-pred, prob, and classes. These are lists of details for making predictions on numbers, class probabilities, or hard class predictions (respectively).
      • 
-
    
-
    We’ll look at each. The convention here is to name this {model name}_{engine}_data. We’ll start with:
    
-
-
    
-
    
-The fit module
    
-
    The main arguments are:
    
-
      
-
    • 
-interface a single character value that could be “formula”, “data.frame”, or “matrix”. This defines the type of interface used by the underlying fit function (mda::mda, in this case). This helps the translation of the data to be in an appropriate format for the that function.
      • 
-
    • 
-protect is an optional list of function arguments that should not be changeable by the user. In this case, we probably don’t want users to pass data values to these arguments (until the fit function is called).
      • 
-
    • 
-func is the package and name of the function that will be called. If you are using a locally defined function, only fun is required.
      • 
-
    • 
-defaults is an optional list of arguments to the fit function that the user can change, but whose defaults can be set here. This isn’t needed in this case, but is describe later in this document.
      • 
+
    • interface a single character value that could be “formula”, “data.frame”, or “matrix”. This defines the type of interface used by the underlying fit function (mda::mda, in this case). This helps the translation of the data to be in an appropriate format for the that function.
      • 
+
    • protect is an optional list of function arguments that should not be changeable by the user. In this case, we probably don’t want users to pass data values to these arguments (until the fit function is called).
      • 
+
    • func is the package and name of the function that will be called. If you are using a locally defined function, only fun is required.
      • 
+
    • defaults is an optional list of arguments to the fit function that the user can change, but whose defaults can be set here. This isn’t needed in this case, but is describe later in this document.
      • 
 
    
 
    For the first engine:
    
-
    mixture_da_mda_data$fit <-
-  list(
-    interface = "formula",
-    protect = c("formula", "data", "weights"),
-    func = c(pkg = "mda", fun = "mda"),
-    defaults = list()
-  )
    
-
    
-
    
-
    
-The numeric module
    
-
    This is defined only for regression models (so is not added to the list). The convention used here is very similar to the two that are detailed in the next section. For numeric, the model requires an unnamed numeric vector output (usually).
    
-
    Examples are here and here.
    
-
    For multivariate models, the return value should be a matrix or data frame (otherwise a vector should be the results).
    
-
    Note that the numeric module maps to the predict_numeric function in parsnip. However, the user-facing predict function is used to generate predictions and returns a tibble with a column named .pred (see the example below). When creating new models, you don’t have to write code for that part.
    
+
 
    
-
    
-
    
-The class module
    
-
    To make hard class predictions, the class object contains the details. The elements of the list are:
    
+
    
+
    
+Step 3. Add Modules for Prediction
    
+
    Similar to the fitting module, we specify the code for making different types of predictions. To make hard class predictions, the class object contains the details. The elements of the list are:
    
 
      
 
    • 
 pre and post are optional functions that can preprocess the data being fed to the prediction code and to postprocess the raw output of the predictions. These won’t be need for this example, but a section below has examples of how these can be used when the model code is not easy to use. If the data being predicted has a simple type requirement, you can avoid using a pre function with the args below.
      • 
 
    • 
 func is the prediction function (in the same format as above). In many cases, packages have a predict method for their model’s class but this is typically not exported. In this case (and the example below), it is simple enough to make a generic call to predict with no associated package.
      • 
 
    • 
-args is a list of arguments to pass to the prediction function. These will mostly likely be wrapped in rlang::expr so that they are not evaluated when defining the method. For mda, the code would be predict(object, newdata, type = "class"). What is actually given to the function is the parsnip model fit object, which includes a sub-object called fit and this houses the mda model object. If the data need to be a matrix or data frame, you could also use new_data = quote(as.data.frame(new_data)) and so on.
      • 
+args is a list of arguments to pass to the prediction function. These will mostly likely be wrapped in rlang::expr so that they are not evaluated when defining the method. For mda, the code would be predict(object, newdata, type = "class"). What is actually given to the function is the parsnip model fit object, which includes a sub-object called fit and this houses the mda model object. If the data need to be a matrix or data frame, you could also use newdata = quote(as.data.frame(newdata)) and so on.
 
    
-
-
    The predict_class function will expect the result to be an unnamed character string or factor. This will be coerced to a factor with the same levels as the original data. As with the pred module, the user doesn’t call predict_class but uses predict instead and this produces a tibble with a column named .pred_class per the model guidlines.
    
-
    
-
    
-
    
-The classprob module
    
-
    This defines the class probabilities (if they can be computed). The format is identical to the class module but the output is expected to be a tibble with columns for each factor level.
    
+
    The parsnip prediction code will expect the result to be an unnamed character string or factor. This will be coerced to a factor with the same levels as the original data.
    
+
    To add this method to the model environment, a similar set function is used:
    
+
+
    A similar call can be used to define the class probability module (if they can be computed). The format is identical to the class module but the output is expected to be a tibble with columns for each factor level.
    
 
    As an example of the post function, the data frame created by mda:::predict.mda will be converted to a tibble. The arguments are x (the raw results coming from the predict method) and object (the parsnip model fit object). The latter has a sub-object called lvl which is a character string of the outcome’s factor levels (if any).
    
-
-
    The post element converts the output to a tibble but the main predict method does proper naming of the column names.
    
-
    
+
    We register the probability module. There is a template function that makes this slightly easier to format the objects:
    
+
+
    If this model could be used for regression situations, we could also add a “numeric” module. The convention used here is very similar to the two that are detailed in the next section. For pred, the model requires an unnamed numeric vector output (usually).
    
+
    Examples are here and here.
    
 
    
 
    
 
    
 Does it Work?
    
 
    As a developer, one thing that may come in handy is the translate function. This will tell you what the model’s eventual syntax will be.
    
 
    For example:
    
-
+
 
    Let’s try it on the iris data:
    
-
    set.seed(4622)
-iris_split <- initial_split(iris, prop = 0.90)
-iris_train <- training(iris_split)
-iris_test  <-  testing(iris_split)
-
-mda_spec <- mixture_da(sub_classes = 2)
-
-mda_fit <- mda_spec %>%
-  set_engine("mda") %>%
-  fit(Species ~ ., data = iris_train)
-mda_fit
-#> parsnip model object
-#> 
-#> Call:
-#> mda::mda(formula = formula, data = data, sub_classes = ~2)
-#> 
-#> Dimension: 4 
-#> 
-#> Percent Between-Group Variance Explained:
-#>    v1    v2    v3    v4 
-#>  95.8  98.3  99.9 100.0 
-#> 
-#> Degrees of Freedom (per dimension): 5 
-#> 
-#> Training Misclassification Error: 0.0221 ( N = 136 )
-#> 
-#> Deviance: 13.4
-
-predict(mda_fit, new_data = iris_test) %>%
-  bind_cols(iris_test %>% select(Species))
-#> # A tibble: 14 x 2
-#>    .pred_class Species   
-#>    <fct>       <fct>     
-#>  1 setosa      setosa    
-#>  2 setosa      setosa    
-#>  3 setosa      setosa    
-#>  4 setosa      setosa    
-#>  5 versicolor  versicolor
-#>  6 versicolor  versicolor
-#>  7 versicolor  versicolor
-#>  8 versicolor  versicolor
-#>  9 versicolor  versicolor
-#> 10 virginica   virginica 
-#> 11 virginica   virginica 
-#> 12 virginica   virginica 
-#> 13 virginica   virginica 
-#> 14 virginica   virginica
-
-predict(mda_fit, new_data = iris_test, type = "prob") %>% 
-  bind_cols(iris_test %>% select(Species))
-#> # A tibble: 14 x 4
-#>    .pred_setosa .pred_versicolor .pred_virginica Species   
-#>           <dbl>            <dbl>           <dbl> <fct>     
-#>  1    1.00e+  0         5.97e-24        1.64e-65 setosa    
-#>  2    1.00e+  0         8.08e-28        2.99e-65 setosa    
-#>  3    1.00e+  0         1.09e-22        1.16e-60 setosa    
-#>  4    1.00e+  0         6.82e-29        1.09e-72 setosa    
-#>  5    2.21e- 53         9.99e- 1        9.37e- 4 versicolor
-#>  6    3.02e- 30        10.00e- 1        4.74e- 8 versicolor
-#>  7    4.61e- 33        10.00e- 1        1.77e- 6 versicolor
-#>  8    2.78e- 45         9.99e- 1        1.03e- 3 versicolor
-#>  9    1.33e- 21        10.00e- 1        3.46e-14 versicolor
-#> 10    2.87e- 76         3.93e- 4       10.00e- 1 virginica 
-#> 11    3.73e- 71         4.69e- 5       10.00e- 1 virginica 
-#> 12    2.26e-111         4.48e-15       10.00e- 1 virginica 
-#> 13    1.12e- 58         2.24e- 1        7.76e- 1 virginica 
-#> 14    2.54e- 56         1.86e- 1        8.14e- 1 virginica
    
+
    library(rsample)
+library(tibble)
+
+set.seed(4622)
+iris_split <- initial_split(iris, prop = 0.95)
+iris_train <- training(iris_split)
+iris_test  <-  testing(iris_split)
+
+mda_spec <- mixture_da(sub_classes = 2) %>% 
+  set_engine("mda")
+
+mda_fit <- mda_spec %>%
+  fit(Species ~ ., data = iris_train, engine = "mda")
+mda_fit
+#> parsnip model object
+#> 
+#> Call:
+#> mda::mda(formula = formula, data = data, subclasses = ~2)
+#> 
+#> Dimension: 4 
+#> 
+#> Percent Between-Group Variance Explained:
+#>    v1    v2    v3    v4 
+#>  97.8  99.4 100.0 100.0 
+#> 
+#> Degrees of Freedom (per dimension): 5 
+#> 
+#> Training Misclassification Error: 0.021 ( N = 143 )
+#> 
+#> Deviance: 11.6
+
+predict(mda_fit, new_data = iris_test, type = "prob") %>%
+  mutate(Species = iris_test$Species)
+#> # A tibble: 7 x 4
+#>   .pred_setosa .pred_versicolor .pred_virginica Species   
+#>          <dbl>            <dbl>           <dbl> <fct>     
+#> 1     1.00e+ 0         1.72e-28        1.09e-57 setosa    
+#> 2     1.44e-24        10.00e- 1        3.93e- 8 versicolor
+#> 3     1.35e-24        10.00e- 1        6.23e- 7 versicolor
+#> 4     7.30e-18        10.00e- 1        6.34e-12 versicolor
+#> 5     8.59e-56         1.23e- 4       10.00e- 1 virginica 
+#> 6     8.46e-80         9.71e-11       10.00e- 1 virginica 
+#> 7     3.72e-42         3.03e- 1        6.97e- 1 virginica
+
+predict(mda_fit, new_data = iris_test) %>% 
+  mutate(Species = iris_test$Species)
+#> # A tibble: 7 x 2
+#>   .pred_class Species   
+#>   <fct>       <fct>     
+#> 1 setosa      setosa    
+#> 2 versicolor  versicolor
+#> 3 versicolor  versicolor
+#> 4 versicolor  versicolor
+#> 5 virginica   virginica 
+#> 6 virginica   virginica 
+#> 7 virginica   virginica
    
+
    
+
    
+
    
+Adding an engine to an existing model
    
+
    The process here is almost the same but more simple. You would only need to add the engine-specific aspects of the model. For example, if we wanted to fit a linear regression model using M-estimation, we could only add a new engine. The code for the rlm function in MASS is pretty similar to lm, so we can copy that code and change the package/function names:
    
+
 
    
 
    
 
    
 Pro-tips, what-ifs, exceptions, FAQ, and minutiae
    
 
    There are various things that came to mind while writing this document.
    
-
    
+
    
 
    
-Do I have to return a simple vector for predict_num and predict_class?
    
-
    Previously, when discussing the numeric information:
    
+Do I have to return a simple vector for predict and predict_class?
+
    Previously, when discussing the pred information:
    
 
    
-
    For numeric, the model requires an unnamed numeric vector output (usually).
    
+
    For pred, the model requires an unnamed numeric vector output (usually).
    
 
    
-
    There are some occasions where a prediction for a single new sample may be multidimensional. Examples are enumerated here but some easy examples are:
    
-
      
-
    • confidence or prediction intervals
      • 
-
    • quantile regression predictions.
      • 
-
    
-
    and so on. These can be accomodated via predict.model_fit using different type arguments.
    
-
    However, there are some models (e.g. glmnet, plsr, Cubist, etc.) that can make predictions for different models from the same fitted model object. The regular predict method requires prediction from a single model but the multi_predict can. The guideline is to always return the same number of rows as in new_data. This means that the .pred column is a list-column of tibbles.
    
-
    For example, for a multinomial glmnet model, we leave penalty unspecified when fitting and get predictions on a sequence of values:
    
-
-
    This can be easily expanded to remove the list columns:
    
-
-
    multi_predict doesn’t exist for every model and needs to be implmented by the developer. See methods("multi_predict") for examples in this package.
    
+
    There are some models (e.g. glmnet, plsr, Cubust, etc.) that can make predictions for different models from the same fitted model object. We want to facilitate that here so that, for these cases, the current convention is to return a tibble with the prediction in a column called values and have extra columns for any parameters that define the different sub-models.
    
+
    For example, if I fit a linear regression model via glmnet and get four values of the regularization parameter (lambda):
    
+
    linear_reg() %>%
+  set_engine("glmnet", nlambda = 4) %>% 
+  fit(mpg ~ ., data = mtcars) %>%
+  multi_predict(new_data = mtcars[1:3, -1])
    
+
    However, the api is still being developed. Currently, there is not an interface in the prediction functions to pass in the values of the parameters to make predictions with (lambda, in this case).
    
 
    
-
    
+
    
 
    
-What is the defaults slot and why do I need it?
    
+What is defaults slot and why do I need them?
 
    You might want to set defaults that can be overridden by the user. For example, for logistic regression with glm, it make sense to default family = binomial. However, if someone wants to use a different link function, they should be able to do that. For that model/engine definition, it has
    
-
    defaults = list(family = expr(stats::binomial))
    
+
    defaults = list(family = expr(binomial))
    
 
    so that is the default:
    
-
+
    logistic_reg() %>% translate(engine = "glm")
+
+# but you can change it:
+
+logistic_reg() %>%
+  set_engine("glm", family = expr(binomial(link = "probit"))) %>% 
+  translate()
    
 
    That’s what defaults are for.
    
-
    Note that I wrapped binomial inside of expr. If I didn’t, it would substitute the results of executing binomial inside of the expression (and that’s a mess). Using namespaces is a good idea here.
    
+
    Note that I wrapped binomial inside of expr. If I didn’t, it would substitute the results of executing binomial inside of the expression (and that’s a mess).
    
 
    
 
    
 
    
 What if I want more complex defaults?
    
 
    The translate function can be used to check values or set defaults once the model’s mode is known. To do this, you can create a model-specific S3 method that first calls the general method (translate.model_spec) and then makes modifications or conducts error traps.
    
 
    For example, the ranger and randomForest package functions have arguments for calculating importance. One is a logical and the other is a string. Since this is likely to lead to a bunch of frustration and GH issues, we can put in a check:
    
-
-
    As another example, nnet::nnet has an option for the final layer to be linear (called linout). If mode = "regression", that should probably be set to TRUE. You couldn’t do this with the args (described above) since you need the function translated first.
    
-
    In cases where the model requires different defaults, the translate method can also be used. See the code for the mars function to see how to check and potentially switch arguments for classification models.
    
+
+
    As another example, nnet::nnet has an option for the final layer to be linear (called linout). If mode = "regression", that should probably be set to TRUE. You couldn’t do this with the args (described above) since you need the function translated first.
    
 
    
 
    
 
    
 My model fit requires more than one function call. So….?
    
-
    The best course of action is to write wrapper so that it can be one call. This was the case with xgboost, C5.0, and keras.
    
+
    The best course of action is to write wrapper so that it can be one call. This was the case with xgboost and keras.
    
 
    
 
    
 
    
@@ -500,7 +545,7 @@ 
    
 
    
 
    
 Why would I postprocess my predictions?
    
-
    What comes back from some R functions make be somewhat… arcane or problematic. As an example, for xgboost, if you fit a multiclass boosted tree, you might expect the class probabilities to come back as a matrix1. If you have four classes and make predictions on three samples, you get a vector of 12 probability values. You need to convert these to a rectangular data set.
    
+
    What comes back from some R functions make be somewhat… arcane or problematic. As an example, for xgboost, if you fit a multiclass boosted tree, you might expect the class probabilities to come back as a matrix (narrator: they don’t). If you have four classes and make predictions on three samples, you get a vector of 12 probability values. You need to convert these to a rectangular data set.
    
 
    Another example is the predict method for ranger, which encapsulates the actual predictions in a more complex object structure.
    
 
    These are the types of problems that the postprocessor will solve.
    
 
    
@@ -510,12 +555,6 @@ 
    
 
    Not yet but there will be. For example, it might make sense to have a different mode when doing risk-based modeling via Cox regression models. That would enable different classes of objects and those might be needed since the types of models don’t make direct predictions of the outcome.
    
 
    If you have a suggestion, please ad a GitHub issue to discuss it.
    
 
    
-
    
-
    
-
    
-
      
-
    1. narrator: they don’t↩
      2. 
-
    
 
    
   
    
 
@@ -524,10 +563,14 @@ 
    
       
    
 Contents
    
       
 
    
@@ -536,6 +579,7 @@ 
    
 
    
 
 
+
       
    
   
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
 
    
@@ -546,10 +590,12 @@ 
    
     Site built by pkgdown.
   
    
 
    
+
       
 
    
 
   
 
+
   
 
diff --git a/docs/articles/articles/Submodels.html b/docs/articles/articles/Submodels.html
new file mode 100644
index 000000000..3dfbccfd2
--- /dev/null
+++ b/docs/articles/articles/Submodels.html
@@ -0,0 +1,252 @@
+
+
+
+
+
+
+
+Evaluating Submodels with the Same Model Object • parsnip
+
+
+
+
+
+
+
+
+
+
+    
    
+      
    
+
+
+      
+
+      
    
+  
    
+    
    
+      
    Evaluating Submodels with the Same Model Object
    
+            
+      
+      
+      
+
+    
    
+
+    
+    
+
    Some R packages can create predictions from models that are different than the one that was fit. For example, if a boosted tree is fit with 10 iterations of boosting, the model can usually make predictions on submodels that have less than 10 trees (all other parameters being static). This is helpful for model tuning since you can cheap evaluate tuning parameter combinations and can often results in a large speed-up in the computations.
    
+
    In parsnip, there is a method called multi_predict() that can do this. It’s current methods are:
    
+
    library(parsnip)
+methods("multi_predict")
    
+
    ## [1] multi_predict._C5.0*        multi_predict._earth*      
+## [3] multi_predict._elnet*       multi_predict._lognet*     
+## [5] multi_predict._multnet*     multi_predict._train.kknn* 
+## [7] multi_predict._xgb.Booster* multi_predict.default*     
+## see '?methods' for accessing help and source code
    
+
    We’ll use the attrition data in rsample to illustrate:
    
+
    library(tidymodels)
    
+
    ## ── Attaching packages ─────────────────────────────────────── tidymodels 0.0.2 ──
    
+
    ## ✔ broom     0.5.1          ✔ recipes   0.1.6     
+## ✔ dials     0.0.2.9000     ✔ rsample   0.0.5     
+## ✔ dplyr     0.8.3          ✔ tibble    2.1.3     
+## ✔ infer     0.4.0          ✔ yardstick 0.0.3.9000
+## ✔ purrr     0.3.2
    
+
    ## ── Conflicts ────────────────────────────────────────── tidymodels_conflicts() ──
+## ✖ purrr::discard() masks scales::discard()
+## ✖ dplyr::filter()  masks stats::filter()
+## ✖ dplyr::lag()     masks stats::lag()
+## ✖ dials::margin()  masks ggplot2::margin()
+## ✖ recipes::step()  masks stats::step()
    
+
+
    A boosted classification tree is one of the most low-maintenance approaches that we could take to these data:
    
+
    # requires the xgboost package
+attrition_boost <- 
+  boost_tree(mode = "classification", trees = 100) %>% 
+  set_engine("C5.0")
    
+
    Suppose that 10-fold cross-validation was being used to tune the model over the number of trees:
    
+
+
    The process would fit a model on 90% of the data and predict on the remaining 10%. Using rsample:
    
+
+
    For multi_predict(), the same semantics of predict() are used but, for this model, there is an extra argument called trees. Candidate submodel values can be passed in with trees:
    
+
+
    ## # A tibble: 111 x 1
+##    .pred             
+##    <list>            
+##  1 <tibble [100 × 3]>
+##  2 <tibble [100 × 3]>
+##  3 <tibble [100 × 3]>
+##  4 <tibble [100 × 3]>
+##  5 <tibble [100 × 3]>
+##  6 <tibble [100 × 3]>
+##  7 <tibble [100 × 3]>
+##  8 <tibble [100 × 3]>
+##  9 <tibble [100 × 3]>
+## 10 <tibble [100 × 3]>
+## # … with 101 more rows
    
+
    The results is a tibble that has as many rows as the data being predicted (n = 111). The .pred column contains a list of tibbles and each has the predictions across the different number of trees:
    
+
+
    ## # A tibble: 100 x 3
+##    trees .pred_No .pred_Yes
+##    <int>    <dbl>     <dbl>
+##  1     1    0.691     0.309
+##  2     2    1         0    
+##  3     3    1         0    
+##  4     4    1         0    
+##  5     5    1         0    
+##  6     6    0.862     0.138
+##  7     7    0.882     0.118
+##  8     8    0.770     0.230
+##  9     9    0.689     0.311
+## 10    10    0.724     0.276
+## # … with 90 more rows
    
+
    To get this into a format that is more usable, we can use tidyr::unnest() but we first add row numbers so that we can track the predictions by test sample as well as the actual classes:
    
+
+
    ## # A tibble: 11,100 x 5
+##    Attrition  .row trees .pred_No .pred_Yes
+##    <fct>     <int> <int>    <dbl>     <dbl>
+##  1 Yes           1     1    0.691     0.309
+##  2 Yes           1     2    1         0    
+##  3 Yes           1     3    1         0    
+##  4 Yes           1     4    1         0    
+##  5 Yes           1     5    1         0    
+##  6 Yes           1     6    0.862     0.138
+##  7 Yes           1     7    0.882     0.118
+##  8 Yes           1     8    0.770     0.230
+##  9 Yes           1     9    0.689     0.311
+## 10 Yes           1    10    0.724     0.276
+## # … with 11,090 more rows
    
+
    For two samples, what do these look like over trees?
    
+
+
    
+
    What does performance look like over trees (using the area under the ROC curve)?
    
+
+
    
+  
    
+
+  
+
+
    
+
+
+
+      
    
+  
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
+
    
+
+
    
+  
    
+    Developed by Max Kuhn, Davis Vaughan.
+    Site built by pkgdown.
+  
    
+
    
+
+      
    
+
    
+
+  
+
+
+  
+
diff --git a/docs/articles/articles/Submodels_files/figure-html/auc-plot-1.png b/docs/articles/articles/Submodels_files/figure-html/auc-plot-1.png
new file mode 100644
index 000000000..6f91a5b27
Binary files /dev/null and b/docs/articles/articles/Submodels_files/figure-html/auc-plot-1.png differ
diff --git a/docs/articles/articles/Submodels_files/figure-html/prob-plot-1.png b/docs/articles/articles/Submodels_files/figure-html/prob-plot-1.png
new file mode 100644
index 000000000..ac42253d0
Binary files /dev/null and b/docs/articles/articles/Submodels_files/figure-html/prob-plot-1.png differ
diff --git a/docs/articles/index.html b/docs/articles/index.html
index 686818908..b6350f83b 100644
--- a/docs/articles/index.html
+++ b/docs/articles/index.html
@@ -52,6 +52,7 @@
 
  gtag('config', 'UA-115082821-1');
 
+
   
 
   
@@ -71,7 +72,7 @@
         parsnip
         
    
           part of tidymodels
-          0.0.2
+          0.0.3
    @@ -99,6 +100,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -115,6 +119,7 @@ +
    @@ -131,13 +136,15 @@

    All vignettes

  • Classification Example
  • List of Models
  • Regression Example
    • -
  • Making a `parsnip` model from scratch
    • +
  • Making a parsnip model from scratch
    • +
  • Evaluating Submodels with the Same Model Object
  • parsnip Basics
    • +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -149,11 +156,14 @@

    All vignettes

    Site built by pkgdown.

    +
    + + diff --git a/docs/articles/parsnip_Intro.html b/docs/articles/parsnip_Intro.html index 9029d6c62..bc744eef7 100644 --- a/docs/articles/parsnip_Intro.html +++ b/docs/articles/parsnip_Intro.html @@ -40,7 +40,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -68,6 +68,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -86,6 +89,7 @@ +
    @@ -107,14 +111,14 @@

    parsnip tries to solve this by providing similar interfaces to models. For example, if you are fitting a random forest model and would like to adjust the number of trees in the forest there are different argument names to remember:

    Rather than remembering these values, a common interface to these models can be used with

    -
    library(parsnip)
+
    library(parsnip)
 rf_mod <- rand_forest(trees = 2000)
    
 
    The package makes the translation between trees and the real names in each of the implementations.
    
 
    Some terminology:
    
@@ -151,7 +155,7 @@ 
    
 min_n: The minimum number of data points in a node that are required for the node to be split further.

    • The arguments to the default function are:

    -
    args(rand_forest)
+
 
    However, there might be other arguments that you would like to change or allow to vary. These are accessible using set_engine. For example, ranger has an option to set the internal random number seed. To set this to a specific value:
    
@@ -183,13 +187,13 @@ 
    
 
    rf_with_seed %>% 
   set_args(mtry = 4) %>% 
   set_engine("ranger") %>%
-  fit(mpg ~ ., data = mtcars)
    
+  fit(mpg ~ ., data = mtcars)
    #> parsnip model object
 #> 
 #> Ranger result
 #> 
 #> Call:
-#>  ranger::ranger(formula = formula, data = data, mtry = ~4, num.trees = ~2000,      seed = ~63233, num.threads = 1, verbose = FALSE) 
+#>  ranger::ranger(formula = formula, data = data, mtry = ~4, num.trees = ~2000,      seed = sample.int(10^5, 1), num.threads = 1, verbose = FALSE) 
 #> 
 #> Type:                             Regression 
 #> Number of trees:                  2000 
@@ -202,24 +206,24 @@ 
    
 #> OOB prediction error (MSE):       5.57 
 #> R squared (OOB):                  0.847

    Or, using the randomForest package:

    -
    set.seed(56982)
+
    set.seed(56982)
 rf_with_seed %>% 
   set_args(mtry = 4) %>% 
   set_engine("randomForest") %>%
-  fit(mpg ~ ., data = mtcars)
    
+  fit(mpg ~ ., data = mtcars)
    #> parsnip model object
 #> 
 #> 
 #> Call:
-#>  randomForest(x = as.data.frame(x), y = y, ntree = ~2000, mtry = ~4,      seed = ~63233) 
+#>  randomForest(x = as.data.frame(x), y = y, ntree = ~2000, mtry = ~4) 
 #>                Type of random forest: regression
 #>                      Number of trees: 2000
 #> No. of variables tried at each split: 4
 #> 
 #>           Mean of squared residuals: 5.52
 #>                     % Var explained: 84.3
    -

    Note that the call objects show num.trees = ~2000. The tilde is the consequence of parsnip using quosures to process the model specification’s arguments.

    -

    Normally, when a function is executed, the function’s arguments are immediately evaluated. In the case of parsnip, the model specification’s arguments are not; the expression is captured along with the environment where it should be evaluated. That is what a quosure does.

    +

    Note that the call objects show num.trees = ~2000. The tilde is the consequence of parsnip using quosures to process the model specification’s arguments.

    +

    Normally, when a function is executed, the function’s arguments are immediately evaluated. In the case of parsnip, the model specification’s arguments are not; the expression is captured along with the environment where it should be evaluated. That is what a quosure does.

    parsnip uses these expressions to make a model fit call that is evaluated. The tilde in the call above reflects that the argument was captured using a quosure.

    @@ -230,6 +234,7 @@

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -240,10 +245,12 @@

    Site built by pkgdown.

    +

    + diff --git a/docs/authors.html b/docs/authors.html index e765cb7b2..56a4ed47d 100644 --- a/docs/authors.html +++ b/docs/authors.html @@ -52,6 +52,7 @@ gtag('config', 'UA-115082821-1'); + @@ -71,7 +72,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -99,6 +100,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -115,6 +119,7 @@ +
    @@ -143,6 +148,7 @@

    Authors

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -154,11 +160,14 @@

    Authors

    Site built by pkgdown.

    +
    + + diff --git a/docs/index.html b/docs/index.html index 0acaf4da7..af96dbb82 100644 --- a/docs/index.html +++ b/docs/index.html @@ -11,7 +11,7 @@ - + +
    -

    Travis build status Coverage status

    +

    One issue with different functions available in R that do the same thing is that they can have different interfaces and arguments. For example, to fit a random forest classification model, we might have:

    @@ -122,20 +126,20 @@

    The idea of parsnip is to:

    • Separate the definition of a model from its evaluation.
      • -
    • Decouple the model specification from the implementation (whether the implementation is in R, spark, or something else). For example, the user would call rand_forest instead of ranger::ranger or other specific packages.
      • +
    • Decouple the model specification from the implementation (whether the implementation is in R, spark, or something else). For example, the user would call rand_forest instead of ranger::ranger or other specific packages.
    • Harmonize the argument names (e.g. n.trees, ntrees, trees) so that users can remember a single name. This will help across model types too so that trees will be the same argument across random forest as well as boosting or bagging.

    Using the example above, the parsnip approach would be

    rand_forest(mtry = 12, trees = 2000) %>%
   set_engine("ranger", importance = 'impurity') %>%
-  fit(y ~ ., data = dat)
    + fit(y ~ ., data = dat)

    The engine can be easily changed and the mode can be determined when fit is called. To use Spark, the change is simple:

    rand_forest(mtry = 12, trees = 2000) %>%
   set_engine("spark") %>%
-  fit(y ~ ., data = dat)
    + fit(y ~ ., data = dat)

    To install it, use:

    -
    require(devtools)
-install_github("tidymodels/parsnip")
    +
    require(devtools)
+install_github("tidymodels/parsnip")
    @@ -164,9 +168,20 @@

    Developers

    - +
    +

    Dev status

    +
      +
    • Build Status
      • +
    • Coverage status
      • +
    • CRAN_Status_Badge
      • +
    • Downloads
      • +
    • lifecycle
      • +
    +
    + +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -177,10 +192,12 @@

    Developers

    Site built by pkgdown.

    +
    + diff --git a/docs/news/index.html b/docs/news/index.html index 11ab1b5b1..712020170 100644 --- a/docs/news/index.html +++ b/docs/news/index.html @@ -52,6 +52,7 @@ gtag('config', 'UA-115082821-1'); + @@ -71,7 +72,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -99,6 +100,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -115,6 +119,7 @@ +
    @@ -124,36 +129,62 @@

    Changelog

    -
    +

    -parsnip 0.0.2 Unreleased +parsnip 0.0.3 Unreleased

    -

    Small release driven by changes in sample() in the current r-devel.

    +

    Unplanned release based on CRAN requirements for Solaris.

    +
    +

    +Breaking Changes

    +
      +

    • The method that parsnip stores 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_env() and the package vignette for adding models.

      • +

    • The mode need to be declared for models that can be used for more than one mode prior to fitting and/or translation.

      • +

    • For surv_reg(), the engine that uses the survival package is now called survival instead of survreg.

      • +

    • For glmnet models, the full regularization path is always fit regardless of the value given to penalty. Previously, the model was fit with passing penalty to glmnet’s lambda argument and the model could only make predictions at those specific values. (#195)

      • +
    +

    New Features

      +

    • add_rowindex() can create a column called .row to 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.

      • +

    • nearest_neighbor gained a multi_predict method. The multi_predict() documentation is a little better organized.

      • +

    • A suite of internal functions were added to help with upcoming model tuning features.

      • +
    +
    +
    +
    +

    +parsnip 0.0.2 2019-03-22 +

    +

    Small release driven by changes in sample() in the current r-devel.

    +
    +

    +New Features

    +

    • 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 y without error

      • +

    • fit_xy() can take a single column data frame or matrix for y without error

    Other Changes

      -

    • varying_args() now has a full argument to control whether the full set of possible varying arguments is returned (as opposed to only the arguments that are actually varying).

      • +

    • varying_args() now has a full argument 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_class, 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 check using the “r-devel-windows-ix86+x86_64` flavor. Packages that errored are censored.

      • +

    • An event time data set (check_times) was included that is the time (in seconds) to run R CMD check using the "r-devel-windows-ix86+x86_64` flavor. Packages that errored are censored.

    Bug Fixes

      -

    • varying_args() now uses the version from the generics package. This means that the first argument, x, has been renamed to object to align with generics.

      • -

    • For the recipes step method of varying_args(), there is now error checking to catch if a user tries to specify an argument that cannot be varying as varying (for example, the id) (#132).

      • +

    • varying_args() now uses the version from the generics package. This means that the first argument, x, has been renamed to object to align with generics.

      • +

    • For the recipes step method of varying_args(), there is now error checking to catch if a user tries to specify an argument that cannot be varying as varying (for example, the id) (#132).

    • 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).

    • For multinomial regression using glmnet, multi_predict() now pulls the correct default penalty (#108).

      • @@ -221,7 +252,7 @@

      parsnip 0.0.0.9000 Unreleased

        -
      • The fit interface was previously used to cover both the x/y interface as well as the formula interface. Now, fit() is the formula interface and fit_xy() is for the x/y interface.
        • +
      • The fit interface was previously used to cover both the x/y interface as well as the formula interface. Now, fit() is the formula interface and fit_xy() is for the x/y interface.
      • Added a NEWS.md file to track changes to the package.
      • predict methods were overhauled to be consistent.
        • @@ -234,6 +265,7 @@

        Contents

        +

        parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

        @@ -259,11 +292,14 @@

        Contents

        Site built by pkgdown.

        +

    + + diff --git a/docs/pkgdown.css b/docs/pkgdown.css index 904884727..88181704b 100644 --- a/docs/pkgdown.css +++ b/docs/pkgdown.css @@ -102,21 +102,13 @@ a.anchor { margin-top: -40px; } -/* Static header placement on mobile devices */ -@media (max-width: 767px) { - .navbar-fixed-top { - position: absolute; - } - .navbar { - padding: 0; - } -} - - /* Sidebar --------------------------*/ #sidebar { margin-top: 30px; + position: -webkit-sticky; + position: sticky; + top: 70px; } #sidebar h2 { font-size: 1.5em; @@ -133,6 +125,9 @@ a.anchor { .orcid { height: 16px; + /* margins are required by official ORCID trademark and display guidelines */ + margin-left:4px; + margin-right:4px; vertical-align: middle; } @@ -222,6 +217,19 @@ a.sourceLine:hover { visibility: visible; } +/* headroom.js ------------------------ */ + +.headroom { + will-change: transform; + transition: transform 200ms linear; +} +.headroom--pinned { + transform: translateY(0%); +} +.headroom--unpinned { + transform: translateY(-100%); +} + /* mark.js ----------------------------*/ mark { diff --git a/docs/pkgdown.js b/docs/pkgdown.js index eb7e83d2f..3451dfc28 100644 --- a/docs/pkgdown.js +++ b/docs/pkgdown.js @@ -2,14 +2,7 @@ (function($) { $(function() { - $("#sidebar") - .stick_in_parent({offset_top: 40}) - .on('sticky_kit:bottom', function(e) { - $(this).parent().css('position', 'static'); - }) - .on('sticky_kit:unbottom', function(e) { - $(this).parent().css('position', 'relative'); - }); + $('.navbar-fixed-top').headroom(); $('body').scrollspy({ target: '#sidebar', diff --git a/docs/pkgdown.yml b/docs/pkgdown.yml index 3a80ccd02..7aea716a6 100644 --- a/docs/pkgdown.yml +++ b/docs/pkgdown.yml @@ -1,10 +1,11 @@ -pandoc: 2.0.5 -pkgdown: 1.3.0.9000 -pkgdown_sha: 0b2c27d41b869ac31b6a46815940f55b12f6171f +pandoc: 2.3.1 +pkgdown: 1.3.0.9100 +pkgdown_sha: 71070c3aa5cbff0bffd5bdc599f0fffbd5065142 articles: Classification: articles/Classification.html Models: articles/Models.html Regression: articles/Regression.html Scratch: articles/Scratch.html + Submodels: articles/Submodels.html parsnip_Intro: parsnip_Intro.html diff --git a/docs/reference/C5.0_train.html b/docs/reference/C5.0_train.html index 84988dad0..b93eede93 100644 --- a/docs/reference/C5.0_train.html +++ b/docs/reference/C5.0_train.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -202,6 +207,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -213,11 +219,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/add_rowindex.html b/docs/reference/add_rowindex.html new file mode 100644 index 000000000..0e153d883 --- /dev/null +++ b/docs/reference/add_rowindex.html @@ -0,0 +1,228 @@ + + + + + + + + +Add a column of row numbers to a data frame — add_rowindex • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + + +
    + +
    +
    +
    +

    Add a column of row numbers to a data frame

    + + +
    + +
    + +

    Add a column of row numbers to a data frame

    + +
    + + 
    add_rowindex(x)
    + +

    Arguments

    +

    • -model +engine -engine +mode -class +model -classprob +class conf_int @@ -128,6 +132,9 @@

    List of Models

    pred_int     		+prob + quantile @@ -136,20 +143,17 @@

    List of Models
    -boost_tree() - + C5.0 -✔ +classification -✔ +boost_tree() -× +✔ × @@ -163,19 +167,19 @@

    List of Models

    		 ✔
    -spark +× ✔
    -✔ +classification -× +decision_tree() ✔ @@ -189,20 +193,6 @@

    List of Models

    		 ×
    -xgboost - -✔ - -✔ - -× - ✔ -× - ✔
    -decision_tree() + +earth -C5.0 +classification -✔ +mars() ✔ @@ -239,6 +226,9 @@

    List of Models

    ×     		+✔ + × @@ -247,13 +237,13 @@

    List of Models

    -rpart +regression -✔ +mars() -✔ +× × @@ -268,18 +258,24 @@

    List of Models

    ×     		+× + ✔
    -spark +flexsurv -✔ +regression -✔ +surv_reg() + +× × @@ -294,24 +290,24 @@

    List of Models

    ×     		+✔ + ×
    -linear_reg() - -glmnet +glm -× +classification -× +logistic_reg() -× +✔ ✔ @@ -325,37 +321,29 @@

    List of Models

    		 ✔
    -keras - × -× +✔ -× +
    +glmnet -✔ +classification -× +logistic_reg() -× +✔ ×
    -lm - × -✔ +× ✔
    -× +classification -✔ +multinom_reg()
    -spark +✔ × @@ -398,15 +386,15 @@

    List of Models

    ×     		-× - -× +✔
    -stan +regression + +linear_reg() × @@ -418,10 +406,10 @@

    List of Models

    ✔     		-✔ +× -✔ +× × @@ -432,19 +420,19 @@

    List of Models

    -logistic_reg() +keras -glm +classification -✔ +logistic_reg() ✔ -✔ +× × @@ -453,18 +441,21 @@

    List of Models

    ×     		+✔ + × -✔ +×
    -glmnet +classification -✔ +mlp() ✔ @@ -479,6 +470,9 @@

    List of Models

    ×     		+✔ + × @@ -487,10 +481,10 @@

    List of Models

    -keras +classification -✔ +multinom_reg() ✔ @@ -505,6 +499,9 @@

    List of Models

    ×     		+✔ + × @@ -513,18 +510,21 @@

    List of Models

    -spark +regression -✔ +linear_reg() -✔ +× × +✔ + × @@ -539,13 +539,16 @@

    List of Models

    -stan +regression -✔ +mlp() -✔ +× + +× ✔ @@ -554,7 +557,7 @@

    List of Models

    ×     		-✔ +× × @@ -564,14 +567,14 @@

    List of Models
    -mars() + +kernlab -earth +classification -✔ +svm_poly() ✔ @@ -580,12 +583,15 @@

    List of Models

    ×     		-✔ +× × +✔ + × @@ -593,14 +599,11 @@

    List of Models
    -mlp() - -keras +classification -✔ +svm_rbf() ✔ @@ -609,12 +612,15 @@

    List of Models

    ×     		-✔ +× × +✔ + × @@ -623,13 +629,13 @@

    List of Models

    -nnet +regression -✔ +svm_poly() -✔ +× × @@ -644,26 +650,29 @@

    List of Models

    ×     		+× + ✔
    -multinom_reg() - -glmnet +regression -✔ +svm_rbf() -✔ +× × +✔ + × @@ -677,11 +686,14 @@

    List of Models
    +kknn + -keras +classification -✔ +nearest_neighbor() ✔ @@ -696,27 +708,30 @@

    List of Models

    ×     		-× +✔ × +✔ +
    -spark +regression -✔ +nearest_neighbor() -✔ +× × -× +✔ × @@ -727,22 +742,28 @@

    List of Models

    		 × +✔ +
    -nearest_neighbor() +lm -kknn +regression -✔ +linear_reg() + +× ✔ -× +✔ ✔ @@ -758,14 +779,14 @@

    List of Models
    -null_model() + +nnet -parsnip +classification -✔ +mlp() ✔ @@ -774,12 +795,15 @@

    List of Models

    ×     		-✔ +× × +✔ + × @@ -787,18 +811,53 @@

    List of Models
    -rand_forest() + +regression -randomForest +mlp() + +× + +× ✔ +× + +× + +× + ✔
    +parsnip + +classification + +null_model() + +✔ + +× + +× + × +✔ +
    +regression + +null_model() + +× + +× + +✔ + +× + +× + +× + +✔ +
    +randomForest + +classification + +rand_forest() + +✔ + +× + +× + +× + +✔ + × @@ -817,14 +934,75 @@

    List of Models

    +regression + +rand_forest() + +× + +× + +✔ + +× + +× + +× + +✔ +
    ranger +classification + +rand_forest() + +✔ + +✔ + +× + +× + ✔ +× + ✔
    +regression + +rand_forest() + +× + ✔ +× + ✔
    +rpart + +classification + +decision_tree() + +✔ + +× + +× + +× + +✔ + +× + +✔ +
    +regression + +decision_tree() + +× + +× + +✔ + +× + +× + +× + +✔ +
    spark +classification + +boost_tree() + ✔ +× + +× + +× + ✔ × +× +
    +classification + +decision_tree() + ✔ @@ -866,13 +1134,25 @@

    List of Models

    		 × +✔ + +× + +× +
    -surv_reg() + +classification -flexsurv +logistic_reg() + +✔ × @@ -890,15 +1170,47 @@

    List of Models

    ×     		+× +
    +classification + +multinom_reg() + +✔ + +× + +× + +× + ✔ × +× +
    -survreg +classification + +rand_forest() + +✔ × @@ -916,29 +1228,110 @@

    List of Models

    ×     		+× +
    +regression + +boost_tree() + +× + +× + ✔ × +× + +× + +× +
    -svm_poly() +regression -kernlab +decision_tree() + +× + +× ✔ +× + +× + +× + +× +
    +regression + +linear_reg() + +× + +× + ✔ × +× + +× + +× +
    +regression + +rand_forest() + +× + +× + ✔ @@ -948,20 +1341,128 @@

    List of Models

    ×     		+× + +× +
    +stan + +classification + +logistic_reg() + +✔ + +✔ + +× + +✔ + +✔ + +× + ✔
    -svm_rbf() +regression -kernlab +linear_reg() + +× + +✔ + +✔ + +✔ + +× + +× + +✔ +
    +survival + +regression + +surv_reg() + +× + +× + +✔ + +× + +× + +✔ + +× +
    +xgboost + +classification + +boost_tree() ✔ +× + +× + +× + ✔ @@ -970,6 +1471,26 @@

    List of Models

    		 ✔
    +regression + +boost_tree() + +× + +× + +✔ + +× + ×
    + + + + + +
    x

    A data frame

    + +

    Value

    + +

    The same data frame with a column of 1-based integers named .row.

    + + +

    Examples

    + 
    mtcars %>% add_rowindex()
    #>     mpg cyl  disp  hp drat    wt  qsec vs am gear carb .row
+#> 1  21.0   6 160.0 110 3.90 2.620 16.46  0  1    4    4    1
+#> 2  21.0   6 160.0 110 3.90 2.875 17.02  0  1    4    4    2
+#> 3  22.8   4 108.0  93 3.85 2.320 18.61  1  1    4    1    3
+#> 4  21.4   6 258.0 110 3.08 3.215 19.44  1  0    3    1    4
+#> 5  18.7   8 360.0 175 3.15 3.440 17.02  0  0    3    2    5
+#> 6  18.1   6 225.0 105 2.76 3.460 20.22  1  0    3    1    6
+#> 7  14.3   8 360.0 245 3.21 3.570 15.84  0  0    3    4    7
+#> 8  24.4   4 146.7  62 3.69 3.190 20.00  1  0    4    2    8
+#> 9  22.8   4 140.8  95 3.92 3.150 22.90  1  0    4    2    9
+#> 10 19.2   6 167.6 123 3.92 3.440 18.30  1  0    4    4   10
+#> 11 17.8   6 167.6 123 3.92 3.440 18.90  1  0    4    4   11
+#> 12 16.4   8 275.8 180 3.07 4.070 17.40  0  0    3    3   12
+#> 13 17.3   8 275.8 180 3.07 3.730 17.60  0  0    3    3   13
+#> 14 15.2   8 275.8 180 3.07 3.780 18.00  0  0    3    3   14
+#> 15 10.4   8 472.0 205 2.93 5.250 17.98  0  0    3    4   15
+#> 16 10.4   8 460.0 215 3.00 5.424 17.82  0  0    3    4   16
+#> 17 14.7   8 440.0 230 3.23 5.345 17.42  0  0    3    4   17
+#> 18 32.4   4  78.7  66 4.08 2.200 19.47  1  1    4    1   18
+#> 19 30.4   4  75.7  52 4.93 1.615 18.52  1  1    4    2   19
+#> 20 33.9   4  71.1  65 4.22 1.835 19.90  1  1    4    1   20
+#> 21 21.5   4 120.1  97 3.70 2.465 20.01  1  0    3    1   21
+#> 22 15.5   8 318.0 150 2.76 3.520 16.87  0  0    3    2   22
+#> 23 15.2   8 304.0 150 3.15 3.435 17.30  0  0    3    2   23
+#> 24 13.3   8 350.0 245 3.73 3.840 15.41  0  0    3    4   24
+#> 25 19.2   8 400.0 175 3.08 3.845 17.05  0  0    3    2   25
+#> 26 27.3   4  79.0  66 4.08 1.935 18.90  1  1    4    1   26
+#> 27 26.0   4 120.3  91 4.43 2.140 16.70  0  1    5    2   27
+#> 28 30.4   4  95.1 113 3.77 1.513 16.90  1  1    5    2   28
+#> 29 15.8   8 351.0 264 4.22 3.170 14.50  0  1    5    4   29
+#> 30 19.7   6 145.0 175 3.62 2.770 15.50  0  1    5    6   30
+#> 31 15.0   8 301.0 335 3.54 3.570 14.60  0  1    5    8   31
+#> 32 21.4   4 121.0 109 4.11 2.780 18.60  1  1    4    2   32
    +
    + +
    + + +
    +
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    +
    + +
    +

    + Developed by Max Kuhn, Davis Vaughan. + Site built by pkgdown. +

    +
    + +
    +
    + + + + + + + + diff --git a/docs/reference/boost_tree.html b/docs/reference/boost_tree.html index 67f545c94..11ec5d621 100644 --- a/docs/reference/boost_tree.html +++ b/docs/reference/boost_tree.html @@ -76,6 +76,7 @@ gtag('config', 'UA-115082821-1'); + @@ -95,7 +96,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -123,6 +124,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -139,6 +143,7 @@ +
    @@ -171,7 +176,7 @@

    General Interface for Boosted Trees

    time that the model is fit. Other options and argument can be set using the set_engine() function. 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.

    @@ -181,7 +186,7 @@

    General Interface for Boosted Trees

    loss_reduction = NULL, sample_size = NULL) # S3 method for boost_tree -update(object, mtry = NULL, trees = NULL, +update(object, mtry = NULL, trees = NULL, min_n = NULL, tree_depth = NULL, learn_rate = NULL, loss_reduction = NULL, sample_size = NULL, fresh = FALSE, ...) @@ -242,7 +247,7 @@

    Arg ... -

    Not used for update().

    +

    Not used for update().

    @@ -255,9 +260,9 @@

    Details

    The data given to the function are not saved and are only used to determine the mode of the model. For boost_tree(), the possible modes are "regression" and "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: "xgboost", "C5.0"

      • +

    • R: "xgboost" (the default), "C5.0"

    • Spark: "spark"

    @@ -265,13 +270,13 @@

    Note

    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.

    @@ -309,7 +314,7 @@

    See also

    -

    varying(), fit(), set_engine()

    +

    varying(), fit(), set_engine()

    Examples

    @@ -328,12 +333,12 @@

    Examp #> Main Arguments: #> mtry = 10 #> min_n = 3 -#>
    update(model, mtry = 1)
    #> Boosted Tree Model Specification (unknown) +#>
    update(model, mtry = 1)
    #> Boosted Tree Model Specification (unknown) #> #> Main Arguments: #> mtry = 1 #> min_n = 3 -#>
    update(model, mtry = 1, fresh = TRUE)
    #> Boosted Tree Model Specification (unknown) +#>
    update(model, mtry = 1, fresh = TRUE)
    #> Boosted Tree Model Specification (unknown) #> #> Main Arguments: #> mtry = 1 @@ -360,6 +365,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -371,11 +377,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/check_empty_ellipse.html b/docs/reference/check_empty_ellipse.html index a1f7f9959..2e23494f0 100644 --- a/docs/reference/check_empty_ellipse.html +++ b/docs/reference/check_empty_ellipse.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -118,6 +122,7 @@ +
    @@ -162,6 +167,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -173,11 +179,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/check_times.html b/docs/reference/check_times.html index 89ff3919d..e68781981 100644 --- a/docs/reference/check_times.html +++ b/docs/reference/check_times.html @@ -59,6 +59,7 @@ gtag('config', 'UA-115082821-1'); + @@ -78,7 +79,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -106,6 +107,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -122,6 +126,7 @@ +
    @@ -195,8 +200,8 @@

    Details

    Examples

    - 
    data(check_times)
-str(check_times)
    #> Classes ‘tbl_df’, ‘tbl’ and 'data.frame':	13626 obs. of  25 variables:
+    
    data(check_times)
+str(check_times)
    #> 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 ...
@@ -239,6 +244,7 @@ 
    Contents
    
   
    
 
    
 
+
       
    
       
    
   
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
@@ -250,11 +256,14 @@ 
    Contents
    
     Site built by pkgdown.
   
    
 
    
+
    + + diff --git a/docs/reference/decision_tree.html b/docs/reference/decision_tree.html index cf3a7d3c8..63a8e9e62 100644 --- a/docs/reference/decision_tree.html +++ b/docs/reference/decision_tree.html @@ -70,6 +70,7 @@ gtag('config', 'UA-115082821-1'); + @@ -89,7 +90,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -117,6 +118,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -133,6 +137,7 @@ +
    @@ -159,7 +164,7 @@

    General Interface for Decision Tree Models

    time that the model is fit. Other options and argument can be set using 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.

    @@ -168,7 +173,7 @@

    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 +210,15 @@

    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 rpart models, but cost_complexity and @@ -226,13 +231,13 @@

    Note

    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.

    @@ -259,17 +264,17 @@

    spark classification

     sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(), 
-    seed = sample.int(10^5, 1), type = "classification")
+    seed = sample.int(10^5, 1))

    spark regression

     sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(), 
-    seed = sample.int(10^5, 1), type = "regression")
+    seed = sample.int(10^5, 1))

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -288,12 +293,12 @@

    Examp #> Main Arguments: #> cost_complexity = 10 #> min_n = 3 -#>
    update(model, cost_complexity = 1)
    #> Decision Tree Model Specification (unknown) +#>
    update(model, cost_complexity = 1)
    #> Decision Tree Model Specification (unknown) #> #> Main Arguments: #> cost_complexity = 1 #> min_n = 3 -#>
    update(model, cost_complexity = 1, fresh = TRUE)
    #> Decision Tree Model Specification (unknown) +#>
    update(model, cost_complexity = 1, fresh = TRUE)
    #> Decision Tree Model Specification (unknown) #> #> Main Arguments: #> cost_complexity = 1 @@ -318,6 +323,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -329,11 +335,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/descriptors.html b/docs/reference/descriptors.html index 4fcba8cb4..dfaba2ab3 100644 --- a/docs/reference/descriptors.html +++ b/docs/reference/descriptors.html @@ -59,6 +59,7 @@ gtag('config', 'UA-115082821-1'); + @@ -78,7 +79,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -106,6 +107,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -122,6 +126,7 @@ +
    @@ -134,7 +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 +179,7 @@

    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. If fit_xy() was used, the outcomes are attached as the +outcomes. If fit_xy() was used, the outcomes are attached as the column, ..y.

    • For example, if you use the model formula Sepal.Width ~ . with the iris @@ -200,7 +205,7 @@

    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 via fit() (and the variables listed above are available). For example:

         data("lending_club")
@@ -221,6 +226,7 @@ 
    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -232,11 +238,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/fit.html b/docs/reference/fit.html index 874199db2..6a48eb842 100644 --- a/docs/reference/fit.html +++ b/docs/reference/fit.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -132,18 +137,18 @@

    Fit a Model Specification to a Dataset

    -

    fit() and fit_xy() take a model specification, translate the required +

    fit() and fit_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 +211,24 @@

    Value

    Details

    -

    fit() and fit_xy() substitute the current arguments in the model +

    fit() and fit_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 or x/y) and these functions translate -between the interface used when fit() or fit_xy() were invoked and the one +between the interface used when fit() or fit_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 as x/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 verbosity option of +fit_control() is greater than zero, a warning will be produced.

    See also

    @@ -231,28 +239,26 @@

    Examp 
    # Although `glm()` only has a formula interface, different
 # methods for specifying the model can be used
 
-library(dplyr)
    #> 
+library(dplyr)
    #> 
 #> Attaching package: ‘dplyr’
    #> The following object is masked from ‘package:testthat’:
 #> 
 #>     matches
    #> The following objects are masked from ‘package:stats’:
 #> 
 #>     filter, lag
    #> The following objects are masked from ‘package:base’:
 #> 
-#>     intersect, setdiff, setequal, union
    data("lending_club")
-
-lr_mod <- logistic_reg()
+#>     intersect, setdiff, setequal, union
    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
@@ -296,6 +302,7 @@ 
    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -307,11 +314,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/fit_control.html b/docs/reference/fit_control.html index d1bfd9790..96240c9d1 100644 --- a/docs/reference/fit_control.html +++ b/docs/reference/fit_control.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -131,7 +136,7 @@

    Control the fit function

    -

    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 computations

    @@ -154,7 +159,7 @@

    Arg catch

    A logical where a value of TRUE will evaluate -the model inside of try(, silent = TRUE). If the model fails, +the model inside of try(, silent = TRUE). If the model fails, an object is still returned (without an error) that inherits the class "try-error".

    @@ -178,6 +183,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -189,11 +195,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/get_model_env.html b/docs/reference/get_model_env.html new file mode 100644 index 000000000..813637906 --- /dev/null +++ b/docs/reference/get_model_env.html @@ -0,0 +1,259 @@ + + + + + + + + +Working with the parsnip model environment — get_model_env • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + + +
    + +
    +
    +
    +

    Working with the parsnip model environment

    + + +
    + +
    + +

    These functions read and write to the environment where the package stores +information about model specifications.

    + +
    + + 
    get_model_env()
+
+get_from_env(items)
+
+set_in_env(...)
+
+set_env_val(name, value)
    + +

    Arguments

    + + + + + + + + + + + + + + + + + + +
    items

    A character string of objects in the model environment.

    ...

    Named values that will be assigned to the model environment.

    name

    A single character value for a new symbol in the model environment.

    value

    A single value for a new value in the model environment.

    + +

    References

    + +

    "Making a parsnip model from scratch" +https://tidymodels.github.io/parsnip/articles/articles/Scratch.html

    + + +

    Examples

    + 
    # Access the model data:
+current_code <- get_model_env()
+ls(envir = current_code)
    #>  [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"         
    
+
    +
    + +
    + + +
    +
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    +
    + +
    +

    + Developed by Max Kuhn, Davis Vaughan. + Site built by pkgdown. +

    +
    + +
    +
    + + + + + + + + diff --git a/docs/reference/has_multi_predict.html b/docs/reference/has_multi_predict.html new file mode 100644 index 000000000..a366f6203 --- /dev/null +++ b/docs/reference/has_multi_predict.html @@ -0,0 +1,248 @@ + + + + + + + + +Tools for models that predict on sub-models — has_multi_predict • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + + +
    + +
    +
    +
    +

    Tools for models that predict on sub-models

    + + +
    + +
    + +

    has_multi_predict() tests to see if an object can make multiple +predictions on submodels from the same object. multi_predict_args() +returns the names of the argments to multi_predict() for this model +(if any).

    + +
    + + 
    has_multi_predict(object, ...)
+
+# S3 method for default
+has_multi_predict(object, ...)
+
+# S3 method for model_fit
+has_multi_predict(object, ...)
+
+# S3 method for workflow
+has_multi_predict(object, ...)
+
+multi_predict_args(object, ...)
+
+# S3 method for default
+multi_predict_args(object, ...)
+
+# S3 method for model_fit
+multi_predict_args(object, ...)
+
+# S3 method for workflow
+multi_predict_args(object, ...)
    + +

    Arguments

    + + + + + + + + + + +
    object

    An object to test.

    ...

    Not currently used.

    + +

    Value

    + +

    has_multi_predict() returns single logical value while +multi_predict() returns a character vector of argument names (or NA +if none exist).

    + + +

    Examples

    + 
    lm_model_idea <- linear_reg() %>% set_engine("lm")
+has_multi_predict(lm_model_idea)
    #> [1] FALSE
    lm_model_fit <- fit(lm_model_idea, mpg ~ ., data = mtcars)
+has_multi_predict(lm_model_fit)
    #> [1] FALSE
    
+multi_predict_args(lm_model_fit)
    #> [1] NA
    
+library(kknn)
+
+knn_fit <-
+  nearest_neighbor(mode = "regression", neighbors = 5) %>%
+  set_engine("kknn") %>%
+  fit(mpg ~ ., mtcars)
+
+multi_predict_args(knn_fit)
    #> [1] "neighbors"
    
+multi_predict(knn_fit, mtcars[1, -1], neighbors = 1:4)$.pred
    #> [[1]]
+#> # A tibble: 4 x 2
+#>   neighbors .pred
+#>       <int> <dbl>
+#> 1         1  21.2
+#> 2         2  21.2
+#> 3         3  21.2
+#> 4         4  21.2
+#>
    +
    + +
    + + +
    +
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    +
    + +
    +

    + Developed by Max Kuhn, Davis Vaughan. + Site built by pkgdown. +

    +
    + +
    +
    + + + + + + + + diff --git a/docs/reference/index.html b/docs/reference/index.html index d54213b6b..7f4ebac1e 100644 --- a/docs/reference/index.html +++ b/docs/reference/index.html @@ -52,6 +52,7 @@ gtag('config', 'UA-115082821-1'); + @@ -71,7 +72,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -99,6 +100,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -115,6 +119,7 @@ +
    @@ -226,6 +231,12 @@

    add_rowindex()

    + +

    Add a column of row numbers to a data frame

    + +

    .cols() .preds() .obs() .lvls() .facts() .x() .y() .dat()

    @@ -269,7 +280,7 @@

    multi_predict()

    +

    multi_predict

    Model predictions across many sub-models

    @@ -329,6 +340,20 @@

    <

    Execution Time Data

    + + + +

    Developer Tools

    +

    + + + + + +

    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() pred_value_template()

    + +

    Tools to Register Models

    +

    @@ -339,10 +364,12 @@

    Contents

  • Models
  • Infrastructure
  • Data
    • +
  • Developer Tools
    • +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -354,11 +381,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/keras_mlp.html b/docs/reference/keras_mlp.html index 229785727..e6a8811ba 100644 --- a/docs/reference/keras_mlp.html +++ b/docs/reference/keras_mlp.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -139,7 +144,7 @@

    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),
   ...)

    Arguments

    @@ -203,6 +208,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -214,11 +220,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/lending_club.html b/docs/reference/lending_club.html index 70120d370..b79f08723 100644 --- a/docs/reference/lending_club.html +++ b/docs/reference/lending_club.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -118,6 +122,7 @@ +
    @@ -158,8 +163,8 @@

    Details

    Examples

    - 
    data(lending_club)
-str(lending_club)
    #> Classes ‘tbl_df’, ‘tbl’ and 'data.frame':	9857 obs. of  23 variables:
+    
    data(lending_club)
+str(lending_club)
    #> 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 ...
@@ -200,6 +205,7 @@ 
    Contents
    
   
    
 
    
 
+
       
    
       
    
   
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
@@ -211,11 +217,14 @@ 
    Contents
    
     Site built by pkgdown.
   
    
 
    
+
    + + diff --git a/docs/reference/linear_reg.html b/docs/reference/linear_reg.html index c9e5dd46d..8f16085b0 100644 --- a/docs/reference/linear_reg.html +++ b/docs/reference/linear_reg.html @@ -68,6 +68,7 @@ gtag('config', 'UA-115082821-1'); + @@ -87,7 +88,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -115,6 +116,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -131,6 +135,7 @@ +
    @@ -155,7 +160,7 @@

    General Interface for Linear Regression Models

    time that the model is fit. Other options and argument can be set using 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.

    @@ -163,7 +168,7 @@

    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 +205,7 @@

    Arg ... -

    Not used for update().

    +

    Not used for update().

    @@ -209,9 +214,9 @@

    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"

    • keras: "keras"

      • @@ -221,13 +226,13 @@

      Note

      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.

      @@ -262,11 +267,11 @@

      -

      When using glmnet models, there is the option to pass -multiple values (or no values) to the penalty argument. This -can have an effect on the model object results. When using the -predict() method in these cases, the return value depends on -the value of penalty. When using predict(), only a single +

      For glmnet models, the full regularization path is always fit regardless +of the value given to penalty. Also, there is the option to pass +multiple values (or no values) to the penalty argument. When using the +predict() method in these cases, the return value depends on +the value of penalty. When using predict(), only a single value of the penalty can be used. When predicting on multiple penalties, the multi_predict() function can be used. It returns a tibble with a list column called .pred that contains @@ -280,7 +285,7 @@

      See also

      -

      varying(), fit(), set_engine()

      +

      varying(), fit(), set_engine()

      Examples

      @@ -296,12 +301,12 @@

      Examp #> Main Arguments: #> penalty = 10 #> mixture = 0.1 -#>
      update(model, penalty = 1)
      #> Linear Regression Model Specification (regression) +#>
      update(model, penalty = 1)
      #> Linear Regression Model Specification (regression) #> #> Main Arguments: #> penalty = 1 #> mixture = 0.1 -#>
      update(model, penalty = 1, fresh = TRUE)
      #> Linear Regression Model Specification (regression) +#>
      update(model, penalty = 1, fresh = TRUE)
      #> Linear Regression Model Specification (regression) #> #> Main Arguments: #> penalty = 1 @@ -326,6 +331,7 @@

      Contents

      +

      parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

      @@ -337,11 +343,14 @@

      Contents

      Site built by pkgdown.

      +
      + + diff --git a/docs/reference/logistic_reg.html b/docs/reference/logistic_reg.html index cdfc30736..1f5268311 100644 --- a/docs/reference/logistic_reg.html +++ b/docs/reference/logistic_reg.html @@ -68,6 +68,7 @@ gtag('config', 'UA-115082821-1'); + @@ -87,7 +88,7 @@ parsnip
      part of tidymodels - 0.0.2 + 0.0.3
      @@ -115,6 +116,9 @@
    • Making a parsnip model from scratch
      • +
    • + Evaluating submodels with the same model object +

  • @@ -131,6 +135,7 @@ +
    @@ -155,7 +160,7 @@

    General Interface for Logistic Regression Models

    time that the model is fit. Other options and argument can be set using 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.

    @@ -163,7 +168,7 @@

    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 +205,16 @@

    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"

      • @@ -219,13 +224,13 @@

      Note

      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.

      @@ -261,11 +266,11 @@

      -

      When using glmnet models, there is the option to pass -multiple values (or no values) to the penalty argument. This -can have an effect on the model object results. When using the -predict() method in these cases, the return value depends on -the value of penalty. When using predict(), only a single +

      For glmnet models, the full regularization path is always fit regardless +of the value given to penalty. Also, there is the option to pass +multiple values (or no values) to the penalty argument. When using the +predict() method in these cases, the return value depends on +the value of penalty. When using predict(), only a single value of the penalty can be used. When predicting on multiple penalties, the multi_predict() function can be used. It returns a tibble with a list column called .pred that contains @@ -280,7 +285,7 @@

      See also

      -

      varying(), fit()

      +

      varying(), fit()

      Examples

      @@ -296,12 +301,12 @@

      Examp #> Main Arguments: #> penalty = 10 #> mixture = 0.1 -#>
      update(model, penalty = 1)
      #> Logistic Regression Model Specification (classification) +#>
      update(model, penalty = 1)
      #> Logistic Regression Model Specification (classification) #> #> Main Arguments: #> penalty = 1 #> mixture = 0.1 -#>
      update(model, penalty = 1, fresh = TRUE)
      #> Logistic Regression Model Specification (classification) +#>
      update(model, penalty = 1, fresh = TRUE)
      #> Logistic Regression Model Specification (classification) #> #> Main Arguments: #> penalty = 1 @@ -326,6 +331,7 @@

      Contents

      +

      parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

      @@ -337,11 +343,14 @@

      Contents

      Site built by pkgdown.

      +
      + + diff --git a/docs/reference/make_classes.html b/docs/reference/make_classes.html index 5cc4b4bca..aa7fb9b3e 100644 --- a/docs/reference/make_classes.html +++ b/docs/reference/make_classes.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
      part of tidymodels - 0.0.2 + 0.0.3
      @@ -102,6 +103,9 @@
    • Making a parsnip model from scratch
      • +
    • + Evaluating submodels with the same model object +

  • @@ -118,6 +122,7 @@ +
    @@ -162,6 +167,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -173,11 +179,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/mars.html b/docs/reference/mars.html index 09a4882e4..acf85eb74 100644 --- a/docs/reference/mars.html +++ b/docs/reference/mars.html @@ -71,6 +71,7 @@ gtag('config', 'UA-115082821-1'); + @@ -90,7 +91,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -118,6 +119,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -134,6 +138,7 @@ +
    @@ -161,7 +166,7 @@

    General Interface for MARS

    time that the model is fit. Other options and argument can be set using 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.

    @@ -170,7 +175,7 @@

    General Interface for MARS

    prune_method = NULL) # S3 method for mars -update(object, num_terms = NULL, prod_degree = NULL, +update(object, num_terms = NULL, prod_degree = NULL, prune_method = NULL, fresh = FALSE, ...)

    Arguments

    @@ -206,15 +211,15 @@

    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: "earth"

      • +

    • R: "earth" (the default)

    Engine Details

    @@ -239,7 +244,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -253,12 +258,12 @@

    Examp #> Main Arguments: #> num_terms = 10 #> prune_method = none -#>
    update(model, num_terms = 1)
    #> MARS Model Specification (unknown) +#>
    update(model, num_terms = 1)
    #> MARS Model Specification (unknown) #> #> Main Arguments: #> num_terms = 1 #> prune_method = none -#>
    update(model, num_terms = 1, fresh = TRUE)
    #> MARS Model Specification (unknown) +#>
    update(model, num_terms = 1, fresh = TRUE)
    #> MARS Model Specification (unknown) #> #> Main Arguments: #> num_terms = 1 @@ -281,6 +286,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -292,11 +298,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/min_grid.html b/docs/reference/min_grid.html new file mode 100644 index 000000000..be6333af9 --- /dev/null +++ b/docs/reference/min_grid.html @@ -0,0 +1,223 @@ + + + + + + + + +Determine the minimum set of model fits — min_grid • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + + +
    + +
    +
    +
    +

    Determine the minimum set of model fits

    + + +
    + +
    + +

    min_grid determines exactly what models should be fit in order to +evaluate the entire set of tuning parameter combinations. This is for +internal use only.

    + +
    + + 
    min_grid(x, grid, ...)
+
+# S3 method for boost_tree
+min_grid(x, grid, ...)
+
+# S3 method for linear_reg
+min_grid(x, grid, ...)
+
+# S3 method for logistic_reg
+min_grid(x, grid, ...)
+
+# S3 method for mars
+min_grid(x, grid, ...)
+
+# S3 method for multinom_reg
+min_grid(x, grid, ...)
+
+# S3 method for nearest_neighbor
+min_grid(x, grid, ...)
    + +

    Arguments

    + + + + + + + + + + + + + + +
    x

    A model specification.

    grid

    A tibble with tuning parameter combinations.

    ...

    Not currently used.

    + +

    Value

    + +

    A tibble with the minimum tuning parameters to fit and an additional +list column with the parameter combinations used for prediction.

    + + +
    + +
    + + +
    +
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    +
    + +
    +

    + Developed by Max Kuhn, Davis Vaughan. + Site built by pkgdown. +

    +
    + +
    +
    + + + + + + + + diff --git a/docs/reference/mlp.html b/docs/reference/mlp.html index 460c32d47..67f392e51 100644 --- a/docs/reference/mlp.html +++ b/docs/reference/mlp.html @@ -71,6 +71,7 @@ gtag('config', 'UA-115082821-1'); + @@ -90,7 +91,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -118,6 +119,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -134,6 +138,7 @@ +
    @@ -170,7 +175,7 @@

    General Interface for Single Layer Neural Network

    dropout = NULL, epochs = NULL, activation = NULL) # S3 method for mlp -update(object, hidden_units = NULL, penalty = NULL, +update(object, hidden_units = NULL, penalty = NULL, dropout = NULL, epochs = NULL, activation = NULL, fresh = FALSE, ...) @@ -220,7 +225,7 @@

    Arg ... -

    Not used for update().

    +

    Not used for update().

    @@ -230,15 +235,15 @@

    Details time that the model is fit. Other options and argument can be set using 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 +276,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -290,12 +295,12 @@

    Examp #> Main Arguments: #> hidden_units = 10 #> dropout = 0.3 -#>

    update(model, hidden_units = 2)
    #> Single Layer Neural Network Specification (unknown) +#>
    update(model, hidden_units = 2)
    #> Single Layer Neural Network Specification (unknown) #> #> Main Arguments: #> hidden_units = 2 #> dropout = 0.3 -#>
    update(model, hidden_units = 2, fresh = TRUE)
    #> Single Layer Neural Network Specification (unknown) +#>
    update(model, hidden_units = 2, fresh = TRUE)
    #> Single Layer Neural Network Specification (unknown) #> #> Main Arguments: #> hidden_units = 2 @@ -318,6 +323,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -329,11 +335,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/model_fit.html b/docs/reference/model_fit.html index aa4317313..c5cf3eaf2 100644 --- a/docs/reference/model_fit.html +++ b/docs/reference/model_fit.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -168,7 +173,7 @@

    Examp # Keep the `x` matrix if the data are not too big. spec_obj <- linear_reg() %>% - set_engine("lm", x = ifelse(.obs() < 500, TRUE, FALSE)) + set_engine("lm", x = ifelse(.obs() < 500, TRUE, FALSE)) spec_obj

    #> Linear Regression Model Specification (regression) #> #> Engine-Specific Arguments: @@ -176,7 +181,7 @@

    Examp #> #> Computational engine: lm #>

    -fit_obj <- fit(spec_obj, mpg ~ ., data = mtcars) +fit_obj <- fit(spec_obj, mpg ~ ., data = mtcars) fit_obj
    #> parsnip model object #> #> @@ -190,7 +195,7 @@

    Examp #> qsec vs am gear carb #> 0.82104 0.31776 2.52023 0.65541 -0.19942 #>

    -nrow(fit_obj$fit$x)
    #> [1] 32
    +nrow(fit_obj$fit$x)
    #> [1] 32
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -215,11 +221,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/model_printer.html b/docs/reference/model_printer.html index 06d3c8441..a0cc9076c 100644 --- a/docs/reference/model_printer.html +++ b/docs/reference/model_printer.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -162,6 +167,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -173,11 +179,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/model_spec.html b/docs/reference/model_spec.html index 9b3772e75..268ec0597 100644 --- a/docs/reference/model_spec.html +++ b/docs/reference/model_spec.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -177,7 +182,7 @@

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -267,11 +273,14 @@

    Contents

    Site built by     pkgdown.

    + + + diff --git a/docs/reference/multi_predict.html b/docs/reference/multi_predict.html index 3477f2790..a167d4ac2 100644 --- a/docs/reference/multi_predict.html +++ b/docs/reference/multi_predict.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -118,6 +122,7 @@ +
    @@ -137,7 +142,35 @@

    Model predictions across many sub-models

    multi_predict(object, ...)
 
 # S3 method for default
-multi_predict(object, ...)
    +multi_predict(object, ...) + +# S3 method for _xgb.Booster +multi_predict(object, new_data, type = NULL, + trees = NULL, ...) + +# S3 method for _C5.0 +multi_predict(object, new_data, type = NULL, + trees = NULL, ...) + +# S3 method for _elnet +multi_predict(object, new_data, type = NULL, + penalty = NULL, ...) + +# S3 method for _lognet +multi_predict(object, new_data, type = NULL, + penalty = NULL, ...) + +# S3 method for _earth +multi_predict(object, new_data, type = NULL, + num_terms = NULL, ...) + +# S3 method for _multnet +multi_predict(object, new_data, type = NULL, + penalty = NULL, ...) + +# S3 method for _train.kknn +multi_predict(object, new_data, type = NULL, + neighbors = NULL, ...)

    Arguments

    @@ -151,6 +184,33 @@

    Arg

    + + + + + + + + + + + + + + + + + + + + + + + +

    Optional arguments to pass to predict.model_fit(type = "raw") such as type.
    new_data

    A rectangular data object, such as a data frame.

    type

    A single character value or NULL. Possible values +are "numeric", "class", "prob", "conf_int", "pred_int", "quantile", +or "raw". When NULL, predict() will choose an appropriate value +based on the model's mode.

    trees

    An integer vector for the number of trees in the ensemble.

    penalty

    An numeric vector of penalty values.

    num_terms

    An integer vector for the number of MARS terms to retain.

    neighbors

    An integer vector for the number of nearest neighbors.

    Value

    @@ -172,6 +232,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -183,11 +244,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/multinom_reg.html b/docs/reference/multinom_reg.html index 3f6668f8c..782ad6c87 100644 --- a/docs/reference/multinom_reg.html +++ b/docs/reference/multinom_reg.html @@ -68,6 +68,7 @@ gtag('config', 'UA-115082821-1'); + @@ -87,7 +88,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -115,6 +116,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -131,6 +135,7 @@ +
    @@ -155,7 +160,7 @@

    General Interface for Multinomial Regression Models

    time that the model is fit. Other options and argument can be set using 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.

    @@ -163,7 +168,7 @@

    General Interface for Multinomial Regression Models

    multinom_reg(mode = "classification", penalty = NULL, mixture = NULL)
 
 # S3 method for multinom_reg
-update(object, penalty = NULL, mixture = NULL,
+update(object, penalty = NULL, mixture = NULL,
   fresh = FALSE, ...)

    Arguments

    @@ -200,16 +205,16 @@

    Arg ... -

    Not used for update().

    +

    Not used for update().

    Details

    For multinom_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: "glmnet"

      • +

    • R: "glmnet" (the default)

    • Stan: "stan"

    • keras: "keras"

    @@ -218,13 +223,13 @@

    Note

    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.

    @@ -250,11 +255,11 @@

    -

    When using glmnet models, there is the option to pass -multiple values (or no values) to the penalty argument. This -can have an effect on the model object results. When using the -predict() method in these cases, the return value depends on -the value of penalty. When using predict(), only a single +

    For glmnet models, the full regularization path is always fit regardless +of the value given to penalty. Also, there is the option to pass +multiple values (or no values) to the penalty argument. When using the +predict() method in these cases, the return value depends on +the value of penalty. When using predict(), only a single value of the penalty can be used. When predicting on multiple penalties, the multi_predict() function can be used. It returns a tibble with a list column called .pred that contains @@ -262,7 +267,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -278,12 +283,12 @@

    Examp #> Main Arguments: #> penalty = 10 #> mixture = 0.1 -#>
    update(model, penalty = 1)
    #> Multinomial Regression Model Specification (classification) +#>
    update(model, penalty = 1)
    #> Multinomial Regression Model Specification (classification) #> #> Main Arguments: #> penalty = 1 #> mixture = 0.1 -#>
    update(model, penalty = 1, fresh = TRUE)
    #> Multinomial Regression Model Specification (classification) +#>
    update(model, penalty = 1, fresh = TRUE)
    #> Multinomial Regression Model Specification (classification) #> #> Main Arguments: #> penalty = 1 @@ -308,6 +313,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -319,11 +325,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/nearest_neighbor.html b/docs/reference/nearest_neighbor.html index ca04511c1..a47780a2c 100644 --- a/docs/reference/nearest_neighbor.html +++ b/docs/reference/nearest_neighbor.html @@ -71,6 +71,7 @@ gtag('config', 'UA-115082821-1'); + @@ -90,7 +91,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -118,6 +119,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -134,6 +138,7 @@ +
    @@ -161,7 +166,7 @@

    General Interface for K-Nearest Neighbor Models

    time that the model is fit. Other options and argument can be set using 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.

    @@ -181,7 +186,8 @@

    Arg neighbors

    A single integer for the number of neighbors -to consider (often called k).

    +to consider (often called k). For kknn, a value of 5 +is used if neighbors is not specified.

    weight_func @@ -199,9 +205,9 @@

    Arg

    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: "kknn"

      • +

    • R: "kknn" (the default)

    Note

    @@ -221,12 +227,12 @@

    kknn (classification or regression)

     kknn::train.kknn(formula = missing_arg(), data = missing_arg(), 
-    kmax = missing_arg())
+    ks = 5)

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -256,6 +262,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -267,11 +274,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/null_model.html b/docs/reference/null_model.html index d433a77df..064e0b64b 100644 --- a/docs/reference/null_model.html +++ b/docs/reference/null_model.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -153,7 +158,7 @@

    Arg

    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: "parsnip"

    @@ -175,7 +180,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -199,6 +204,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -210,11 +216,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/nullmodel.html b/docs/reference/nullmodel.html index edc9778e0..473af96a9 100644 --- a/docs/reference/nullmodel.html +++ b/docs/reference/nullmodel.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -142,10 +147,10 @@

    Fit a simple, non-informative model

    nullmodel(x = NULL, y, ...) # S3 method for nullmodel -print(x, ...) +print(x, ...) # S3 method for nullmodel -predict(object, new_data = NULL, type = NULL, ...) +predict(object, new_data = NULL, type = NULL, ...)

    Arguments

    @@ -207,13 +212,13 @@

    Details

    Examples

    
-outcome <- factor(sample(letters[1:2],
+outcome <- factor(sample(letters[1:2],
                          size = 100,
-                         prob = c(.1, .9),
+                         prob = c(.1, .9),
                          replace = TRUE))
 useless <- nullmodel(y = outcome)
 useless
    #> Null Regression Model
-#> Predicted Value: b 
    predict(useless, matrix(NA, nrow = 5))
    #> [1] b b b b b
+#> Predicted Value: b 
    predict(useless, matrix(NA, nrow = 5))
    #> [1] b b b b b
 #> Levels: a b
    @@ -232,6 +237,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -243,11 +249,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/predict.model_fit.html b/docs/reference/predict.model_fit.html index 7f0a5b2e3..c727e0747 100644 --- a/docs/reference/predict.model_fit.html +++ b/docs/reference/predict.model_fit.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -133,14 +138,14 @@

    Model predictions

    Apply a model to create different types of predictions. -predict() can be used for all types of models and used the +predict() can be used for all types of models and used the "type" argument for more specificity.

    # S3 method for model_fit
-predict(object, new_data, type = NULL,
-  opts = list(), ...)
    +predict(object, new_data, type = NULL, + opts = list(), ...)

    Arguments

    • @@ -157,7 +162,7 @@

    Arg

    @@ -213,16 +218,16 @@

    Value

    appear in names and 2) vectors are never returned but type-specific prediction functions.

    When the model fit failed and the error was captured, the -predict() function will return the same structure as above but +predict() function will return the same structure as above but filled with missing values. This does not currently work for multivariate models.

    Details

    -

    If "type" is not supplied to predict(), then a choice +

    If "type" is not supplied to predict(), then a choice is made (type = "numeric" for regression models and type = "class" for classification).

    -

    predict() is designed to provide a tidy result (see "Value" +

    predict() is designed to provide a tidy result (see "Value" section below) in a tibble output format.

    When using type = "conf_int" and type = "pred_int", the options level and std_error can be used. The latter is a logical for an @@ -230,54 +235,54 @@

    Details

    Examples

    - 
    library(dplyr)
+    
    library(dplyr)
 
 lm_model <-
   linear_reg() %>%
   set_engine("lm") %>%
-  fit(mpg ~ ., data = mtcars %>% slice(11:32))
+  fit(mpg ~ ., data = mtcars %>% slice(11:32))
 
 pred_cars <-
   mtcars %>%
   slice(1:10) %>%
   select(-mpg)
 
-predict(lm_model, pred_cars)
    #> # A tibble: 10 x 1
+predict(lm_model, pred_cars)
    #> # A tibble: 10 x 1
 #>    .pred
-#>    <dbl>
-#>  1  23.4
-#>  2  23.3
-#>  3  27.6
-#>  4  21.5
-#>  5  17.6
-#>  6  21.6
-#>  7  13.9
-#>  8  21.7
-#>  9  25.6
-#> 10  17.1
    
-predict(
+#>    <dbl>
+#>  1  23.4
+#>  2  23.3
+#>  3  27.6
+#>  4  21.5
+#>  5  17.6
+#>  6  21.6
+#>  7  13.9
+#>  8  21.7
+#>  9  25.6
+#> 10  17.1
    
+predict(
   lm_model,
   pred_cars,
   type = "conf_int",
   level = 0.90
-)
    #> # A tibble: 10 x 2
+)
    #> # A tibble: 10 x 2
 #>    .pred_lower .pred_upper
-#>          <dbl>       <dbl>
-#>  1       17.9         29.0
-#>  2       18.1         28.5
-#>  3       24.0         31.3
-#>  4       17.5         25.6
-#>  5       14.3         20.8
-#>  6       17.0         26.2
-#>  7        9.65        18.2
-#>  8       16.2         27.2
-#>  9       14.2         37.0
-#> 10       11.5         22.7
    
-predict(
+#>          <dbl>       <dbl>
+#>  1       17.9         29.0
+#>  2       18.1         28.5
+#>  3       24.0         31.3
+#>  4       17.5         25.6
+#>  5       14.3         20.8
+#>  6       17.0         26.2
+#>  7        9.65        18.2
+#>  8       16.2         27.2
+#>  9       14.2         37.0
+#> 10       11.5         22.7
    
+predict(
   lm_model,
   pred_cars,
   type = "raw",
-  opts = list(type = "terms")
+  opts = list(type = "terms")
 )
    #>             cyl       disp         hp        drat         wt      qsec
 #> 1  -0.001433177 -0.8113275  0.6303467 -0.06120265  2.4139815 -1.567729
 #> 2  -0.001433177 -0.8113275  0.6303467 -0.06120265  1.4488706 -0.736286
@@ -318,6 +323,7 @@ 
    Contents
    
   
    
 
    
 
+
       
    
       
    
   
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
@@ -329,11 +335,14 @@ 
    Contents
    
     Site built by pkgdown.
   
    
 
    
+
       
    
    
 
   
 
+
   
 
 
+
diff --git a/docs/reference/rand_forest.html b/docs/reference/rand_forest.html
index dd273eebe..496636a4f 100644
--- a/docs/reference/rand_forest.html
+++ b/docs/reference/rand_forest.html
@@ -69,6 +69,7 @@
 
  gtag('config', 'UA-115082821-1');
 
+
   
 
   
@@ -88,7 +89,7 @@
         parsnip
         
    
           part of tidymodels
-          0.0.2
+          0.0.3
         
    
       
     
@@ -116,6 +117,9 @@
     
  • 
       Making a parsnip model from scratch
     
    • 
+    
  • 
+      Evaluating submodels with the same model object
+    
    • 
   
 
 
  • 
@@ -132,6 +136,7 @@
 
 
       
+
       
 
 
    
@@ -157,7 +162,7 @@ 
    General Interface for Random Forest Models
    
 time that the model is fit. Other options and argument can be
 set using 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.
    
     
     
    
@@ -166,7 +171,7 @@ 
    General Interface for Random Forest Models
    
   min_n = NULL)
 
 # S3 method for rand_forest
-update(object, mtry = NULL, trees = NULL,
+update(object, mtry = NULL, trees = NULL,
   min_n = NULL, fresh = FALSE, ...)

    • Arguments

    @@ -204,15 +209,15 @@

    Arg

    - +
    type

    A single character value or NULL. Possible values are "numeric", "class", "prob", "conf_int", "pred_int", "quantile", -or "raw". When NULL, predict() will choose an appropriate value +or "raw". When NULL, predict() will choose an appropriate value based on the model's mode.
    ...

    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: "ranger" or "randomForest"

      • +

    • R: "ranger" (the default) or "randomForest"

    • Spark: "spark"

    @@ -220,7 +225,7 @@

    Note

    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 +279,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -293,12 +298,12 @@

    Examp #> Main Arguments: #> mtry = 10 #> min_n = 3 -#>

    update(model, mtry = 1)
    #> Random Forest Model Specification (unknown) +#>
    update(model, mtry = 1)
    #> Random Forest Model Specification (unknown) #> #> Main Arguments: #> mtry = 1 #> min_n = 3 -#>
    update(model, mtry = 1, fresh = TRUE)
    #> Random Forest Model Specification (unknown) +#>
    update(model, mtry = 1, fresh = TRUE)
    #> Random Forest Model Specification (unknown) #> #> Main Arguments: #> mtry = 1 @@ -323,6 +328,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -334,11 +340,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/reexports.html b/docs/reference/reexports.html index c1cd5361b..703390577 100644 --- a/docs/reference/reexports.html +++ b/docs/reference/reexports.html @@ -61,6 +61,7 @@ gtag('config', 'UA-115082821-1'); + @@ -80,7 +81,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -108,6 +109,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -124,6 +128,7 @@ +
    @@ -139,7 +144,7 @@

    Objects exported from other packages

    These objects are imported from other packages. Follow the links below to see their documentation.

    -
    generics

    fit, fit_xy, varying_args

    +
    generics

    fit, fit_xy, varying_args

    magrittr

    %>%

    @@ -157,6 +162,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -168,11 +174,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/rpart_train.html b/docs/reference/rpart_train.html index aa288e7bd..f68b429f1 100644 --- a/docs/reference/rpart_train.html +++ b/docs/reference/rpart_train.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -202,6 +207,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -213,11 +219,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/set_args.html b/docs/reference/set_args.html index 9c8bf428d..544462587 100644 --- a/docs/reference/set_args.html +++ b/docs/reference/set_args.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -197,6 +202,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -208,11 +214,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/set_engine.html b/docs/reference/set_engine.html index 4aaee273a..d5ade5c26 100644 --- a/docs/reference/set_engine.html +++ b/docs/reference/set_engine.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -196,6 +201,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -207,11 +213,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/set_new_model.html b/docs/reference/set_new_model.html new file mode 100644 index 000000000..b21284fa4 --- /dev/null +++ b/docs/reference/set_new_model.html @@ -0,0 +1,356 @@ + + + + + + + + +Tools to Register Models — set_new_model • parsnip + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + + +
    + +
    +
    +
    +

    Tools to Register Models

    + + +
    + +
    + +

    These 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.

    + +
    + + 
    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)
+
+pred_value_template(pre = NULL, post = NULL, func, ...)
    + +

    Arguments

    + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    model

    A single character string for the model type (e.g. +"rand_forest", etc).

    mode

    A single character string for the model mode (e.g. "regression").

    eng

    A single character string for the model engine.

    parsnip

    A single character string for the "harmonized" argument name +that parsnip exposes.

    original

    A single character string for the argument name that +underlying model function uses.

    func

    A named character vector that describes how to call +a function. func should have elements pkg and fun. The +former is optional but is recommended and the latter is +required. For example, c(pkg = "stats", fun = "lm") would be +used to invoke the usual linear regression function. In some +cases, it is helpful to use c(fun = "predict") when using a +package's predict method.

    has_submodel

    A single logical for whether the argument +can make predictions on mutiple submodels at once.

    pkg

    An options character string for a package name.

    value

    A list that conforms to the fit_obj or pred_obj description +above, depending on context.

    type

    A single character value for the type of prediction. Possible +values are: class, conf_int, numeric, pred_int, prob, quantile, +and raw.

    pre, post

    Optional functions for pre- and post-processing of prediction +results.

    ...

    Optional arguments that should be passed into the args slot for +prediction objects.

    arg

    A single character string for the model argument name.

    fit_obj

    A list with elements interface, protect, +func and defaults. See the package vignette "Making a +parsnip model from scratch".

    pred_obj

    A list with elements pre, post, func, and args. +See the package vignette "Making a parsnip model from scratch".

    + +

    Details

    + +

    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.

    +

    check_model_exists() checks the model value and ensures that the model has +already been registered. check_model_doesnt_exist() checks the model value +and also checks to see if it is novel in the environment.

    + +

    References

    + +

    "Making a parsnip model from scratch" +https://tidymodels.github.io/parsnip/articles/articles/Scratch.html

    + + +

    Examples

    + 
    # set_new_model("shallow_learning_model")
+
+# 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
+#>
    +
    + +
    + + +
    +
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    +
    + +
    +

    + Developed by Max Kuhn, Davis Vaughan. + Site built by pkgdown. +

    +
    + +
    +
    + + + + + + + + diff --git a/docs/reference/show_call.html b/docs/reference/show_call.html index bdf75f3d4..138699371 100644 --- a/docs/reference/show_call.html +++ b/docs/reference/show_call.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -118,6 +122,7 @@ +
    @@ -162,6 +167,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -173,11 +179,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/surv_reg.html b/docs/reference/surv_reg.html index 8cf118c1f..1b236a988 100644 --- a/docs/reference/surv_reg.html +++ b/docs/reference/surv_reg.html @@ -66,6 +66,7 @@ gtag('config', 'UA-115082821-1'); + @@ -85,7 +86,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -113,6 +114,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -129,6 +133,7 @@ +
    @@ -159,7 +164,7 @@

    General Interface for Parametric Survival Models

    surv_reg(mode = "regression", dist = NULL)
 
 # S3 method for surv_reg
-update(object, dist = NULL, fresh = FALSE, ...)
    +update(object, dist = NULL, fresh = FALSE, ...)

    Arguments

    @@ -185,7 +190,7 @@

    Arg

    - +
    ...

    Not used for update().

    Not used for update().
    @@ -195,15 +200,15 @@

    Details to determine the mode of the model. 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)

    Engine Details

    @@ -217,7 +222,7 @@

    survreg

    +

    survival

     survival::survreg(formula = missing_arg(), data = missing_arg(), 
     weights = missing_arg(), model = TRUE)
@@ -233,7 +238,7 @@ 
    R
     
     
    See also
    
 
-    
    varying(), fit(), survival::Surv()
    
+    
    varying(), fit(), survival::Surv()
    
     
 
     
    Examples
    
@@ -249,7 +254,7 @@ 
    Examp
 #> 
 #> Main Arguments:
 #>   dist = weibull
-#>

    update(model, dist = "lnorm")
    #> Parametric Survival Regression Model Specification (regression) +#>
    update(model, dist = "lnorm")
    #> Parametric Survival Regression Model Specification (regression) #> #> Main Arguments: #> dist = lnorm @@ -274,6 +279,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -285,11 +291,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/svm_poly.html b/docs/reference/svm_poly.html index 5a443acc3..bb439af99 100644 --- a/docs/reference/svm_poly.html +++ b/docs/reference/svm_poly.html @@ -70,6 +70,7 @@ gtag('config', 'UA-115082821-1'); + @@ -89,7 +90,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -117,6 +118,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -133,6 +137,7 @@ +
    @@ -159,7 +164,7 @@

    General interface for polynomial support vector machines

    time that the model is fit. Other options and argument can be set using 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.

    @@ -168,7 +173,7 @@

    General interface for polynomial support vector machines

    scale_factor = NULL, margin = NULL) # S3 method for svm_poly -update(object, cost = NULL, degree = NULL, +update(object, cost = NULL, degree = NULL, scale_factor = NULL, margin = NULL, fresh = FALSE, ...)

    Arguments

    @@ -209,15 +214,15 @@

    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: "kernlab"

      • +

    • R: "kernlab" (the default)

    Engine Details

    @@ -238,7 +243,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -257,12 +262,12 @@

    Examp #> Main Arguments: #> cost = 10 #> scale_factor = 0.1 -#>
    update(model, cost = 1)
    #> Polynomial Support Vector Machine Specification (unknown) +#>
    update(model, cost = 1)
    #> Polynomial Support Vector Machine Specification (unknown) #> #> Main Arguments: #> cost = 1 #> scale_factor = 0.1 -#>
    update(model, cost = 1, fresh = TRUE)
    #> Polynomial Support Vector Machine Specification (unknown) +#>
    update(model, cost = 1, fresh = TRUE)
    #> Polynomial Support Vector Machine Specification (unknown) #> #> Main Arguments: #> cost = 1 @@ -285,6 +290,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -296,11 +302,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/svm_rbf.html b/docs/reference/svm_rbf.html index ae59ef985..6f7a4b325 100644 --- a/docs/reference/svm_rbf.html +++ b/docs/reference/svm_rbf.html @@ -70,6 +70,7 @@ gtag('config', 'UA-115082821-1'); + @@ -89,7 +90,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -117,6 +118,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -133,6 +137,7 @@ +
    @@ -159,7 +164,7 @@

    General interface for radial basis function support vector machines

    time that the model is fit. Other options and argument can be set using 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.

    @@ -168,7 +173,7 @@

    General interface for radial basis function support vector machines

    margin = NULL) # S3 method for svm_rbf -update(object, cost = NULL, rbf_sigma = NULL, +update(object, cost = NULL, rbf_sigma = NULL, margin = NULL, fresh = FALSE, ...)

    Arguments

    @@ -205,15 +210,15 @@

    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: "kernlab"

      • +

    • R: "kernlab" (the default)

    Engine Details

    @@ -234,7 +239,7 @@

    See also

    -

    varying(), fit()

    +

    varying(), fit()

    Examples

    @@ -253,12 +258,12 @@

    Examp #> Main Arguments: #> cost = 10 #> rbf_sigma = 0.1 -#>
    update(model, cost = 1)
    #> Radial Basis Function Support Vector Machine Specification (unknown) +#>
    update(model, cost = 1)
    #> Radial Basis Function Support Vector Machine Specification (unknown) #> #> Main Arguments: #> cost = 1 #> rbf_sigma = 0.1 -#>
    update(model, cost = 1, fresh = TRUE)
    #> Radial Basis Function Support Vector Machine Specification (unknown) +#>
    update(model, cost = 1, fresh = TRUE)
    #> Radial Basis Function Support Vector Machine Specification (unknown) #> #> Main Arguments: #> cost = 1 @@ -281,6 +286,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -292,11 +298,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/tidy.model_fit.html b/docs/reference/tidy.model_fit.html index 80ca1eb6b..f20ed312c 100644 --- a/docs/reference/tidy.model_fit.html +++ b/docs/reference/tidy.model_fit.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +

  • @@ -118,6 +122,7 @@ +
    @@ -166,6 +171,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -177,11 +183,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/translate.html b/docs/reference/translate.html index 4fd50043e..ddc26d238 100644 --- a/docs/reference/translate.html +++ b/docs/reference/translate.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -157,7 +162,7 @@

    Details

    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 @@ -183,7 +188,7 @@

    Examp #> #> Model fit template: #> glmnet::glmnet(x = missing_arg(), y = missing_arg(), weights = missing_arg(), -#> lambda = 0.01, family = "gaussian")

    +#> family = "gaussian")
    # `penalty` not applicable for this model. translate(lm_spec, engine = "lm")
    #> Linear Regression Model Specification (regression) #> @@ -231,6 +236,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -242,11 +248,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/type_sum.model_spec.html b/docs/reference/type_sum.model_spec.html index edb82df48..942928721 100644 --- a/docs/reference/type_sum.model_spec.html +++ b/docs/reference/type_sum.model_spec.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -179,6 +184,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -190,11 +196,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/varying.html b/docs/reference/varying.html index 6d638f577..61b3cfe2b 100644 --- a/docs/reference/varying.html +++ b/docs/reference/varying.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -118,6 +122,7 @@ +
    @@ -146,6 +151,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -157,11 +163,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/varying_args.model_spec.html b/docs/reference/varying_args.model_spec.html index 8e46e6a4a..4ae6746a5 100644 --- a/docs/reference/varying_args.model_spec.html +++ b/docs/reference/varying_args.model_spec.html @@ -57,6 +57,7 @@ gtag('config', 'UA-115082821-1'); + @@ -76,7 +77,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -104,6 +105,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -120,6 +124,7 @@ +
    @@ -132,20 +137,20 @@

    Determine varying arguments

    -

    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, ...)

    Arguments

    @@ -182,50 +187,50 @@

    Details

    Examples

    
 # List all possible varying args for the random forest spec
-rand_forest() %>% varying_args()
    #> # A tibble: 3 x 4
+rand_forest() %>% varying_args()
    #> # A tibble: 3 x 4
 #>   name  varying id          type      
-#>   <chr> <lgl>   <chr>       <chr>     
-#> 1 mtry  FALSE   rand_forest model_spec
-#> 2 trees FALSE   rand_forest model_spec
-#> 3 min_n FALSE   rand_forest model_spec
    
+#>   <chr> <lgl>   <chr>       <chr>     
+#> 1 mtry  FALSE   rand_forest model_spec
+#> 2 trees FALSE   rand_forest model_spec
+#> 3 min_n FALSE   rand_forest model_spec
    
 # mtry is now recognized as varying
-rand_forest(mtry = varying()) %>% varying_args()
    #> # A tibble: 3 x 4
+rand_forest(mtry = varying()) %>% varying_args()
    #> # A tibble: 3 x 4
 #>   name  varying id          type      
-#>   <chr> <lgl>   <chr>       <chr>     
-#> 1 mtry  TRUE    rand_forest model_spec
-#> 2 trees FALSE   rand_forest model_spec
-#> 3 min_n FALSE   rand_forest model_spec
    
+#>   <chr> <lgl>   <chr>       <chr>     
+#> 1 mtry  TRUE    rand_forest model_spec
+#> 2 trees FALSE   rand_forest model_spec
+#> 3 min_n FALSE   rand_forest model_spec
    
 # Even engine specific arguments can vary
 rand_forest() %>%
   set_engine("ranger", sample.fraction = varying()) %>%
-  varying_args()
    #> # A tibble: 4 x 4
+  varying_args()
    #> # A tibble: 4 x 4
 #>   name            varying id          type      
-#>   <chr>           <lgl>   <chr>       <chr>     
-#> 1 mtry            FALSE   rand_forest model_spec
-#> 2 trees           FALSE   rand_forest model_spec
-#> 3 min_n           FALSE   rand_forest model_spec
-#> 4 sample.fraction TRUE    rand_forest model_spec
    
+#>   <chr>           <lgl>   <chr>       <chr>     
+#> 1 mtry            FALSE   rand_forest model_spec
+#> 2 trees           FALSE   rand_forest model_spec
+#> 3 min_n           FALSE   rand_forest model_spec
+#> 4 sample.fraction TRUE    rand_forest model_spec
    
 # List only the arguments that actually vary
 rand_forest() %>%
   set_engine("ranger", sample.fraction = varying()) %>%
-  varying_args(full = FALSE)
    #> # A tibble: 1 x 4
+  varying_args(full = FALSE)
    #> # A tibble: 1 x 4
 #>   name            varying id          type      
-#>   <chr>           <lgl>   <chr>       <chr>     
-#> 1 sample.fraction TRUE    rand_forest model_spec
    
+#>   <chr>           <lgl>   <chr>       <chr>     
+#> 1 sample.fraction TRUE    rand_forest model_spec
    
 rand_forest() %>%
   set_engine(
     "randomForest",
     strata = Class,
     sampsize = varying()
   ) %>%
-  varying_args()
    #> # A tibble: 5 x 4
+  varying_args()
    #> # A tibble: 5 x 4
 #>   name     varying id          type      
-#>   <chr>    <lgl>   <chr>       <chr>     
-#> 1 mtry     FALSE   rand_forest model_spec
-#> 2 trees    FALSE   rand_forest model_spec
-#> 3 min_n    FALSE   rand_forest model_spec
-#> 4 strata   FALSE   rand_forest model_spec
-#> 5 sampsize TRUE    rand_forest model_spec
    
+#>   <chr>    <lgl>   <chr>       <chr>     
+#> 1 mtry     FALSE   rand_forest model_spec
+#> 2 trees    FALSE   rand_forest model_spec
+#> 3 min_n    FALSE   rand_forest model_spec
+#> 4 strata   FALSE   rand_forest model_spec
+#> 5 sampsize TRUE    rand_forest model_spec
    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -254,11 +260,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/docs/reference/wa_churn.html b/docs/reference/wa_churn.html index 85a0e4832..cfb1ac282 100644 --- a/docs/reference/wa_churn.html +++ b/docs/reference/wa_churn.html @@ -55,6 +55,7 @@ gtag('config', 'UA-115082821-1'); + @@ -74,7 +75,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -102,6 +103,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -118,6 +122,7 @@ +
    @@ -158,8 +163,8 @@

    Details

    Examples

    - 
    data(wa_churn)
-str(wa_churn)
    #> Classes ‘tbl_df’, ‘tbl’ and 'data.frame':	7043 obs. of  20 variables:
+    
    data(wa_churn)
+str(wa_churn)
    #> Classes ‘tbl_df’, ‘tbl’ and 'data.frame':	7043 obs. of  20 variables:
 #>  $ churn            : Factor w/ 2 levels "Yes","No": 2 2 1 2 1 1 2 2 1 2 ...
 #>  $ female           : num  1 0 0 0 1 1 0 1 1 0 ...
 #>  $ senior_citizen   : int  0 0 0 0 0 0 0 0 0 0 ...
@@ -197,6 +202,7 @@ 
    Contents
    
   
    
 
    
 
+
       
    
       
    
   
    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.
    
@@ -208,11 +214,14 @@ 
    Contents
    
     Site built by pkgdown.
   
    
 
    
+
    + + diff --git a/docs/reference/xgb_train.html b/docs/reference/xgb_train.html index a0fa57a55..454ea325a 100644 --- a/docs/reference/xgb_train.html +++ b/docs/reference/xgb_train.html @@ -56,6 +56,7 @@ gtag('config', 'UA-115082821-1'); + @@ -75,7 +76,7 @@ parsnip
    part of tidymodels - 0.0.2 + 0.0.3
    @@ -103,6 +104,9 @@
  • Making a parsnip model from scratch
    • +
  • + Evaluating submodels with the same model object +
  • @@ -119,6 +123,7 @@ +
    @@ -204,6 +209,7 @@

    Contents

    +

    parsnip is a part of the tidymodels ecosystem, a collection of modeling packages designed with common APIs and a shared philosophy.

    @@ -215,11 +221,14 @@

    Contents

    Site built by pkgdown.

    +
    + + diff --git a/man/linear_reg.Rd b/man/linear_reg.Rd index b58c1d631..791ffb650 100644 --- a/man/linear_reg.Rd +++ b/man/linear_reg.Rd @@ -105,9 +105,9 @@ model, the template of the fit calls are: \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::linear_reg(), "keras")} -When using \code{glmnet} models, there is the option to pass -multiple values (or no values) to the \code{penalty} argument. This -can have an effect on the model object results. When using the +For \code{glmnet} models, the full regularization path is always fit regardless +of the value given to \code{penalty}. Also, there is the option to pass +multiple values (or no values) to the \code{penalty} argument. When using the \code{predict()} method in these cases, the return value depends on the value of \code{penalty}. When using \code{predict()}, only a single value of the penalty can be used. When predicting on multiple diff --git a/man/logistic_reg.Rd b/man/logistic_reg.Rd index 43aa599e9..137da1094 100644 --- a/man/logistic_reg.Rd +++ b/man/logistic_reg.Rd @@ -103,9 +103,9 @@ model, the template of the fit calls are: \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::logistic_reg(), "keras")} -When using \code{glmnet} models, there is the option to pass -multiple values (or no values) to the \code{penalty} argument. This -can have an effect on the model object results. When using the +For \code{glmnet} models, the full regularization path is always fit regardless +of the value given to \code{penalty}. Also, there is the option to pass +multiple values (or no values) to the \code{penalty} argument. When using the \code{predict()} method in these cases, the return value depends on the value of \code{penalty}. When using \code{predict()}, only a single value of the penalty can be used. When predicting on multiple diff --git a/man/min_grid.Rd b/man/min_grid.Rd new file mode 100644 index 000000000..e79250a2b --- /dev/null +++ b/man/min_grid.Rd @@ -0,0 +1,44 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/aaa.R, R/boost_tree.R, R/linear_reg.R, +% R/logistic_reg.R, R/mars.R, R/multinom_reg.R, R/nearest_neighbor.R +\name{min_grid} +\alias{min_grid} +\alias{min_grid.boost_tree} +\alias{min_grid.linear_reg} +\alias{min_grid.logistic_reg} +\alias{min_grid.mars} +\alias{min_grid.multinom_reg} +\alias{min_grid.nearest_neighbor} +\title{Determine the minimum set of model fits} +\usage{ +min_grid(x, grid, ...) + +\method{min_grid}{boost_tree}(x, grid, ...) + +\method{min_grid}{linear_reg}(x, grid, ...) + +\method{min_grid}{logistic_reg}(x, grid, ...) + +\method{min_grid}{mars}(x, grid, ...) + +\method{min_grid}{multinom_reg}(x, grid, ...) + +\method{min_grid}{nearest_neighbor}(x, grid, ...) +} +\arguments{ +\item{x}{A model specification.} + +\item{grid}{A tibble with tuning parameter combinations.} + +\item{...}{Not currently used.} +} +\value{ +A tibble with the minimum tuning parameters to fit and an additional +list column with the parameter combinations used for prediction. +} +\description{ +\code{min_grid} determines exactly what models should be fit in order to +evaluate the entire set of tuning parameter combinations. This is for +internal use only and the API may change in the near future. +} +\keyword{internal} diff --git a/man/multinom_reg.Rd b/man/multinom_reg.Rd index 6f2b4af05..6b711058a 100644 --- a/man/multinom_reg.Rd +++ b/man/multinom_reg.Rd @@ -94,9 +94,9 @@ model, the template of the fit calls are: \Sexpr[results=rd]{parsnip:::show_fit(parsnip:::multinom_reg(), "keras")} -When using \code{glmnet} models, there is the option to pass -multiple values (or no values) to the \code{penalty} argument. This -can have an effect on the model object results. When using the +For \code{glmnet} models, the full regularization path is always fit regardless +of the value given to \code{penalty}. Also, there is the option to pass +multiple values (or no values) to the \code{penalty} argument. When using the \code{predict()} method in these cases, the return value depends on the value of \code{penalty}. When using \code{predict()}, only a single value of the penalty can be used. When predicting on multiple diff --git a/tests/testthat/test_boost_tree_xgboost.R b/tests/testthat/test_boost_tree_xgboost.R index 27f15ee9b..7cbe52750 100644 --- a/tests/testthat/test_boost_tree_xgboost.R +++ b/tests/testthat/test_boost_tree_xgboost.R @@ -207,3 +207,36 @@ test_that('default engine', { ) expect_true(inherits(fit$fit, "xgb.Booster")) }) + +test_that('boosted tree grid reduction', { + reg_grid <- expand.grid(trees = 1:3, learn_rate = (1:5)/5) + reg_grid_smol <- min_grid(boost_tree() %>% set_engine("xgboost"), reg_grid) + + expect_equal(reg_grid_smol$trees, rep(3, 5)) + expect_equal(reg_grid_smol$learn_rate, (1:5)/5) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(trees = 1:2)) + } + + reg_ish_grid <- expand.grid(trees = 1:3, learn_rate = (1:5)/5)[-3,] + reg_ish_grid_smol <- min_grid(boost_tree() %>% set_engine("xgboost"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$trees, c(2, rep(3, 4))) + expect_equal(reg_ish_grid_smol$learn_rate, (1:5)/5) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(trees = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(trees = 1:2)) + } + + reg_grid_extra <- expand.grid(trees = 1:3, learn_rate = (1:5)/5, blah = 10:12) + reg_grid_extra_smol <- min_grid(boost_tree() %>% set_engine("xgboost"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$trees, rep(3, 15)) + expect_equal(reg_grid_extra_smol$learn_rate, rep((1:5)/5, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 5)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(trees = 1:2)) + } + +}) + diff --git a/tests/testthat/test_linear_reg.R b/tests/testthat/test_linear_reg.R index e2830d96a..20416717f 100644 --- a/tests/testthat/test_linear_reg.R +++ b/tests/testthat/test_linear_reg.R @@ -76,7 +76,6 @@ test_that('primary arguments', { x = expr(missing_arg()), y = expr(missing_arg()), weights = expr(missing_arg()), - lambda = new_empty_quosure(1), family = "gaussian" ) ) diff --git a/tests/testthat/test_linear_reg_glmnet.R b/tests/testthat/test_linear_reg_glmnet.R index 46ff3658d..fb2365a13 100644 --- a/tests/testthat/test_linear_reg_glmnet.R +++ b/tests/testthat/test_linear_reg_glmnet.R @@ -67,10 +67,10 @@ test_that('glmnet prediction, single lambda', { y = iris$Sepal.Length ) - uni_pred <- c(5.05124049139868, 4.87103404621362, 4.91028250633598, 4.9399094532023, - 5.08728178043569) + uni_pred <- c(5.05125589060219, 4.86977761622526, 4.90912345599309, 4.93931874108359, + 5.08755154547758) - expect_equal(uni_pred, predict(res_xy, iris[1:5, num_pred])$.pred) + expect_equal(uni_pred, predict(res_xy, iris[1:5, num_pred])$.pred, tolerance = 0.0001) res_form <- fit( iris_basic, @@ -79,10 +79,10 @@ test_that('glmnet prediction, single lambda', { control = ctrl ) - form_pred <- c(5.24228948237804, 5.09448280355765, 5.15636527125752, 5.12592317615935, - 5.26930099973607) + form_pred <- c(5.23960117346944, 5.08769210344022, 5.15129212608077, 5.12000510716518, + 5.26736239856889) - expect_equal(form_pred, predict(res_form, iris[1:5,])$.pred) + expect_equal(form_pred, predict(res_form, iris[1:5,])$.pred, tolerance = 0.0001) }) @@ -132,7 +132,8 @@ test_that('glmnet prediction, multiple lambda', { as.data.frame(mult_pred), multi_predict(res_xy, new_data = iris[1:5, num_pred], lambda = lams) %>% unnest() %>% - as.data.frame() + as.data.frame(), + tolerance = 0.0001 ) res_form <- fit( @@ -176,7 +177,8 @@ test_that('glmnet prediction, multiple lambda', { as.data.frame(form_pred), multi_predict(res_form, new_data = iris[1:5, ], lambda = lams) %>% unnest() %>% - as.data.frame() + as.data.frame(), + tolerance = 0.0001 ) }) @@ -249,7 +251,7 @@ test_that('submodel prediction', { ) reg_fit <- - linear_reg(penalty = c(0, 0.01, 0.1)) %>% + linear_reg() %>% set_engine("glmnet") %>% fit(mpg ~ ., data = mtcars[-(1:4), ]) @@ -274,12 +276,6 @@ test_that('error traps', { skip_if_not_installed("glmnet") - expect_error( - linear_reg(penalty = .1) %>% - set_engine("glmnet") %>% - fit(mpg ~ ., data = mtcars[-(1:4), ]) %>% - predict(mtcars[-(1:4), ], penalty = .2) - ) expect_error( linear_reg() %>% set_engine("glmnet") %>% @@ -295,3 +291,36 @@ test_that('error traps', { }) + +test_that('glmnet grid reduction', { + reg_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5) + reg_grid_smol <- min_grid(linear_reg() %>% set_engine("glmnet"), reg_grid) + + expect_equal(reg_grid_smol$penalty, rep(3, 5)) + expect_equal(reg_grid_smol$mixture, (1:5)/5) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_ish_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5)[-3,] + reg_ish_grid_smol <- min_grid(linear_reg() %>% set_engine("glmnet"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$penalty, c(2, rep(3, 4))) + expect_equal(reg_ish_grid_smol$mixture, (1:5)/5) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(penalty = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_grid_extra <- expand.grid(penalty = 1:3, mixture = (1:5)/5, blah = 10:12) + reg_grid_extra_smol <- min_grid(linear_reg() %>% set_engine("glmnet"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$penalty, rep(3, 15)) + expect_equal(reg_grid_extra_smol$mixture, rep((1:5)/5, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 5)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(penalty = 1:2)) + } + +}) + diff --git a/tests/testthat/test_logistic_reg.R b/tests/testthat/test_logistic_reg.R index c287ccbf6..236d93bd7 100644 --- a/tests/testthat/test_logistic_reg.R +++ b/tests/testthat/test_logistic_reg.R @@ -79,7 +79,6 @@ test_that('primary arguments', { x = expr(missing_arg()), y = expr(missing_arg()), weights = expr(missing_arg()), - lambda = new_empty_quosure(1), family = "binomial" ) ) diff --git a/tests/testthat/test_logistic_reg_glmnet.R b/tests/testthat/test_logistic_reg_glmnet.R index d2db3d5ed..f08392732 100644 --- a/tests/testthat/test_logistic_reg_glmnet.R +++ b/tests/testthat/test_logistic_reg_glmnet.R @@ -62,8 +62,7 @@ test_that('glmnet prediction, one lambda', { uni_pred <- predict(xy_fit$fit, newx = as.matrix(lending_club[1:7, num_pred]), - s = 0.1, type = "response") - uni_pred <- predict(xy_fit$fit, newx = as.matrix(lending_club[1:7, num_pred]), type = "response") + s = 0.1, type = "response")[,1] uni_pred <- ifelse(uni_pred >= 0.5, "good", "bad") uni_pred <- factor(uni_pred, levels = levels(lending_club$Class)) uni_pred <- unname(uni_pred) @@ -83,7 +82,7 @@ test_that('glmnet prediction, one lambda', { form_pred <- predict(res_form$fit, newx = form_mat, - s = 0.1) + s = 0.1, type = "response")[,1] form_pred <- ifelse(form_pred >= 0.5, "good", "bad") form_pred <- factor(form_pred, levels = levels(lending_club$Class)) form_pred <- unname(form_pred) @@ -357,7 +356,7 @@ test_that('glmnet probabilities, no lambda', { expect_equal( mult_pred, multi_predict(xy_fit, lending_club[1:7, num_pred], type = "prob") %>% unnest() - ) + ) res_form <- fit( logistic_reg() %>% set_engine("glmnet"), @@ -419,4 +418,35 @@ test_that('submodel prediction', { }) +test_that('glmnet grid reduction', { + reg_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5) + reg_grid_smol <- min_grid(logistic_reg() %>% set_engine("glmnet"), reg_grid) + + expect_equal(reg_grid_smol$penalty, rep(3, 5)) + expect_equal(reg_grid_smol$mixture, (1:5)/5) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_ish_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5)[-3,] + reg_ish_grid_smol <- min_grid(logistic_reg() %>% set_engine("glmnet"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$penalty, c(2, rep(3, 4))) + expect_equal(reg_ish_grid_smol$mixture, (1:5)/5) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(penalty = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_grid_extra <- expand.grid(penalty = 1:3, mixture = (1:5)/5, blah = 10:12) + reg_grid_extra_smol <- min_grid(logistic_reg() %>% set_engine("glmnet"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$penalty, rep(3, 15)) + expect_equal(reg_grid_extra_smol$mixture, rep((1:5)/5, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 5)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(penalty = 1:2)) + } + +}) diff --git a/tests/testthat/test_logistic_reg_spark.R b/tests/testthat/test_logistic_reg_spark.R index 6c47487c8..a225a3df7 100644 --- a/tests/testthat/test_logistic_reg_spark.R +++ b/tests/testthat/test_logistic_reg_spark.R @@ -87,3 +87,14 @@ test_that('spark execution', { }) + +test_that('spark grid reduction', { + reg_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5) + reg_grid_smol <- min_grid(logistic_reg() %>% set_engine("spark"), reg_grid) + + expect_equal(reg_grid_smol$penalty, reg_grid$penalty) + expect_equal(reg_grid_smol$mixture, reg_grid$mixture) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list()) + } +}) diff --git a/tests/testthat/test_mars.R b/tests/testthat/test_mars.R index b17d46ac7..6c9b88f6c 100644 --- a/tests/testthat/test_mars.R +++ b/tests/testthat/test_mars.R @@ -286,3 +286,34 @@ test_that('classification', { expect_equal(parsnip_pred$.pred_good, earth_pred) }) +test_that('earth grid reduction', { + reg_grid <- expand.grid(num_terms = 1:3, prod_degree = 1:2) + reg_grid_smol <- min_grid(mars() %>% set_engine("earth"), reg_grid) + + expect_equal(reg_grid_smol$num_terms, rep(3, 2)) + expect_equal(reg_grid_smol$prod_degree, 1:2) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(num_terms = 1:2)) + } + + reg_ish_grid <- expand.grid(num_terms = 1:3, prod_degree = 1:2)[-3,] + reg_ish_grid_smol <- min_grid(mars() %>% set_engine("earth"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$num_terms, 2:3) + expect_equal(reg_ish_grid_smol$prod_degree, 1:2) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(num_terms = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(num_terms = 1:2)) + } + + reg_grid_extra <- expand.grid(num_terms = 1:3, prod_degree = 1:2, blah = 10:12) + reg_grid_extra_smol <- min_grid(mars() %>% set_engine("earth"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$num_terms, rep(3, 6)) + expect_equal(reg_grid_extra_smol$prod_degree, rep(1:2, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 2)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(num_terms = 1:2)) + } + +}) diff --git a/tests/testthat/test_multinom_reg.R b/tests/testthat/test_multinom_reg.R index 31b8c72bb..eba11a0b6 100644 --- a/tests/testthat/test_multinom_reg.R +++ b/tests/testthat/test_multinom_reg.R @@ -40,7 +40,6 @@ test_that('primary arguments', { x = expr(missing_arg()), y = expr(missing_arg()), weights = expr(missing_arg()), - lambda = new_empty_quosure(1), family = "multinomial" ) ) diff --git a/tests/testthat/test_multinom_reg_glmnet.R b/tests/testthat/test_multinom_reg_glmnet.R index 4115abbda..b612b44ef 100644 --- a/tests/testthat/test_multinom_reg_glmnet.R +++ b/tests/testthat/test_multinom_reg_glmnet.R @@ -61,7 +61,6 @@ test_that('glmnet prediction, one lambda', { uni_pred <- factor(uni_pred[,1], levels = levels(iris$Species)) uni_pred <- unname(uni_pred) - expect_equal(uni_pred, parsnip:::predict_class.model_fit(xy_fit, iris[rows, 1:4])) expect_equal(uni_pred, predict(xy_fit, iris[rows, 1:4], type = "class")$.pred_class) res_form <- fit( @@ -151,4 +150,38 @@ test_that('glmnet probabilities, mulitiple lambda', { }) +# ------------------------------------------------------------------------------ + +test_that('glmnet grid reduction', { + reg_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5) + reg_grid_smol <- min_grid(multinom_reg() %>% set_engine("glmnet"), reg_grid) + + expect_equal(reg_grid_smol$penalty, rep(3, 5)) + expect_equal(reg_grid_smol$mixture, (1:5)/5) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_ish_grid <- expand.grid(penalty = 1:3, mixture = (1:5)/5)[-3,] + reg_ish_grid_smol <- min_grid(multinom_reg() %>% set_engine("glmnet"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$penalty, c(2, rep(3, 4))) + expect_equal(reg_ish_grid_smol$mixture, (1:5)/5) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(penalty = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(penalty = 1:2)) + } + + reg_grid_extra <- expand.grid(penalty = 1:3, mixture = (1:5)/5, blah = 10:12) + reg_grid_extra_smol <- min_grid(multinom_reg() %>% set_engine("glmnet"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$penalty, rep(3, 15)) + expect_equal(reg_grid_extra_smol$mixture, rep((1:5)/5, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 5)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(penalty = 1:2)) + } + +}) + diff --git a/tests/testthat/test_nearest_neighbor_kknn.R b/tests/testthat/test_nearest_neighbor_kknn.R index 3a0039b70..9971b1c64 100644 --- a/tests/testthat/test_nearest_neighbor_kknn.R +++ b/tests/testthat/test_nearest_neighbor_kknn.R @@ -191,3 +191,35 @@ test_that('kknn multi-predict', { dplyr::select(.pred) expect_equal(pred_uni, pred_uni_obs) }) + +test_that('kkknn grid reduction', { + reg_grid <- expand.grid(neighbors = 1:3, prod_degree = 1:2) + reg_grid_smol <- min_grid(nearest_neighbor() %>% set_engine("kknn"), reg_grid) + + expect_equal(reg_grid_smol$neighbors, rep(3, 2)) + expect_equal(reg_grid_smol$prod_degree, 1:2) + for (i in 1:nrow(reg_grid_smol)) { + expect_equal(reg_grid_smol$.submodels[[i]], list(neighbors = 1:2)) + } + + reg_ish_grid <- expand.grid(neighbors = 1:3, prod_degree = 1:2)[-3,] + reg_ish_grid_smol <- min_grid(nearest_neighbor() %>% set_engine("kknn"), reg_ish_grid) + + expect_equal(reg_ish_grid_smol$neighbors, 2:3) + expect_equal(reg_ish_grid_smol$prod_degree, 1:2) + expect_equal(reg_ish_grid_smol$.submodels[[1]], list(neighbors = 1)) + for (i in 2:nrow(reg_ish_grid_smol)) { + expect_equal(reg_ish_grid_smol$.submodels[[i]], list(neighbors = 1:2)) + } + + reg_grid_extra <- expand.grid(neighbors = 1:3, prod_degree = 1:2, blah = 10:12) + reg_grid_extra_smol <- min_grid(nearest_neighbor() %>% set_engine("kknn"), reg_grid_extra) + + expect_equal(reg_grid_extra_smol$neighbors, rep(3, 6)) + expect_equal(reg_grid_extra_smol$prod_degree, rep(1:2, each = 3)) + expect_equal(reg_grid_extra_smol$blah, rep(10:12, 2)) + for (i in 1:nrow(reg_grid_extra_smol)) { + expect_equal(reg_grid_extra_smol$.submodels[[i]], list(neighbors = 1:2)) + } + +}) diff --git a/vignettes/articles/Submodels.Rmd b/vignettes/articles/Submodels.Rmd new file mode 100644 index 000000000..aeb22a468 --- /dev/null +++ b/vignettes/articles/Submodels.Rmd @@ -0,0 +1,118 @@ +--- +title: "Evaluating Submodels with the Same Model Object" +vignette: > + %\VignetteEngine{knitr::rmarkdown} + %\VignetteIndexEntry{Evaluating Submodels with the Same Model Object} +output: + knitr:::html_vignette: + toc: yes +--- + +```{r startup, include = FALSE} +library(utils) +library(ggplot2) +theme_set(theme_bw()) +``` + +Some R packages can create predictions from models that are different than the one that was fit. For example, if a boosted tree is fit with 10 iterations of boosting, the model can usually make predictions on _submodels_ that have less than 10 trees (all other parameters being static). This is helpful for model tuning since you can cheap evaluate tuning parameter combinations and can often results in a large speed-up in the computations. + +In `parsnip`, there is a method called `multi_predict()` that can do this. It's current methods are: + +```{r methods} +library(parsnip) +methods("multi_predict") +``` + +We'll use the attrition data in `rsample` to illustrate: + +```{r} +library(tidymodels) +data(attrition) + +set.seed(4595) +data_split <- initial_split(attrition, strata = "Attrition") +attrition_train <- training(data_split) +attrition_test <- testing(data_split) +``` + +A boosted classification tree is one of the most low-maintenance approaches that we could take to these data: + +```{r boost-model} +# requires the xgboost package +attrition_boost <- + boost_tree(mode = "classification", trees = 100) %>% + set_engine("C5.0") +``` + +Suppose that 10-fold cross-validation was being used to tune the model over the number of trees: + +```{r folds} +set.seed(616) +folds <- vfold_cv(attrition_train) +``` + +The process would fit a model on 90% of the data and predict on the remaining 10%. Using `rsample`: + +```{r fold-1} +model_data <- analysis(folds$splits[[1]]) +pred_data <- assessment(folds$splits[[1]]) + +fold_1_model <- + attrition_boost %>% + fit_xy(x = model_data %>% dplyr::select(-Attrition), y = model_data$Attrition) + +``` + +For `multi_predict()`, the same semantics of `predict()` are used but, for this model, there is an extra argument called `trees`. Candidate submodel values can be passed in with `trees`: + +```{r fold-1-pred} + +fold_1_pred <- + multi_predict( + fold_1_model, + new_data = pred_data %>% dplyr::select(-Attrition), + trees = 1:100, + type = "prob" + ) +fold_1_pred +``` + +The results is a tibble that has as many rows as the data being predicted (_n_ = `r nrow(pred_data)`). The `.pred` column contains a list of tibbles and each has the predictions across the different number of trees: + +```{r obs-1-pred} +fold_1_pred$.pred[[1]] +``` + +To get this into a format that is more usable, we can use `tidyr::unnest()` but we first add row numbers so that we can track the predictions by test sample as well as the actual classes: + +```{r unnest} +fold_1_df <- + fold_1_pred %>% + bind_cols(pred_data %>% dplyr::select(Attrition)) %>% + add_rowindex() %>% + unnest(.pred) +fold_1_df +``` + +For two samples, what do these look like over trees? + +```{r prob-plot} +fold_1_df %>% + dplyr::filter(.row %in% c(1, 88)) %>% + ggplot(aes(x = trees, y = .pred_No, col = Attrition, group = .row)) + + geom_step() + + ylim(0:1) + + theme(legend.position = "top") +``` + +What does performance look like over trees (using the area under the ROC curve)? + +```{r auc-plot} +fold_1_df %>% + group_by(trees) %>% + roc_auc(truth = Attrition, .pred_No) %>% + ggplot(aes(x = trees, y = .estimate)) + + geom_step() +``` + +