From 4a8b83100a4efee2d6ab8c08fd3e82d0472d234b Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Wed, 27 Feb 2019 15:38:26 -0500
Subject: [PATCH 01/34] changes to #115

---
 .gitignore                                    |  2 +
 NEWS.md                                       |  2 +
 R/fit_helpers.R                               | 42 ++++++++++++-------
 tests/testthat/test_boost_tree_C50.R          | 16 +++++--
 tests/testthat/test_linear_reg_stan.R         | 12 +++---
 tests/testthat/test_logistic_reg.R            | 31 +++++++-------
 tests/testthat/test_logistic_reg_glmnet.R     | 14 +++----
 tests/testthat/test_logistic_reg_stan.R       |  6 ++-
 tests/testthat/test_multinom_reg_glmnet.R     | 14 +++----
 tests/testthat/test_nearest_neighbor_kknn.R   |  7 ++--
 tests/testthat/test_predict_formats.R         | 19 +++++++++
 .../testthat/test_rand_forest_randomForest.R  | 13 +++---
 tests/testthat/test_rand_forest_ranger.R      | 13 +++---
 13 files changed, 119 insertions(+), 72 deletions(-)

diff --git a/.gitignore b/.gitignore
index 058237471..99bec7b71 100644
--- a/.gitignore
+++ b/.gitignore
@@ -5,3 +5,5 @@
 tests/testthat/derby.log
 tests/testthat/logs/
 *.history
+derby.log
+logs/*
diff --git a/NEWS.md b/NEWS.md
index 775ee3b28..53b2a260c 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -13,6 +13,8 @@ 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). 
+
 ## Bug Fixes
 
 * `varying_args()` now uses the version from the `generics` package. This means
diff --git a/R/fit_helpers.R b/R/fit_helpers.R
index 23f8cb9e0..74f614ede 100644
--- a/R/fit_helpers.R
+++ b/R/fit_helpers.R
@@ -6,13 +6,13 @@
 #' @importFrom  stats model.frame model.response terms as.formula model.matrix
 form_form <-
   function(object, control, env, ...) {
-    opts <- quos(...)
 
-    if (object$mode != "regression") {
-      y_levels <- levels_from_formula( # prob rewrite this as simple subset/levels
-        env$formula,
-        env$data
-      )
+    if (object$mode == "classification") {
+      # prob rewrite this as simple subset/levels
+      y_levels <- levels_from_formula(env$formula, env$data)
+      if (!inherits(env$data, "tbl_spark") && is.null(y_levels))
+        stop("For classification models, the outcome should be a factor.",
+             call. =  FALSE)
     } else {
       y_levels <- NULL
     }
@@ -20,7 +20,7 @@ form_form <-
     object <- check_mode(object, y_levels)
 
     # if descriptors are needed, update descr_env with the calculated values
-    if(requires_descrs(object)) {
+    if (requires_descrs(object)) {
       data_stats <- get_descr_form(env$formula, env$data)
       scoped_descrs(data_stats)
     }
@@ -71,8 +71,14 @@ xy_xy <- function(object, env, control, target = "none", ...) {
 
   object <- check_mode(object, levels(env$y))
 
+  if (object$mode == "classification") {
+    if (is.null(levels(env$y)))
+      stop("For classification models, the outcome should be a factor.",
+           call. =  FALSE)
+  }
+
   # if descriptors are needed, update descr_env with the calculated values
-  if(requires_descrs(object)) {
+  if (requires_descrs(object)) {
     data_stats <- get_descr_form(env$formula, env$data)
     scoped_descrs(data_stats)
   }
@@ -125,13 +131,12 @@ form_xy <- function(object, control, env,
   env$x <- data_obj$x
   env$y <- data_obj$y
 
-  res <- list(
-    lvl = levels_from_formula(
-      env$formula,
-      env$data
-    ),
-    spec = object
-  )
+  res <- list(lvl = levels_from_formula(env$formula, env$data), spec = object)
+  if (object$mode == "classification") {
+    if (is.null(res$lvl))
+      stop("For classification models, the outcome should be a factor.",
+           call. =  FALSE)
+  }
 
   res <- xy_xy(
     object = object,
@@ -148,6 +153,13 @@ form_xy <- function(object, control, env,
 }
 
 xy_form <- function(object, env, control, ...) {
+
+  if (object$mode == "classification") {
+    if (is.null(levels(env$y)))
+      stop("For classification models, the outcome should be a factor.",
+           call. =  FALSE)
+  }
+
   data_obj <-
     convert_xy_to_form_fit(
       x = env$x,
diff --git a/tests/testthat/test_boost_tree_C50.R b/tests/testthat/test_boost_tree_C50.R
index 81e30fd62..e712c5ab1 100644
--- a/tests/testthat/test_boost_tree_C50.R
+++ b/tests/testthat/test_boost_tree_C50.R
@@ -1,6 +1,7 @@
 library(testthat)
 library(parsnip)
 library(tibble)
+library(dplyr)
 
 # ------------------------------------------------------------------------------
 
@@ -8,6 +9,9 @@ context("boosted tree execution with C5.0")
 
 data("lending_club")
 lending_club <- head(lending_club, 200)
+lending_club_fail <-
+  lending_club %>%
+  mutate(bad = Inf, miss = NA)
 num_pred <- c("funded_amnt", "annual_inc", "num_il_tl")
 lc_basic <-
   boost_tree(mode = "classification")  %>%
@@ -41,6 +45,8 @@ test_that('C5.0 execution', {
     ),
     regexp = NA
   )
+
+  # outcome is not a factor:
   expect_error(
     res <- fit(
       lc_basic,
@@ -51,19 +57,21 @@ test_that('C5.0 execution', {
     )
   )
 
+  # Model fails
   C5.0_form_catch <- fit(
     lc_basic,
-    funded_amnt ~ term,
-    data = lending_club,
+    Class ~ miss,
+    data = lending_club_fail,
     control = caught_ctrl
   )
   expect_true(inherits(C5.0_form_catch$fit, "try-error"))
 
+  # Model fails
   C5.0_xy_catch <- fit_xy(
     lc_basic,
     control = caught_ctrl,
-    x = lending_club[, num_pred],
-    y = lending_club$total_bal_il
+    x = lending_club_fail[, "miss"],
+    y = lending_club_fail$Class
   )
   expect_true(inherits(C5.0_xy_catch$fit, "try-error"))
 })
diff --git a/tests/testthat/test_linear_reg_stan.R b/tests/testthat/test_linear_reg_stan.R
index 4891e65e1..08555e0df 100644
--- a/tests/testthat/test_linear_reg_stan.R
+++ b/tests/testthat/test_linear_reg_stan.R
@@ -54,9 +54,9 @@ test_that('stan_glm execution', {
 test_that('stan prediction', {
   skip_if_not_installed("rstanarm")
 
-  uni_stan <- stan_glm(Sepal.Length ~ Sepal.Width + Petal.Width + Petal.Length, data = iris, seed = 123)
+  uni_stan <- rstanarm::stan_glm(Sepal.Length ~ Sepal.Width + Petal.Width + Petal.Length, data = iris, seed = 123)
   uni_pred <- unname(predict(uni_stan, newdata = iris[1:5, ]))
-  inl_stan <- stan_glm(Sepal.Width ~ log(Sepal.Length) + Species, data = iris, seed = 123, chains = 1)
+  inl_stan <- rstanarm::stan_glm(Sepal.Width ~ log(Sepal.Length) + Species, data = iris, seed = 123, chains = 1)
   inl_pred <- unname(predict(inl_stan, newdata = iris[1:5, c("Sepal.Length", "Species")]))
 
   res_xy <- fit_xy(
@@ -103,11 +103,11 @@ test_that('stan intervals', {
             level = 0.93)
 
   prediction_stan <-
-    predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
-                        prob = 0.93)
+    rstanarm::predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
+                                  prob = 0.93)
 
-  stan_post <- posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
-                                 seed = 13)
+  stan_post <- rstanarm::posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
+                                           seed = 13)
   stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
   stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
 
diff --git a/tests/testthat/test_logistic_reg.R b/tests/testthat/test_logistic_reg.R
index c74b0c492..b4d234631 100644
--- a/tests/testthat/test_logistic_reg.R
+++ b/tests/testthat/test_logistic_reg.R
@@ -244,23 +244,24 @@ test_that('glm execution', {
     )
   )
 
-  # passes interactively but not on R CMD check
-  # glm_form_catch <- fit(
-  #   lc_basic,
-  #   funded_amnt ~ term,
-  #   data = lending_club,
-  #
-  #   control = caught_ctrl
-  # )
-  # expect_true(inherits(glm_form_catch$fit, "try-error"))
+  # wrong outcome type
+  expect_error(
+    glm_form_catch <- fit(
+      lc_basic,
+      funded_amnt ~ term,
+      data = lending_club,
+      control = caught_ctrl
+    )
+  )
 
-  glm_xy_catch <- fit_xy(
-    lc_basic,
-    control = caught_ctrl,
-    x = lending_club[, num_pred],
-    y = lending_club$total_bal_il
+  expect_error(
+    glm_xy_catch <- fit_xy(
+      lc_basic,
+      control = caught_ctrl,
+      x = lending_club[, num_pred],
+      y = lending_club$total_bal_il
+    )
   )
-  expect_true(inherits(glm_xy_catch$fit, "try-error"))
 })
 
 test_that('glm prediction', {
diff --git a/tests/testthat/test_logistic_reg_glmnet.R b/tests/testthat/test_logistic_reg_glmnet.R
index e59533209..8dddbb48a 100644
--- a/tests/testthat/test_logistic_reg_glmnet.R
+++ b/tests/testthat/test_logistic_reg_glmnet.R
@@ -34,14 +34,14 @@ test_that('glmnet execution', {
     regexp = NA
   )
 
-  glmnet_xy_catch <- fit_xy(
-    lc_basic,
-    x = lending_club[, num_pred],
-    y = lending_club$total_bal_il,
-    control = caught_ctrl
+  expect_error(
+    glmnet_xy_catch <- fit_xy(
+      lc_basic,
+      x = lending_club[, num_pred],
+      y = lending_club$total_bal_il,
+      control = caught_ctrl
+    )
   )
-  expect_true(inherits(glmnet_xy_catch$fit, "try-error"))
-
 })
 
 test_that('glmnet prediction, one lambda', {
diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index 0a322d1e4..75b8635ce 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -24,6 +24,8 @@ quiet_ctrl <- fit_control(verbosity = 0, catch = TRUE)
 test_that('stan_glm execution', {
   skip_if_not_installed("rstanarm")
 
+  library(rstanarm)
+
   expect_error(
     res <- fit(
       lc_basic,
@@ -149,8 +151,8 @@ test_that('stan intervals', {
             std_error = TRUE)
 
   stan_post <-
-    posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
-                      prob = 0.93, transform = TRUE)
+    rstanarm::posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
+                                prob = 0.93, transform = TRUE)
 
   stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
   stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
diff --git a/tests/testthat/test_multinom_reg_glmnet.R b/tests/testthat/test_multinom_reg_glmnet.R
index 6043aae6e..ab1b1a8d3 100644
--- a/tests/testthat/test_multinom_reg_glmnet.R
+++ b/tests/testthat/test_multinom_reg_glmnet.R
@@ -29,14 +29,14 @@ test_that('glmnet execution', {
     regexp = NA
   )
 
-  glmnet_xy_catch <- fit_xy(
-    multinom_reg() %>% set_engine("glmnet"),
-    x = iris[, 2:5],
-    y = iris$Sepal.Length,
-    ,
-    control = caught_ctrl
+  expect_error(
+    glmnet_xy_catch <- fit_xy(
+      multinom_reg() %>% set_engine("glmnet"),
+      x = iris[, 2:5],
+      y = iris$Sepal.Length,
+      control = caught_ctrl
+    )
   )
-  expect_true(inherits(glmnet_xy_catch$fit, "try-error"))
 
 })
 
diff --git a/tests/testthat/test_nearest_neighbor_kknn.R b/tests/testthat/test_nearest_neighbor_kknn.R
index 1d764d0aa..b0362a5aa 100644
--- a/tests/testthat/test_nearest_neighbor_kknn.R
+++ b/tests/testthat/test_nearest_neighbor_kknn.R
@@ -20,6 +20,7 @@ quiet_ctrl <- fit_control(verbosity = 0, catch = TRUE)
 test_that('kknn execution', {
 
   skip_if_not_installed("kknn")
+  library(kknn)
 
   # continuous
   # expect no error
@@ -78,7 +79,7 @@ test_that('kknn prediction', {
 
   # nominal
   res_xy_nom <- fit_xy(
-    iris_basic,
+    iris_basic %>% set_mode("classification"),
     control = ctrl,
     x = iris[, c("Sepal.Length", "Petal.Width")],
     y = iris$Species
@@ -89,11 +90,11 @@ test_that('kknn prediction', {
     newdata = iris[1:5, c("Sepal.Length", "Petal.Width")]
   )
 
-  expect_equal(uni_pred_nom, predict_class(res_xy_nom, iris[1:5, c("Sepal.Length", "Petal.Width")]))
+  expect_equal(uni_pred_nom, parsnip:::predict_class(res_xy_nom, iris[1:5, c("Sepal.Length", "Petal.Width")]))
 
   # continuous - formula interface
   res_form <- fit(
-    iris_basic,
+    iris_basic %>% set_mode("regression"),
     Sepal.Length ~ log(Sepal.Width) + Species,
     data = iris,
     control = ctrl
diff --git a/tests/testthat/test_predict_formats.R b/tests/testthat/test_predict_formats.R
index cd10d2add..d12743f4b 100644
--- a/tests/testthat/test_predict_formats.R
+++ b/tests/testthat/test_predict_formats.R
@@ -58,3 +58,22 @@ test_that('non-standard levels', {
   expect_equal(names(predict_classprob(lr_fit_2, new_data = class_dat2[1:5,-1])),
                c("2low", "high+values"))
 })
+
+test_that('non-factor classification', {
+  expect_error(
+    logistic_reg() %>%
+      set_engine("glm") %>%
+      fit(Species ~ ., data = iris %>% mutate(Species = Species == "setosa"))
+  )
+  expect_error(
+    logistic_reg() %>%
+      set_engine("glm") %>%
+      fit(Species ~ ., data = iris %>% mutate(Species = ifelse(Species == "setosa", 1, 0)))
+  )
+  expect_error(
+    multinom_reg() %>%
+      set_engine("glmnet") %>%
+      fit(Species ~ ., data = iris %>% mutate(Species = as.character(Species)))
+  )
+})
+
diff --git a/tests/testthat/test_rand_forest_randomForest.R b/tests/testthat/test_rand_forest_randomForest.R
index 35937b244..888dc05da 100644
--- a/tests/testthat/test_rand_forest_randomForest.R
+++ b/tests/testthat/test_rand_forest_randomForest.R
@@ -66,13 +66,14 @@ test_that('randomForest classification execution', {
   # )
   # expect_true(inherits(randomForest_form_catch$fit, "try-error"))
 
-  randomForest_xy_catch <- fit_xy(
-    bad_rf_cls,
-    x = lending_club[, num_pred],
-    y = lending_club$total_bal_il,
-    control = caught_ctrl
+  expect_error(
+    fit_xy(
+      bad_rf_cls,
+      x = lending_club[, num_pred],
+      y = lending_club$total_bal_il,
+      control = caught_ctrl
+    )
   )
-  expect_true(inherits(randomForest_xy_catch$fit, "try-error"))
 
 })
 
diff --git a/tests/testthat/test_rand_forest_ranger.R b/tests/testthat/test_rand_forest_ranger.R
index 82d767c4b..330afb3e5 100644
--- a/tests/testthat/test_rand_forest_ranger.R
+++ b/tests/testthat/test_rand_forest_ranger.R
@@ -86,10 +86,9 @@ test_that('ranger classification prediction', {
   skip_if_not_installed("ranger")
 
   xy_fit <- fit_xy(
-    rand_forest() %>% set_engine("ranger"),
+    rand_forest() %>% set_mode("classification") %>% set_engine("ranger"),
     x = lending_club[, num_pred],
     y = lending_club$Class,
-
     control = ctrl
   )
 
@@ -99,7 +98,7 @@ test_that('ranger classification prediction', {
   expect_equal(xy_pred, predict_class(xy_fit, new_data = lending_club[1:6, num_pred]))
 
   form_fit <- fit(
-    rand_forest() %>% set_engine("ranger"),
+    rand_forest() %>% set_mode("classification") %>% set_engine("ranger"),
     Class ~ funded_amnt + int_rate,
     data = lending_club,
 
@@ -119,7 +118,7 @@ test_that('ranger classification probabilities', {
   skip_if_not_installed("ranger")
 
   xy_fit <- fit_xy(
-    rand_forest() %>% set_engine("ranger", seed = 3566),
+    rand_forest() %>% set_mode("classification") %>% set_engine("ranger", seed = 3566),
     x = lending_club[, num_pred],
     y = lending_club$Class,
 
@@ -134,7 +133,7 @@ test_that('ranger classification probabilities', {
   expect_equivalent(xy_pred[1,], one_row)
 
   form_fit <- fit(
-    rand_forest()  %>% set_engine("ranger", seed = 3566),
+    rand_forest() %>% set_mode("classification") %>% set_engine("ranger", seed = 3566),
     Class ~ funded_amnt + int_rate,
     data = lending_club,
 
@@ -149,7 +148,6 @@ test_that('ranger classification probabilities', {
     rand_forest() %>% set_engine("ranger", probability = FALSE),
     x = lending_club[, num_pred],
     y = lending_club$Class,
-
     control = ctrl
   )
 
@@ -348,7 +346,7 @@ test_that('ranger classification prediction', {
   skip_if_not_installed("ranger")
 
   xy_class_fit <-
-    rand_forest()  %>% set_engine("ranger") %>%
+    rand_forest() %>% set_mode("classification")  %>% set_engine("ranger") %>%
     fit_xy(
       x = iris[, 1:4],
       y = iris$Species,
@@ -366,6 +364,7 @@ test_that('ranger classification prediction', {
 
   xy_prob_fit <-
     rand_forest() %>%
+    set_mode("classification") %>%
     set_engine("ranger") %>%
     fit_xy(
       x = iris[, 1:4],

From a620edb24d9496b7dbe575c157d45c1f5595f433 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Wed, 27 Feb 2019 16:03:41 -0500
Subject: [PATCH 02/34] testing out adding modeling pacakges to suggests

---
 DESCRIPTION | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/DESCRIPTION b/DESCRIPTION
index 61788129c..d21f8279a 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -38,4 +38,12 @@ Suggests:
     keras,
     xgboost,
     covr,
-    sparklyr
+    sparklyr,
+    earth,
+    glmnet,
+    kernlab,
+    kknn,
+    randomForest,
+    ranger,
+    rpart,
+    rstanarm

From 147eea859c9d22d562ba3a51a3a02228f9ac66c3 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Wed, 27 Feb 2019 19:57:47 -0500
Subject: [PATCH 03/34] first attempt at getting around CXX14 travis issues

---
 .travis.yml | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/.travis.yml b/.travis.yml
index ebf585975..b76000fb7 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -14,6 +14,14 @@ r:
 - release
 - devel
 
+
+addons:
+  apt:
+    sources:
+      - ubuntu-toolchain-r-test
+    packages:
+      - g++-6
+
 env:
   global:
   - KERAS_BACKEND="tensorflow"
@@ -69,6 +77,9 @@ before_install:
   - sudo apt-get -y install libnlopt-dev 
   - sudo apt-get update
   - sudo apt-get -y install python3
+  - sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-6 100
+  - echo "** Overriding src/Makevars and removing C++14 on Travis only"
+  - sed -i 's|CXX_STD = CXX14||' src/Makevars  
 
 after_success:
   - Rscript -e 'covr::codecov()'

From d41bef3206ba5601565cf44e3e107c5a6110ba67 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 11:53:21 -0500
Subject: [PATCH 04/34] second attempt at getting around CXX14 travis issues

---
 .travis.yml | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index b76000fb7..86a1ee1d5 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -22,6 +22,7 @@ addons:
     packages:
       - g++-6
 
+
 env:
   global:
   - KERAS_BACKEND="tensorflow"
@@ -77,9 +78,7 @@ before_install:
   - sudo apt-get -y install libnlopt-dev 
   - sudo apt-get update
   - sudo apt-get -y install python3
-  - sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-6 100
-  - echo "** Overriding src/Makevars and removing C++14 on Travis only"
-  - sed -i 's|CXX_STD = CXX14||' src/Makevars  
+
 
 after_success:
   - Rscript -e 'covr::codecov()'

From 5d26fec991c9b9e1f98c445a7dac0ffc7e3789cd Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 13:04:57 -0500
Subject: [PATCH 05/34] third attempt at getting around CXX14 travis issues
 (xenial)

---
 .travis.yml | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 86a1ee1d5..ccbf5c385 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,7 +1,7 @@
 # R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
 language: r
 
-dist: trusty
+dist: xenial
 sudo: true
 
 # until generics is finalized
@@ -15,14 +15,6 @@ r:
 - devel
 
 
-addons:
-  apt:
-    sources:
-      - ubuntu-toolchain-r-test
-    packages:
-      - g++-6
-
-
 env:
   global:
   - KERAS_BACKEND="tensorflow"

From b734323fa0cdc8b08506a223fd60538e45f7cb22 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 13:09:31 -0500
Subject: [PATCH 06/34] fourth attempt at getting around CXX14 travis issues

---
 .travis.yml | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/.travis.yml b/.travis.yml
index ccbf5c385..68590848d 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,7 +1,7 @@
 # R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
 language: r
 
-dist: xenial
+dist: trusty
 sudo: true
 
 # until generics is finalized
@@ -25,6 +25,17 @@ matrix:
   allow_failures:
     - r: 3.1
     - r: 3.2
+  include:
+    - os: linux
+      addons:
+        apt:
+          sources:
+            - ubuntu-toolchain-r-test
+          packages:
+            - g++-7
+      env:
+        - MATRIX_EVAL="CC=gcc-7 && CXX=g++-7"
+
 
 r_binary_packages: 
  - rstan

From f9f6d294a6749d6e7132969429a6100da43fbb86 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 13:37:07 -0500
Subject: [PATCH 07/34] fifth attempt at getting around CXX14 travis issues

---
 .travis.yml | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 68590848d..8dc4b064a 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,9 +1,6 @@
 # R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
 language: r
 
-dist: trusty
-sudo: true
-
 # until generics is finalized
 warnings_are_errors: false
 
@@ -25,17 +22,8 @@ matrix:
   allow_failures:
     - r: 3.1
     - r: 3.2
-  include:
-    - os: linux
-      addons:
-        apt:
-          sources:
-            - ubuntu-toolchain-r-test
-          packages:
-            - g++-7
-      env:
-        - MATRIX_EVAL="CC=gcc-7 && CXX=g++-7"
-
+  include:    
+    - dist: xenial
 
 r_binary_packages: 
  - rstan

From 2d3d21f87bc453946296a8cc1bf6e2691a1b84ad Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 14:11:14 -0500
Subject: [PATCH 08/34] sizth attempt at getting around CXX14 travis issues

---
 .travis.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.travis.yml b/.travis.yml
index 8dc4b064a..5a74d4ee5 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -3,6 +3,7 @@ language: r
 
 # until generics is finalized
 warnings_are_errors: false
+sudo: true
 
 r:
 - 3.1

From 71e6178886de1f21a1f5060ee67a14c9231dcaf3 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 15:39:52 -0500
Subject: [PATCH 09/34] abandoning rstanarm on travis

---
 .travis.yml | 3 ---
 DESCRIPTION | 3 +--
 2 files changed, 1 insertion(+), 5 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 5a74d4ee5..cf16b3905 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -27,8 +27,6 @@ matrix:
     - dist: xenial
 
 r_binary_packages: 
- - rstan
- - rstanarm
  - RCurl
  - dplyr
  - glue
@@ -41,7 +39,6 @@ r_binary_packages:
  - scales
  - tibble
  - ggplot2
- - StanHeaders
  - Rcpp
  - RcppEigen
  - BH
diff --git a/DESCRIPTION b/DESCRIPTION
index d21f8279a..1a94c909b 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -45,5 +45,4 @@ Suggests:
     kknn,
     randomForest,
     ranger,
-    rpart,
-    rstanarm
+    rpart

From 8649a6265ed7cb3e1421bc6f86dbfffba6eefe88 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 15:58:52 -0500
Subject: [PATCH 10/34] cleaned the travis house to see what runs and fails

---
 .travis.yml | 38 +-------------------------------------
 1 file changed, 1 insertion(+), 37 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index cf16b3905..053460d8d 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -5,52 +5,16 @@ language: r
 warnings_are_errors: false
 sudo: true
 
-r:
-- 3.1
-- 3.2
-- oldrel
-- release
-- devel
-
-
 env:
   global:
   - KERAS_BACKEND="tensorflow"
   - MAKEFLAGS="-j 2"
 
-# until we troubleshoot these issues
+
 matrix:
-  allow_failures:
-    - r: 3.1
-    - r: 3.2
   include:    
     - dist: xenial
 
-r_binary_packages: 
- - RCurl
- - dplyr
- - glue
- - magrittr
- - stringi
- - stringr
- - munsell
- - rlang
- - reshape2
- - scales
- - tibble
- - ggplot2
- - Rcpp
- - RcppEigen
- - BH
- - glmnet
- - earth
- - sparklyr
- - flexsurv
- - ranger
- - randomforest
- - xgboost
- - C50
-
 cache:
   packages: true
   directories:

From cb1abfb831ad091bdafae3d0f732390733a91085 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 16:35:25 -0500
Subject: [PATCH 11/34] no rstanarm and back to trusty

---
 .travis.yml | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 053460d8d..724bc7e79 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,9 +1,22 @@
 # R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
 language: r
+dist: trusty
+sudo: true
 
 # until generics is finalized
 warnings_are_errors: false
-sudo: true
+
+r:
+- 3.1
+- 3.2
+- oldrel
+- release
+- devel
+
+matrix:
+  allow_failures:
+    - r: 3.1
+    - r: 3.2
 
 env:
   global:
@@ -11,10 +24,6 @@ env:
   - MAKEFLAGS="-j 2"
 
 
-matrix:
-  include:    
-    - dist: xenial
-
 cache:
   packages: true
   directories:

From 26be9296b606f421767c04df0f7f1a84d4420d98 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Thu, 28 Feb 2019 16:42:36 -0500
Subject: [PATCH 12/34] trying to avoid travis failures for devtools
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

```
* installing *source* package ‘devtools’ ...
** package ‘devtools’ successfully unpacked and MD5 sums checked
** R
** inst
** byte-compile and prepare package for lazy loading
Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]) :
  there is no package called ‘magrittr’
ERROR: lazy loading failed for package ‘devtools’
* removing ‘/home/travis/R/Library/devtools’
* restoring previous ‘/home/travis/R/Library/devtools’
The downloaded source packages are in
	‘/tmp/RtmpUhu1j4/downloaded_packages’
Warning message:
In install.packages(c("devtools")) :
  installation of package ‘devtools’ had non-zero exit status
```
---
 .travis.yml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/.travis.yml b/.travis.yml
index 724bc7e79..56e7a85ac 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -18,6 +18,11 @@ matrix:
     - r: 3.1
     - r: 3.2
 
+r_binary_packages: 
+ - devtools
+ - magrittr
+
+
 env:
   global:
   - KERAS_BACKEND="tensorflow"

From 3ffaa345925f4b05bc1550dcb698aae23978a817 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Fri, 1 Mar 2019 12:52:58 -0500
Subject: [PATCH 13/34] back to original travis file

---
 .travis.yml | 40 ++++++++++++++++++++++++++++++----------
 1 file changed, 30 insertions(+), 10 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 56e7a85ac..ae430a7b8 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,5 +1,6 @@
 # R for travis: see documentation at https://docs.travis-ci.com/user/languages/r
 language: r
+
 dist: trusty
 sudo: true
 
@@ -13,21 +14,41 @@ r:
 - release
 - devel
 
+env:
+  global:
+  - KERAS_BACKEND="tensorflow"
+  - MAKEFLAGS="-j 2"
+
+# until we troubleshoot these issues
 matrix:
   allow_failures:
     - r: 3.1
     - r: 3.2
 
 r_binary_packages: 
- - devtools
+  - RCurl
+ - dplyr
+ - glue
  - magrittr
-
-
-env:
-  global:
-  - KERAS_BACKEND="tensorflow"
-  - MAKEFLAGS="-j 2"
-
+ - stringi
+ - stringr
+ - munsell
+ - rlang
+ - reshape2
+ - scales
+ - tibble
+ - ggplot2
+ - Rcpp
+ - RcppEigen
+ - BH
+ - glmnet
+ - earth
+ - sparklyr
+ - flexsurv
+ - ranger
+ - randomforest
+ - xgboost
+ - C50
 
 cache:
   packages: true
@@ -46,6 +67,5 @@ before_install:
   - sudo apt-get update
   - sudo apt-get -y install python3
 
-
 after_success:
-  - Rscript -e 'covr::codecov()'
+  - Rscript -e 'covr::codecov()'
\ No newline at end of file

From 0102dfd758dd758d78e760b30ecdeffb2f3cd8f6 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Fri, 1 Mar 2019 13:01:46 -0500
Subject: [PATCH 14/34] reformatting

---
 .travis.yml | 57 +++++++++++++++++++++++++++--------------------------
 1 file changed, 29 insertions(+), 28 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index ae430a7b8..e480b2df9 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -8,11 +8,11 @@ sudo: true
 warnings_are_errors: false
 
 r:
-- 3.1
-- 3.2
-- oldrel
-- release
-- devel
+  - 3.1
+  - 3.2
+  - oldrel
+  - release
+  - devel
 
 env:
   global:
@@ -27,28 +27,28 @@ matrix:
 
 r_binary_packages: 
   - RCurl
- - dplyr
- - glue
- - magrittr
- - stringi
- - stringr
- - munsell
- - rlang
- - reshape2
- - scales
- - tibble
- - ggplot2
- - Rcpp
- - RcppEigen
- - BH
- - glmnet
- - earth
- - sparklyr
- - flexsurv
- - ranger
- - randomforest
- - xgboost
- - C50
+  - dplyr
+  - glue
+  - magrittr
+  - stringi
+  - stringr
+  - munsell
+  - rlang
+  - reshape2
+  - scales
+  - tibble
+  - ggplot2
+  - Rcpp
+  - RcppEigen
+  - BH
+  - glmnet
+  - earth
+  - sparklyr
+  - flexsurv
+  - ranger
+  - randomforest
+  - xgboost
+  - C50
 
 cache:
   packages: true
@@ -68,4 +68,5 @@ before_install:
   - sudo apt-get -y install python3
 
 after_success:
-  - Rscript -e 'covr::codecov()'
\ No newline at end of file
+  - Rscript -e 'covr::codecov()'
+

From ec54ba84c30965e2def8d57f9a922196f9bb6f41 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 11:18:21 -0500
Subject: [PATCH 15/34] R-devel complains about calling by namespace without
 loading

---
 tests/testthat/test_boost_tree_C50.R    |  3 ++-
 tests/testthat/test_linear_reg_stan.R   | 13 +++++++------
 tests/testthat/test_logistic_reg_stan.R |  5 +++--
 3 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/tests/testthat/test_boost_tree_C50.R b/tests/testthat/test_boost_tree_C50.R
index e712c5ab1..d6a6e30e1 100644
--- a/tests/testthat/test_boost_tree_C50.R
+++ b/tests/testthat/test_boost_tree_C50.R
@@ -116,11 +116,12 @@ test_that('C5.0 probabilities', {
 test_that('submodel prediction', {
 
   skip_if_not_installed("C50")
+  library(C50)
 
   vars <- c("female", "tenure", "total_charges", "phone_service", "monthly_charges")
   class_fit <-
     boost_tree(trees = 20, mode = "classification") %>%
-    set_engine("C5.0", control = C50::C5.0Control(earlyStopping = FALSE)) %>%
+    set_engine("C5.0", control = C5.0Control(earlyStopping = FALSE)) %>%
     fit(churn ~ ., data = wa_churn[-(1:4), c("churn", vars)])
 
   pred_class <- predict(class_fit$fit, wa_churn[1:4, vars], trials = 4, type = "prob")
diff --git a/tests/testthat/test_linear_reg_stan.R b/tests/testthat/test_linear_reg_stan.R
index 08555e0df..87a71bafa 100644
--- a/tests/testthat/test_linear_reg_stan.R
+++ b/tests/testthat/test_linear_reg_stan.R
@@ -53,10 +53,11 @@ test_that('stan_glm execution', {
 
 test_that('stan prediction', {
   skip_if_not_installed("rstanarm")
+  library(rstanarm)
 
-  uni_stan <- rstanarm::stan_glm(Sepal.Length ~ Sepal.Width + Petal.Width + Petal.Length, data = iris, seed = 123)
+  uni_stan <- stan_glm(Sepal.Length ~ Sepal.Width + Petal.Width + Petal.Length, data = iris, seed = 123)
   uni_pred <- unname(predict(uni_stan, newdata = iris[1:5, ]))
-  inl_stan <- rstanarm::stan_glm(Sepal.Width ~ log(Sepal.Length) + Species, data = iris, seed = 123, chains = 1)
+  inl_stan <- stan_glm(Sepal.Width ~ log(Sepal.Length) + Species, data = iris, seed = 123, chains = 1)
   inl_pred <- unname(predict(inl_stan, newdata = iris[1:5, c("Sepal.Length", "Species")]))
 
   res_xy <- fit_xy(
@@ -103,11 +104,11 @@ test_that('stan intervals', {
             level = 0.93)
 
   prediction_stan <-
-    rstanarm::predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
-                                  prob = 0.93)
+    predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
+                        prob = 0.93)
 
-  stan_post <- rstanarm::posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
-                                           seed = 13)
+  stan_post <- posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
+                                 seed = 13)
   stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
   stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
 
diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index 75b8635ce..a5cc62338 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -126,6 +126,7 @@ test_that('stan_glm probability', {
 test_that('stan intervals', {
   skip_if_not_installed("rstanarm")
 
+  library(rstanarm)
   res_form <- fit(
     logistic_reg() %>%
       set_engine("stan", seed = 1333, chains = 1),
@@ -151,8 +152,8 @@ test_that('stan intervals', {
             std_error = TRUE)
 
   stan_post <-
-    rstanarm::posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
-                                prob = 0.93, transform = TRUE)
+    posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
+                      prob = 0.93, transform = TRUE)
 
   stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
   stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)

From faca552969a4a039aa6d28af26cf13e2b3c5de95 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 11:18:48 -0500
Subject: [PATCH 16/34] seventh attempt at getting around CXX14 travis issues

---
 .travis.yml | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index e480b2df9..14f93255e 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -14,12 +14,7 @@ r:
   - release
   - devel
 
-env:
-  global:
-  - KERAS_BACKEND="tensorflow"
-  - MAKEFLAGS="-j 2"
 
-# until we troubleshoot these issues
 matrix:
   allow_failures:
     - r: 3.1
@@ -56,11 +51,23 @@ cache:
     - $HOME/.keras
     - $HOME/.cache/pip
 
+env:
+  global:
+  - KERAS_BACKEND="tensorflow"
+  - MAKEFLAGS="-j 2"
+
+addons:
+  apt:
+    sources:
+      - ubuntu-toolchain-r-test
+    packages:
+      g++-6
 
 before_script:
   - python -m pip install --upgrade --ignore-installed --user travis pip setuptools wheel virtualenv
   - python -m pip install --upgrade --ignore-installed --user travis keras h5py pyyaml requests Pillow scipy theano
   - R -e 'tensorflow::install_tensorflow()'
+  - mkdir -p ~/.R && echo "CXX14=g++-6` > ~/.R/Makevars
 
 before_install:
   - sudo apt-get -y install libnlopt-dev 

From 98db3086da409d22d3d4e77b9188eb65a21af28b Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 11:34:13 -0500
Subject: [PATCH 17/34] eigth attempt at getting around CXX14 travis issues

---
 .travis.yml | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/.travis.yml b/.travis.yml
index 14f93255e..23010c5ee 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -64,10 +64,12 @@ addons:
       g++-6
 
 before_script:
+  - mkdir -p ~/.R && echo "CXX14=g++-6` > ~/.R/Makevars
+  - cd ~
   - python -m pip install --upgrade --ignore-installed --user travis pip setuptools wheel virtualenv
   - python -m pip install --upgrade --ignore-installed --user travis keras h5py pyyaml requests Pillow scipy theano
   - R -e 'tensorflow::install_tensorflow()'
-  - mkdir -p ~/.R && echo "CXX14=g++-6` > ~/.R/Makevars
+
 
 before_install:
   - sudo apt-get -y install libnlopt-dev 

From d0b77699eff77bfd2ff766ebdff588ee5ea8564c Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 13:30:51 -0500
Subject: [PATCH 18/34] bad backtick!

---
 .travis.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.travis.yml b/.travis.yml
index 23010c5ee..20a4bb779 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -64,7 +64,7 @@ addons:
       g++-6
 
 before_script:
-  - mkdir -p ~/.R && echo "CXX14=g++-6` > ~/.R/Makevars
+  - mkdir -p ~/.R && echo "CXX14=g++-6" > ~/.R/Makevars
   - cd ~
   - python -m pip install --upgrade --ignore-installed --user travis pip setuptools wheel virtualenv
   - python -m pip install --upgrade --ignore-installed --user travis keras h5py pyyaml requests Pillow scipy theano

From e8eee0008ff20c70dea711190528dfcd41901d7a Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 13:47:37 -0500
Subject: [PATCH 19/34] moved makevars stuff to before_install

---
 .travis.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 20a4bb779..f7d8c1934 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -64,8 +64,6 @@ addons:
       g++-6
 
 before_script:
-  - mkdir -p ~/.R && echo "CXX14=g++-6" > ~/.R/Makevars
-  - cd ~
   - python -m pip install --upgrade --ignore-installed --user travis pip setuptools wheel virtualenv
   - python -m pip install --upgrade --ignore-installed --user travis keras h5py pyyaml requests Pillow scipy theano
   - R -e 'tensorflow::install_tensorflow()'
@@ -75,6 +73,8 @@ before_install:
   - sudo apt-get -y install libnlopt-dev 
   - sudo apt-get update
   - sudo apt-get -y install python3
+  - mkdir -p ~/.R && echo "CXX14=g++-6" > ~/.R/Makevars
+ 
 
 after_success:
   - Rscript -e 'covr::codecov()'

From eb3221c3e1faf23bee6223e578259629b358b89c Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 14:03:44 -0500
Subject: [PATCH 20/34] adding C50 and rstanarm back into travis builds

---
 DESCRIPTION | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/DESCRIPTION b/DESCRIPTION
index 1a94c909b..ce249f6be 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -38,6 +38,8 @@ Suggests:
     keras,
     xgboost,
     covr,
+    C50,
+    rstanarm,
     sparklyr,
     earth,
     glmnet,

From fa874d2e198e4176cb91c55cfe1a67dffd8868ac Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 14:30:55 -0500
Subject: [PATCH 21/34] get binaries for stan packages

---
 .travis.yml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.travis.yml b/.travis.yml
index f7d8c1934..91d591718 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -44,6 +44,9 @@ r_binary_packages:
   - randomforest
   - xgboost
   - C50
+  - rstan
+  - StanHeaders
+  - rstanarm
 
 cache:
   packages: true

From 5fbcf2d956c4285f5473b1c284a3c31ec86a4819 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 14:58:53 -0500
Subject: [PATCH 22/34] CXX14FLAGS for rstan

---
 .travis.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.travis.yml b/.travis.yml
index 91d591718..49d889b04 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -77,6 +77,7 @@ before_install:
   - sudo apt-get update
   - sudo apt-get -y install python3
   - mkdir -p ~/.R && echo "CXX14=g++-6" > ~/.R/Makevars
+  - echo "CXX14FLAGS += -fPIC" >> ~/.R/Makevars
  
 
 after_success:

From 75796734955848bd7e28eb73ea45d34732fde4e8 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 15:45:51 -0500
Subject: [PATCH 23/34] more binaries

---
 .travis.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.travis.yml b/.travis.yml
index 49d889b04..d1b91dd73 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -47,6 +47,7 @@ r_binary_packages:
   - rstan
   - StanHeaders
   - rstanarm
+  - RcppEigen
 
 cache:
   packages: true

From cb864e101eee6c8106d079c65a0580dc3abd8e31 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 16:11:40 -0500
Subject: [PATCH 24/34] no stan

---
 .travis.yml                             | 5 +----
 tests/testthat/test_linear_reg_stan.R   | 1 -
 tests/testthat/test_logistic_reg_stan.R | 2 --
 3 files changed, 1 insertion(+), 7 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index d1b91dd73..e45533ed3 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -44,10 +44,7 @@ r_binary_packages:
   - randomforest
   - xgboost
   - C50
-  - rstan
-  - StanHeaders
-  - rstanarm
-  - RcppEigen
+
 
 cache:
   packages: true
diff --git a/tests/testthat/test_linear_reg_stan.R b/tests/testthat/test_linear_reg_stan.R
index 87a71bafa..4891e65e1 100644
--- a/tests/testthat/test_linear_reg_stan.R
+++ b/tests/testthat/test_linear_reg_stan.R
@@ -53,7 +53,6 @@ test_that('stan_glm execution', {
 
 test_that('stan prediction', {
   skip_if_not_installed("rstanarm")
-  library(rstanarm)
 
   uni_stan <- stan_glm(Sepal.Length ~ Sepal.Width + Petal.Width + Petal.Length, data = iris, seed = 123)
   uni_pred <- unname(predict(uni_stan, newdata = iris[1:5, ]))
diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index a5cc62338..7ad3e7f15 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -24,8 +24,6 @@ quiet_ctrl <- fit_control(verbosity = 0, catch = TRUE)
 test_that('stan_glm execution', {
   skip_if_not_installed("rstanarm")
 
-  library(rstanarm)
-
   expect_error(
     res <- fit(
       lc_basic,

From e4b4ff6fafb2e186ab9428af149396726911b342 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 16:35:00 -0500
Subject: [PATCH 25/34] no stan (I mean it this time)

---
 DESCRIPTION | 1 -
 1 file changed, 1 deletion(-)

diff --git a/DESCRIPTION b/DESCRIPTION
index ce249f6be..49ce7449e 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -39,7 +39,6 @@ Suggests:
     xgboost,
     covr,
     C50,
-    rstanarm,
     sparklyr,
     earth,
     glmnet,

From 6a52205059d5a034a8a7074782d9a9d822b7b8af Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 17:54:37 -0500
Subject: [PATCH 26/34] one last library call

---
 tests/testthat/test_logistic_reg_stan.R | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index 7ad3e7f15..0a322d1e4 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -124,7 +124,6 @@ test_that('stan_glm probability', {
 test_that('stan intervals', {
   skip_if_not_installed("rstanarm")
 
-  library(rstanarm)
   res_form <- fit(
     logistic_reg() %>%
       set_engine("stan", seed = 1333, chains = 1),

From 0a334b21f3f99e947e5d1b1b359eef6cef84271f Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 20:17:14 -0500
Subject: [PATCH 27/34] fixed test case

---
 tests/testthat/test_logistic_reg_stan.R | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index 0a322d1e4..efead71a7 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -33,13 +33,14 @@ test_that('stan_glm execution', {
     )
   )
 
-  stan_xy_catch <- fit_xy(
-    lc_basic,
-    control = caught_ctrl,
-    x = lending_club[, num_pred],
-    y = lending_club$total_bal_il
+  expect_error(
+    fit_xy(
+      lc_basic,
+      control = caught_ctrl,
+      x = lending_club[, num_pred],
+      y = lending_club$total_bal_il
+    )
   )
-  expect_true(inherits(stan_xy_catch$fit, "try-error"))
 
 })
 

From 1ff8ad25f01ebcb03c3cb5235f6030c34ccd9d8e Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 20:51:19 -0500
Subject: [PATCH 28/34] more changes for r-devel

---
 DESCRIPTION | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/DESCRIPTION b/DESCRIPTION
index 49ce7449e..dc9b84413 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -46,4 +46,6 @@ Suggests:
     kknn,
     randomForest,
     ranger,
-    rpart
+    rpart,
+    MASS,
+    nlme

From d2c5ea4d951254e182130bbeca10620eb8a5f286 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 21:16:05 -0500
Subject: [PATCH 29/34] removed specific top-level stan calls

---
 tests/testthat/test_linear_reg_stan.R   | 23 ++++---
 tests/testthat/test_logistic_reg_stan.R | 87 +++++++++++++++++--------
 2 files changed, 75 insertions(+), 35 deletions(-)

diff --git a/tests/testthat/test_linear_reg_stan.R b/tests/testthat/test_linear_reg_stan.R
index 4891e65e1..b95629df9 100644
--- a/tests/testthat/test_linear_reg_stan.R
+++ b/tests/testthat/test_linear_reg_stan.R
@@ -102,14 +102,21 @@ test_that('stan intervals', {
             type = "pred_int",
             level = 0.93)
 
-  prediction_stan <-
-    predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
-                        prob = 0.93)
-
-  stan_post <- posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
-                                 seed = 13)
-  stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
-  stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
+  # prediction_stan <-
+  #   predictive_interval(res_xy$fit, newdata = iris[1:5, ], seed = 13,
+  #                       prob = 0.93)
+  #
+  # stan_post <- posterior_linpred(res_xy$fit, newdata = iris[1:5, ],
+  #                                seed = 13)
+  # stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
+  # stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
+
+  stan_lower <- c(`1` = 4.93164991101342, `2` = 4.60197941230393,
+                  `3` = 4.6671442757811, `4` = 4.74402724639963,
+                  `5` = 4.99248110476701)
+  stan_upper <- c(`1` = 5.1002837047058, `2` = 4.77617561853506,
+                  `3` = 4.83183673602725, `4` = 4.90844811805409,
+                  `5` = 5.16979395659009)
 
   expect_equivalent(confidence_parsnip$.pred_lower, stan_lower)
   expect_equivalent(confidence_parsnip$.pred_upper, stan_upper)
diff --git a/tests/testthat/test_logistic_reg_stan.R b/tests/testthat/test_logistic_reg_stan.R
index efead71a7..63dbcf3d9 100644
--- a/tests/testthat/test_logistic_reg_stan.R
+++ b/tests/testthat/test_logistic_reg_stan.R
@@ -74,13 +74,16 @@ test_that('stan_glm prediction', {
     control = ctrl
   )
 
-  form_pred <-
-    predict(res_form$fit,
-            newdata = lending_club[1:7, c("funded_amnt", "int_rate")])
-  form_pred <- xy_fit$fit$family$linkinv(form_pred)
-  form_pred <- unname(form_pred)
-  form_pred <- ifelse(form_pred >= 0.5, "good", "bad")
-  form_pred <- factor(form_pred, levels = levels(lending_club$Class))
+  # form_pred <-
+  #   predict(res_form$fit,
+  #           newdata = lending_club[1:7, c("funded_amnt", "int_rate")])
+  # form_pred <- xy_fit$fit$family$linkinv(form_pred)
+  # form_pred <- unname(form_pred)
+  # form_pred <- ifelse(form_pred >= 0.5, "good", "bad")
+  # form_pred <- factor(form_pred, levels = levels(lending_club$Class))
+  form_pred <- structure(c(2L, 2L, 2L, 2L, 2L, 2L, 2L),
+                         .Label = c("bad", "good"),
+                         class = "factor")
   expect_equal(form_pred, predict_class(res_form, lending_club[1:7, c("funded_amnt", "int_rate")]))
 })
 
@@ -113,12 +116,27 @@ test_that('stan_glm probability', {
     control = ctrl
   )
 
+  # form_pred <-
+  #   predict(res_form$fit,
+  #           newdata = lending_club[1:7, c("funded_amnt", "int_rate")])
+  # form_pred <- xy_fit$fit$family$linkinv(form_pred)
+  # form_pred <- tibble(bad = 1 - form_pred, good = form_pred)
   form_pred <-
-    predict(res_form$fit,
-            newdata = lending_club[1:7, c("funded_amnt", "int_rate")])
-  form_pred <- xy_fit$fit$family$linkinv(form_pred)
-  form_pred <- tibble(bad = 1 - form_pred, good = form_pred)
-  expect_equal(form_pred, predict_classprob(res_form, lending_club[1:7, c("funded_amnt", "int_rate")]))
+    tibble::tribble(
+      ~bad,             ~good,
+      0.0451516541621074, 0.954848345837893,
+      0.0663232780491584, 0.933676721950842,
+      0.0425128897715562, 0.957487110228444,
+      0.0442197030195933, 0.955780296980407,
+      0.00135166763321781, 0.998648332366782,
+      0.013776487556396, 0.986223512443604,
+      0.00359938202445076, 0.996400617975549
+    )
+  expect_equal(
+    form_pred %>% as.data.frame(),
+    predict_classprob(res_form, lending_club[1:7, c("funded_amnt", "int_rate")]) %>%
+      as.data.frame()
+  )
 })
 
 
@@ -149,26 +167,41 @@ test_that('stan intervals', {
             level = 0.93,
             std_error = TRUE)
 
-  stan_post <-
-    posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
-                      prob = 0.93, transform = TRUE)
-
-  stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
-  stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
-  stan_std  <- apply(stan_post, 2, sd)
+  # stan_post <-
+  #   posterior_linpred(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
+  #                     prob = 0.93, transform = TRUE)
+  #
+  # stan_lower <- apply(stan_post, 2, quantile, prob = 0.035)
+  # stan_upper <- apply(stan_post, 2, quantile, prob = 0.965)
+  # stan_std  <- apply(stan_post, 2, sd)
+
+  stan_lower <-
+    c(`1` = 0.913925483690233, `2` = 0.841801274737206, `3` = 0.91056642931229,
+      `4` = 0.913619668586545, `5` = 0.987780279394871)
+  stan_upper <-
+    c(`1` = 0.978674663115785, `2` = 0.975178762720162, `3` = 0.984417491942267,
+      `4` = 0.979606072215269, `5` = 0.9999049778978)
+  stan_std <-
+    c(`1` = 0.0181025303127182, `2` = 0.0388665155739319, `3` = 0.0205886091162274,
+      `4` = 0.0181715224502082, `5` = 0.00405145389896896)
 
   expect_equivalent(confidence_parsnip$.pred_lower, stan_lower)
   expect_equivalent(confidence_parsnip$.pred_upper, stan_upper)
   expect_equivalent(confidence_parsnip$.std_error, stan_std)
 
-  stan_pred_post <-
-    posterior_predict(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
-                      prob = 0.93)
-
-  stan_pred_lower <- apply(stan_pred_post, 2, quantile, prob = 0.035)
-  stan_pred_upper <- apply(stan_pred_post, 2, quantile, prob = 0.965)
-  stan_pred_std  <- apply(stan_pred_post, 2, sd)
-
+  # stan_pred_post <-
+  #   posterior_predict(res_form$fit, newdata = lending_club[1:5, ], seed = 13,
+  #                     prob = 0.93)
+  #
+  # stan_pred_lower <- apply(stan_pred_post, 2, quantile, prob = 0.035)
+  # stan_pred_upper <- apply(stan_pred_post, 2, quantile, prob = 0.965)
+  # stan_pred_std  <- apply(stan_pred_post, 2, sd)
+
+  stan_pred_lower <- c(`1` = 0, `2` = 0, `3` = 0, `4` = 0, `5` = 1)
+  stan_pred_upper <- c(`1` = 1, `2` = 1, `3` = 1, `4` = 1, `5` = 1)
+  stan_pred_std  <-
+    c(`1` = 0.211744742168102, `2` = 0.265130711714607, `3` = 0.209589904165081,
+      `4` = 0.198389410902796, `5` = 0.0446989708829856)
   expect_equivalent(prediction_parsnip$.pred_lower, stan_pred_lower)
   expect_equivalent(prediction_parsnip$.pred_upper, stan_pred_upper)
   expect_equivalent(prediction_parsnip$.std_error, stan_pred_std, tolerance = 0.1)

From 837a55538df2017443e81b52fda07362d9d814bc Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 21:16:14 -0500
Subject: [PATCH 30/34] remove ancillary files

---
 .Rbuildignore | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.Rbuildignore b/.Rbuildignore
index 1bf63fb83..387dedcff 100644
--- a/.Rbuildignore
+++ b/.Rbuildignore
@@ -7,3 +7,6 @@
 ^\.Rproj\.user$
 ^.travis.yml$
 ^R/README\.md$
+derby.log
+^logs$
+^tests/testthat/logs$
\ No newline at end of file

From 22407389969daa50701953383bc9ba3102e25090 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 21:41:01 -0500
Subject: [PATCH 31/34] avodi returning multiple logicals

---
 R/multinom_reg.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/multinom_reg.R b/R/multinom_reg.R
index 33fc37918..de9d5938f 100644
--- a/R/multinom_reg.R
+++ b/R/multinom_reg.R
@@ -168,7 +168,7 @@ check_args.multinom_reg <- function(object) {
 
   args <- lapply(object$args, rlang::eval_tidy)
 
-  if (is.numeric(args$penalty) && args$penalty < 0)
+  if (all(is.numeric(args$penalty)) && any(args$penalty < 0))
     stop("The amount of regularization should be >= 0", call. = FALSE)
   if (is.numeric(args$mixture) && (args$mixture < 0 | args$mixture > 1))
     stop("The mixture proportion should be within [0,1]", call. = FALSE)

From 90b220d930449ea4877817b1468519a2bd762ce2 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 22:03:09 -0500
Subject: [PATCH 32/34] back to computing probability results

---
 tests/testthat/test_svm_poly.R | 24 ++++++++++++------------
 tests/testthat/test_svm_rbf.R  | 22 +++++++++++-----------
 2 files changed, 23 insertions(+), 23 deletions(-)

diff --git a/tests/testthat/test_svm_poly.R b/tests/testthat/test_svm_poly.R
index 8de5827c7..075c55b5a 100644
--- a/tests/testthat/test_svm_poly.R
+++ b/tests/testthat/test_svm_poly.R
@@ -245,19 +245,19 @@ test_that('svm poly classification probabilities', {
     )
   expect_equal(cls_form$fit, cls_xy_form$fit)
 
-  # kern_probs <-
-  #   predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
-  #   as_tibble() %>%
-  #   setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))
-
   kern_probs <-
-    structure(
-      list(
-        .pred_setosa = c(0.982990083267231, 0.0167077303224448, 0.00930879923686657),
-        .pred_versicolor = c(0.00417116710624842, 0.946131931665357, 0.0015524073332013),
-        .pred_virginica = c(0.0128387496265202, 0.0371603380121978, 0.989138793429932)),
-      row.names = c(NA,-3L),
-      class = c("tbl_df", "tbl", "data.frame"))
+    predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
+    as_tibble() %>%
+    setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))
+
+  # kern_probs <-
+  #   structure(
+  #     list(
+  #       .pred_setosa = c(0.982990083267231, 0.0167077303224448, 0.00930879923686657),
+  #       .pred_versicolor = c(0.00417116710624842, 0.946131931665357, 0.0015524073332013),
+  #       .pred_virginica = c(0.0128387496265202, 0.0371603380121978, 0.989138793429932)),
+  #     row.names = c(NA,-3L),
+  #     class = c("tbl_df", "tbl", "data.frame"))
 
   parsnip_probs <- predict(cls_form, iris[ind, -5], type = "prob")
   expect_equal(as.data.frame(kern_probs), as.data.frame(parsnip_probs))
diff --git a/tests/testthat/test_svm_rbf.R b/tests/testthat/test_svm_rbf.R
index b4ab329c3..f364b12f0 100644
--- a/tests/testthat/test_svm_rbf.R
+++ b/tests/testthat/test_svm_rbf.R
@@ -221,18 +221,18 @@ test_that('svm rbf classification probabilities', {
     )
   expect_equal(cls_form$fit, cls_xy_form$fit)
 
-  # kern_probs <-
-  #   predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
-  #   as_tibble() %>%
-  #   setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))
-
   kern_probs <-
-    structure(
-      list(
-        .pred_setosa     = c(0.985403715135807, 0.0158818274678279, 0.00633995479908973),
-        .pred_versicolor = c(0.00818691538722139, 0.359005663318986, 0.0173471664171275),
-        .pred_virginica  = c(0.00640936947697121, 0.625112509213187, 0.976312878783783)),
-      row.names = c(NA,-3L), class = c("tbl_df", "tbl", "data.frame"))
+    predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
+    as_tibble() %>%
+    setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))
+
+  # kern_probs <-
+  #   structure(
+  #     list(
+  #       .pred_setosa     = c(0.985403715135807, 0.0158818274678279, 0.00633995479908973),
+  #       .pred_versicolor = c(0.00818691538722139, 0.359005663318986, 0.0173471664171275),
+  #       .pred_virginica  = c(0.00640936947697121, 0.625112509213187, 0.976312878783783)),
+  #     row.names = c(NA,-3L), class = c("tbl_df", "tbl", "data.frame"))
 
   parsnip_probs <- predict(cls_form, iris[ind, -5], type = "prob")
   expect_equal(as.data.frame(kern_probs), as.data.frame(parsnip_probs))

From d63510aada0a497cad11896b8db806b3f0ae3662 Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 22:20:46 -0500
Subject: [PATCH 33/34] added library calls

---
 tests/testthat/test_svm_poly.R | 1 +
 tests/testthat/test_svm_rbf.R  | 1 +
 2 files changed, 2 insertions(+)

diff --git a/tests/testthat/test_svm_poly.R b/tests/testthat/test_svm_poly.R
index 075c55b5a..50a69ebd8 100644
--- a/tests/testthat/test_svm_poly.R
+++ b/tests/testthat/test_svm_poly.R
@@ -245,6 +245,7 @@ test_that('svm poly classification probabilities', {
     )
   expect_equal(cls_form$fit, cls_xy_form$fit)
 
+  library(kernlab)
   kern_probs <-
     predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
     as_tibble() %>%
diff --git a/tests/testthat/test_svm_rbf.R b/tests/testthat/test_svm_rbf.R
index f364b12f0..50a3289e2 100644
--- a/tests/testthat/test_svm_rbf.R
+++ b/tests/testthat/test_svm_rbf.R
@@ -221,6 +221,7 @@ test_that('svm rbf classification probabilities', {
     )
   expect_equal(cls_form$fit, cls_xy_form$fit)
 
+  library(kernlab)
   kern_probs <-
     predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
     as_tibble() %>%

From 1404357f8f47985836eb2c4309b7448eec45cf9c Mon Sep 17 00:00:00 2001
From: topepo <mxkuhn@gmail.com>
Date: Sat, 2 Mar 2019 23:08:58 -0500
Subject: [PATCH 34/34] namespace the predict function

---
 tests/testthat/test_svm_poly.R | 2 +-
 tests/testthat/test_svm_rbf.R  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/testthat/test_svm_poly.R b/tests/testthat/test_svm_poly.R
index 50a69ebd8..1835a3f5a 100644
--- a/tests/testthat/test_svm_poly.R
+++ b/tests/testthat/test_svm_poly.R
@@ -247,7 +247,7 @@ test_that('svm poly classification probabilities', {
 
   library(kernlab)
   kern_probs <-
-    predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
+    kernlab::predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
     as_tibble() %>%
     setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))
 
diff --git a/tests/testthat/test_svm_rbf.R b/tests/testthat/test_svm_rbf.R
index 50a3289e2..ba78f284b 100644
--- a/tests/testthat/test_svm_rbf.R
+++ b/tests/testthat/test_svm_rbf.R
@@ -223,7 +223,7 @@ test_that('svm rbf classification probabilities', {
 
   library(kernlab)
   kern_probs <-
-    predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
+    kernlab::predict(cls_form$fit, iris[ind, -5], type = "probabilities") %>%
     as_tibble() %>%
     setNames(c('.pred_setosa', '.pred_versicolor', '.pred_virginica'))