## ── Conflicts ──────────────────────────────────────────────────── tidymodels_conflicts() ──
+## ── Attaching packages ───────────────────────────── tidymodels 0.1.0.9000 ──
+## ✓ broom 0.5.6 ✓ rsample 0.0.6.9000
+## ✓ dials 0.0.6.9000 ✓ tibble 3.0.1.9000
+## ✓ dplyr 0.8.5 ✓ tune 0.1.0.9000
+## ✓ infer 0.5.1 ✓ workflows 0.1.1.9000
+## ✓ purrr 0.3.4 ✓ yardstick 0.0.6
+## ✓ recipes 0.1.12
+## ── Conflicts ───────────────────────────────────── tidymodels_conflicts() ──
## x purrr::discard() masks scales::discard()
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
## x dials::margin() masks ggplot2::margin()
## x recipes::step() masks stats::step()
-data(attrition)
+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:
-# requires the xgboost package
+# 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:
-set.seed(616)
+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:
-model_data <- analysis(folds$splits[[1]])
+model_data <- analysis(folds$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:
-fold_1_pred <-
+fold_1_pred <-
multi_predict(
fold_1_model,
new_data = pred_data %>% dplyr::select(-Attrition),
@@ -178,7 +177,7 @@ Evaluating Submodels with the Same Model Object
## 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:
-fold_1_pred$.pred[[1]]
+fold_1_pred$.pred[[1]]
## # A tibble: 100 x 3
## trees .pred_No .pred_Yes
## <int> <dbl> <dbl>
@@ -194,7 +193,7 @@ Evaluating Submodels with the Same Model Object
## 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:
-fold_1_df <-
+fold_1_df <-
fold_1_pred %>%
bind_cols(pred_data %>% dplyr::select(Attrition)) %>%
add_rowindex() %>%
@@ -215,7 +214,7 @@ Evaluating Submodels with the Same Model Object
## 10 10 0.724 0.276 Yes 1
## # … with 11,090 more rows
For two samples, what do these look like over trees?
-fold_1_df %>%
+fold_1_df %>%
dplyr::filter(.row %in% c(1, 88)) %>%
ggplot(aes(x = trees, y = .pred_No, col = Attrition, group = .row)) +
geom_step() +
@@ -223,7 +222,7 @@ Evaluating Submodels with the Same Model Object
theme(legend.position = "top")
What does performance look like over trees (using the area under the ROC curve)?
-fold_1_df %>%
+fold_1_df %>%
group_by(trees) %>%
roc_auc(truth = Attrition, .pred_No) %>%
ggplot(aes(x = trees, y = .estimate)) +
diff --git a/docs/dev/news/index.html b/docs/dev/news/index.html
index 4c4e988db..73319664f 100644
--- a/docs/dev/news/index.html
+++ b/docs/dev/news/index.html
@@ -158,6 +158,8 @@
A new main argument was added to boost_tree() called stop_iter for early stopping. The xgb_train() function gained arguments for early stopping and a percentage of data to leave out for a validation set.
If fit() is used and the underlying model uses a formula, the actual formula is pass to the model (instead of a placeholder). This makes the model call better.
A function named repair_call() was added. This can help change the underlying models call object to better reflect what they would have obtained if the model function had been used directly (instead of via parsnip). This is only useful when the user chooses a formula interface and the model uses a formula interface. It will also be of limited use when a recipes is used to construct the feature set in workflows or tune.
The predict() function now checks to see if required modeling packages are installed. The packages are loaded (but not attached). (#249) (#308) (tidymodels/workflows#45)
The function req_pkgs() is a user interface to determining the required packages. (#308)
diff --git a/docs/dev/pkgdown.css b/docs/dev/pkgdown.css
index 1273238dd..c01e5923b 100644
--- a/docs/dev/pkgdown.css
+++ b/docs/dev/pkgdown.css
@@ -244,14 +244,14 @@ nav[data-toggle='toc'] .nav .nav > .active:focus > a {
.ref-index th {font-weight: normal;}
-.ref-index td {vertical-align: top; min-width: 100px}
+.ref-index td {vertical-align: top;}
.ref-index .icon {width: 40px;}
.ref-index .alias {width: 40%;}
.ref-index-icons .alias {width: calc(40% - 40px);}
.ref-index .title {width: 60%;}
.ref-arguments th {text-align: right; padding-right: 10px;}
-.ref-arguments th, .ref-arguments td {vertical-align: top; min-width: 100px}
+.ref-arguments th, .ref-arguments td {vertical-align: top;}
.ref-arguments .name {width: 20%;}
.ref-arguments .desc {width: 80%;}
diff --git a/docs/dev/pkgdown.yml b/docs/dev/pkgdown.yml
index e6716b06c..40bc87c30 100644
--- a/docs/dev/pkgdown.yml
+++ b/docs/dev/pkgdown.yml
@@ -1,6 +1,6 @@
pandoc: 2.9.2.1
-pkgdown: 1.5.1.9000
-pkgdown_sha: ac78596154e403df5f4e683f2185d88225a0fea6
+pkgdown: 1.5.1
+pkgdown_sha: ~
articles:
Classification: articles/Classification.html
Models: articles/Models.html
@@ -8,7 +8,7 @@ articles:
Scratch: articles/Scratch.html
Submodels: articles/Submodels.html
parsnip_Intro: parsnip_Intro.html
-last_built: 2020-05-27T00:06Z
+last_built: 2020-05-30T23:54Z
urls:
reference: https://parsnip.tidymodels.org/reference
article: https://parsnip.tidymodels.org/articles
diff --git a/docs/dev/reference/boost_tree.html b/docs/dev/reference/boost_tree.html
index 8dd6013bf..254f1d0ed 100644
--- a/docs/dev/reference/boost_tree.html
+++ b/docs/dev/reference/boost_tree.html
@@ -325,34 +325,37 @@ boost_tree() %>%
set_engine("xgboost") %>%
set_mode("regression") %>%
- translate()
## Boosted Tree Model Specification (regression)
-##
-## Computational engine: xgboost
-##
-## Model fit template:
-## parsnip::xgb_train(x = missing_arg(), y = missing_arg(), nthread = 1,
-## verbose = 0)
boost_tree() %>%
+ translate()
## Boosted Tree Model Specification (regression)
+##
+## Computational engine: xgboost
+##
+## Model fit template:
+## parsnip::xgb_train(x = missing_arg(), y = missing_arg(), nthread = 1,
+## verbose = 0)
+
boost_tree() %>%
set_engine("xgboost") %>%
set_mode("classification") %>%
- translate()
## Boosted Tree Model Specification (classification)
-##
-## Computational engine: xgboost
-##
-## Model fit template:
-## parsnip::xgb_train(x = missing_arg(), y = missing_arg(), nthread = 1,
-## verbose = 0)
+ translate()## Boosted Tree Model Specification (classification)
+##
+## Computational engine: xgboost
+##
+## Model fit template:
+## parsnip::xgb_train(x = missing_arg(), y = missing_arg(), nthread = 1,
+## verbose = 0)
+
C5.0
boost_tree() %>%
set_engine("C5.0") %>%
set_mode("classification") %>%
- translate()
## Boosted Tree Model Specification (classification)
-##
-## Computational engine: C5.0
-##
-## Model fit template:
-## parsnip::C5.0_train(x = missing_arg(), y = missing_arg(), weights = missing_arg())
+ translate()
## Boosted Tree Model Specification (classification)
+##
+## Computational engine: C5.0
+##
+## Model fit template:
+## parsnip::C5.0_train(x = missing_arg(), y = missing_arg(), weights = missing_arg())
+
Note that C50::C5.0() does not require factor
predictors to be converted to indicator variables.
@@ -361,22 +364,24 @@ spark
boost_tree() %>%
set_engine("spark") %>%
set_mode("regression") %>%
- translate()
## Boosted Tree Model Specification (regression)
-##
-## Computational engine: spark
-##
-## Model fit template:
-## sparklyr::ml_gradient_boosted_trees(x = missing_arg(), formula = missing_arg(),
-## type = "regression", seed = sample.int(10^5, 1))
boost_tree() %>%
+ translate()
## Boosted Tree Model Specification (regression)
+##
+## Computational engine: spark
+##
+## Model fit template:
+## sparklyr::ml_gradient_boosted_trees(x = missing_arg(), formula = missing_arg(),
+## type = "regression", seed = sample.int(10^5, 1))
+
boost_tree() %>%
set_engine("spark") %>%
set_mode("classification") %>%
- translate()
## Boosted Tree Model Specification (classification)
-##
-## Computational engine: spark
-##
-## Model fit template:
-## sparklyr::ml_gradient_boosted_trees(x = missing_arg(), formula = missing_arg(),
-## type = "classification", seed = sample.int(10^5, 1))
+ translate()## Boosted Tree Model Specification (classification)
+##
+## Computational engine: spark
+##
+## Model fit template:
+## sparklyr::ml_gradient_boosted_trees(x = missing_arg(), formula = missing_arg(),
+## type = "classification", seed = sample.int(10^5, 1))
+
Parameter translations
diff --git a/docs/dev/reference/decision_tree.html b/docs/dev/reference/decision_tree.html
index fe0e95745..ebfa8600f 100644
--- a/docs/dev/reference/decision_tree.html
+++ b/docs/dev/reference/decision_tree.html
@@ -270,20 +270,22 @@ decision_tree() %>%
set_engine("rpart") %>%
set_mode("regression") %>%
- translate()
## Decision Tree Model Specification (regression)
-##
-## Computational engine: rpart
-##
-## Model fit template:
-## rpart::rpart(formula = missing_arg(), data = missing_arg(), weights = missing_arg())
decision_tree() %>%
+ translate()
## Decision Tree Model Specification (regression)
+##
+## Computational engine: rpart
+##
+## Model fit template:
+## rpart::rpart(formula = missing_arg(), data = missing_arg(), weights = missing_arg())
+
decision_tree() %>%
set_engine("rpart") %>%
set_mode("classification") %>%
- translate()
## Decision Tree Model Specification (classification)
-##
-## Computational engine: rpart
-##
-## Model fit template:
-## rpart::rpart(formula = missing_arg(), data = missing_arg(), weights = missing_arg())
+ translate()
## Decision Tree Model Specification (classification)
+##
+## Computational engine: rpart
+##
+## Model fit template:
+## rpart::rpart(formula = missing_arg(), data = missing_arg(), weights = missing_arg())
+
Note that rpart::rpart() does not require factor
predictors to be converted to indicator variables.
@@ -292,13 +294,14 @@ C5.0
decision_tree() %>%
set_engine("C5.0") %>%
set_mode("classification") %>%
- translate()
## Decision Tree Model Specification (classification)
-##
-## Computational engine: C5.0
-##
-## Model fit template:
-## parsnip::C5.0_train(x = missing_arg(), y = missing_arg(), weights = missing_arg(),
-## trials = 1)
+ translate()
## Decision Tree Model Specification (classification)
+##
+## Computational engine: C5.0
+##
+## Model fit template:
+## parsnip::C5.0_train(x = missing_arg(), y = missing_arg(), weights = missing_arg(),
+## trials = 1)
+
Note that C50::C5.0() does not require factor
predictors to be converted to indicator variables.
@@ -307,22 +310,24 @@ spark
decision_tree() %>%
set_engine("spark") %>%
set_mode("regression") %>%
- translate()
## Decision Tree Model Specification (regression)
-##
-## Computational engine: spark
-##
-## Model fit template:
-## sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(),
-## seed = sample.int(10^5, 1))
decision_tree() %>%
+ translate()
## Decision Tree Model Specification (regression)
+##
+## Computational engine: spark
+##
+## Model fit template:
+## sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(),
+## seed = sample.int(10^5, 1))
+
decision_tree() %>%
set_engine("spark") %>%
set_mode("classification") %>%
- translate()
## Decision Tree Model Specification (classification)
-##
-## Computational engine: spark
-##
-## Model fit template:
-## sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(),
-## seed = sample.int(10^5, 1))
+ translate()
## Decision Tree Model Specification (classification)
+##
+## Computational engine: spark
+##
+## Model fit template:
+## sparklyr::ml_decision_tree_classifier(x = missing_arg(), formula = missing_arg(),
+## seed = sample.int(10^5, 1))
+
Parameter translations
diff --git a/docs/dev/reference/descriptors.html b/docs/dev/reference/descriptors.html
index faa7abdaf..6db6fa88f 100644
--- a/docs/dev/reference/descriptors.html
+++ b/docs/dev/reference/descriptors.html
@@ -195,9 +195,9 @@ Details
.obs() = 150
.lvls() = NA (no factor outcome)
.facts() = 1 (the Species predictor)
- .y() = <vector> (Sepal.Width as a vector)
- .x() = <data.frame> (The other 4 columns as a data frame)
- .dat() = <data.frame> (The full data set)
+ .y() = <vector> (Sepal.Width as a vector)
+ .x() = <data.frame> (The other 4 columns as a data frame)
+ .dat() = <data.frame> (The full data set)
If the formula Species ~ . where used:
@@ -206,9 +206,9 @@ Details
.obs() = 150
.lvls() = c(setosa = 50, versicolor = 50, virginica = 50)
.facts() = 0
- .y() = <vector> (Species as a vector)
- .x() = <data.frame> (The other 4 columns as a data frame)
- .dat() = <data.frame> (The full data set)
+ .y() = <vector> (Species as a vector)
+ .x() = <data.frame> (The other 4 columns as a data frame)
+ .dat() = <data.frame> (The full data set)
To use these in a model fit, pass them to a model specification.
diff --git a/docs/dev/reference/fit.html b/docs/dev/reference/fit.html
index 52522745e..309cc94d5 100644
--- a/docs/dev/reference/fit.html
+++ b/docs/dev/reference/fit.html
@@ -261,7 +261,7 @@
Examp
using_formula
#> parsnip model object
#>
-#> Fit time: 26ms
+#> Fit time: 25ms
#>
#> Call: stats::glm(formula = Class ~ funded_amnt + int_rate, family = stats::binomial,
#> data = data)
@@ -274,7 +274,7 @@ Examp
#> Null Deviance: 4055
#> Residual Deviance: 3698 AIC: 3704
using_xy#> parsnip model object
#>
-#> Fit time: 23ms
+#> Fit time: 17ms
#>
#> Call: stats::glm(formula = ..y ~ ., family = stats::binomial, data = data)
#>
diff --git a/docs/dev/reference/get_model_env.html b/docs/dev/reference/get_model_env.html
index e1472fa50..1d114e7ce 100644
--- a/docs/dev/reference/get_model_env.html
+++ b/docs/dev/reference/get_model_env.html
@@ -182,46 +182,53 @@ R
Examples
#> [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" #> [1] "boost_tree" "boost_tree_args"
+#> [3] "boost_tree_encoding" "boost_tree_fit"
+#> [5] "boost_tree_modes" "boost_tree_pkgs"
+#> [7] "boost_tree_predict" "decision_tree"
+#> [9] "decision_tree_args" "decision_tree_encoding"
+#> [11] "decision_tree_fit" "decision_tree_modes"
+#> [13] "decision_tree_pkgs" "decision_tree_predict"
+#> [15] "linear_reg" "linear_reg_args"
+#> [17] "linear_reg_encoding" "linear_reg_fit"
+#> [19] "linear_reg_modes" "linear_reg_pkgs"
+#> [21] "linear_reg_predict" "logistic_reg"
+#> [23] "logistic_reg_args" "logistic_reg_encoding"
+#> [25] "logistic_reg_fit" "logistic_reg_modes"
+#> [27] "logistic_reg_pkgs" "logistic_reg_predict"
+#> [29] "mars" "mars_args"
+#> [31] "mars_encoding" "mars_fit"
+#> [33] "mars_modes" "mars_pkgs"
+#> [35] "mars_predict" "mlp"
+#> [37] "mlp_args" "mlp_encoding"
+#> [39] "mlp_fit" "mlp_modes"
+#> [41] "mlp_pkgs" "mlp_predict"
+#> [43] "models" "modes"
+#> [45] "multinom_reg" "multinom_reg_args"
+#> [47] "multinom_reg_encoding" "multinom_reg_fit"
+#> [49] "multinom_reg_modes" "multinom_reg_pkgs"
+#> [51] "multinom_reg_predict" "nearest_neighbor"
+#> [53] "nearest_neighbor_args" "nearest_neighbor_encoding"
+#> [55] "nearest_neighbor_fit" "nearest_neighbor_modes"
+#> [57] "nearest_neighbor_pkgs" "nearest_neighbor_predict"
+#> [59] "null_model" "null_model_args"
+#> [61] "null_model_encoding" "null_model_fit"
+#> [63] "null_model_modes" "null_model_pkgs"
+#> [65] "null_model_predict" "rand_forest"
+#> [67] "rand_forest_args" "rand_forest_encoding"
+#> [69] "rand_forest_fit" "rand_forest_modes"
+#> [71] "rand_forest_pkgs" "rand_forest_predict"
+#> [73] "surv_reg" "surv_reg_args"
+#> [75] "surv_reg_encoding" "surv_reg_fit"
+#> [77] "surv_reg_modes" "surv_reg_pkgs"
+#> [79] "surv_reg_predict" "svm_poly"
+#> [81] "svm_poly_args" "svm_poly_encoding"
+#> [83] "svm_poly_fit" "svm_poly_modes"
+#> [85] "svm_poly_pkgs" "svm_poly_predict"
+#> [87] "svm_rbf" "svm_rbf_args"
+#> [89] "svm_rbf_encoding" "svm_rbf_fit"
+#> [91] "svm_rbf_modes" "svm_rbf_pkgs"
+#> [93] "svm_rbf_predict"
## Linear Regression Model Specification (regression)
-##
-## Computational engine: lm
-##
-## Model fit template:
-## stats::lm(formula = missing_arg(), data = missing_arg(), weights = missing_arg())
+ translate()