topepo

#' Check to see if a repo has remote dependencies.
#' @param repos A character vector of repos to query (e.g. `tidyverse/dplyr`). 
#' @param hard_depend Should only depends and imports be included in the query.
#' Using `FALSE` will include suggests and others. 
#' @details Packages listed in `Config/Needs/*` will not be returned. 
#' Requires dplyr, pak, and purrr. 
#' @return A tibble with columns for the repo and any github repos that are
#' listed as dependencies. 
#' @examples
#' tests <- paste0("tidymodels/", c("broom", "parsnip", "recipes"))

topepo

library(tidymodels)
tidymodels_prefer()
theme_set(theme_bw())

library(doMC)
registerDoMC(cores = 20)

# ------------------------------------------------------------------------------

data(ad_data)

topepo

library(tidymodels)
library(rules)
tidymodels_prefer()
theme_set(theme_bw())

library(doMC)
registerDoMC(cores = 20)

# ------------------------------------------------------------------------------

topepo

suppressPackageStartupMessages(library(tidymodels))
library(furrr)

plan(multisession, workers = 20)

data(cells)

set.seed(1)
split <- initial_split(cells, prop = .95)
cell_train <- training(split)

topepo

library(tidymodels)
library(patchwork)
library(doMC)
registerDoMC(cores = 20)

theme_set(theme_bw() + theme(legend.position = "top")) 

# ------------------------------------------------------------------------------

url <- "https://github.com/topepo/cars/raw/master/2019_07_03_city/car_data_splits.RData"

topepo

# pak::pak("tidymodels/parsnip@logistic-liblinear")

library(AppliedPredictiveModeling)
library(tidymodels)
theme_set(theme_bw())

# ------------------------------------------------------------------------------

lr_pull <- function(pen, eng = "glmnet", dat, ...) {
  logistic_reg(penalty = pen, ...) %>%

topepo

library(tidymodels)
library(plsmod)
library(DALEX)

theme_set(theme_bw())

## ── Attaching packages ───────────────────────────────── tidymodels 0.1.1 ──
## ✓ broom     0.7.0      ✓ recipes   0.1.13
## ✓ dials     0.0.8      ✓ rsample   0.0.7 
## ✓ dplyr     1.0.0      ✓ tibble    3.0.3 

topepo

#' @param x A prepped recipe or fitted workflow that uses a recipe. The recipe
#'  must have used at least one `step_pca()`. 
#' @param id A single numeric or character value that is used to pick the step
#'  with the PCA results. If a single `step_pca()` was used, this argument is
#'  ignored. *Note*: if used, `id` must be named. 
#' @param ... An optional series of conditional statements used to filter the 
#'  PCA data before plotting. See Details below. 
#' @examples 
#' library(recipes)
#' library(parsnip)

topepo

## -----------------------------------------------------------------------------
# Simple demonstration of stacking using three models with the Ames housing data

## -----------------------------------------------------------------------------

library(tidymodels)
library(rules)
library(doMC)
library(AmesHousing)
library(ggforce)

topepo

library(tidymodels)
library(Matrix)
library(lobstr)


# ------------------------------------------------------------------------------

rand_values <- function (prefix = "step", len = 3, num_vals = Inf) {
  candidates <- c(letters, LETTERS, paste(0:9))
  candidates <- candidates[1:min(length(candidates), num_vals)]