test_that("kknn + predict() works", {
  skip_on_cran()
  skip_if_not_installed("kknn")
  suppressPackageStartupMessages(library(kknn))
  m <- dim(iris)[1]
  val <- sample(1:m, size = round(m / 3), replace = FALSE, prob = rep(1 / m, m))
  iris.learn <- iris[-val, ]
  iris.valid <- iris[val, ]
  kknn_fit <- kknn(
    Species ~ .,
    iris.learn,
    iris.valid,
    distance = 1,
    kernel = "triangular"
  )
  x <- axe_env(kknn_fit)
  expect_identical(attr(x$terms, ".Environment"), rlang::base_env())
  x <- axe_fitted(kknn_fit)
  expect_equal(x$fitted.values, list(NULL))
  x <- butcher(kknn_fit)
  new_data <- data.frame(iris[c(1, 10, 13), 1:4])
  expect_equal(
    predict(x, new_data, type = "prob"),
    predict(kknn_fit, new_data, type = "prob")
  )
  expect_error(predict(x, new_data))
})