/
wrappers_catboost.R
132 lines (130 loc) 路 2.98 KB
/
wrappers_catboost.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
#' Create a dataset
#'
#' @param data Predictors.
#' @param label Labels.
#' @param ... Additional parameters.
#'
#' @return A `catboost.Pool` object.
#'
#' @export
#'
#' @examplesIf is_installed_catboost()
#' sim_data <- msaenet::msaenet.sim.binomial(
#' n = 100,
#' p = 10,
#' rho = 0.6,
#' coef = rnorm(5, mean = 0, sd = 10),
#' snr = 1,
#' p.train = 0.8,
#' seed = 42
#' )
#'
#' catboost_load_pool(data = sim_data$x.tr, label = sim_data$y.tr)
#' catboost_load_pool(data = sim_data$x.tr, label = NULL)
#' catboost_load_pool(data = sim_data$x.te, label = NULL)
catboost_load_pool <- function(data, label = NULL, ...) {
rlang::check_installed("catboost", reason = "to create a dataset")
cl <- rlang::call2(
"catboost.load_pool",
.ns = "catboost",
data = data,
label = label,
...
)
rlang::eval_tidy(cl)
}
#' Train the model
#'
#' @param learn_pool Training dataset.
#' @param test_pool Testing dataset.
#' @param params A list of training parameters.
#'
#' @return A model object.
#'
#' @export
#'
#' @examplesIf is_installed_catboost()
#' sim_data <- msaenet::msaenet.sim.binomial(
#' n = 100,
#' p = 10,
#' rho = 0.6,
#' coef = rnorm(5, mean = 0, sd = 10),
#' snr = 1,
#' p.train = 0.8,
#' seed = 42
#' )
#'
#' x_train <- catboost_load_pool(data = sim_data$x.tr, label = sim_data$y.tr)
#'
#' fit <- catboost_train(
#' x_train,
#' NULL,
#' params = list(
#' loss_function = "Logloss",
#' iterations = 100,
#' depth = 3,
#' logging_level = "Silent"
#' )
#' )
#'
#' fit
catboost_train <- function(learn_pool, test_pool = NULL, params = list()) {
rlang::check_installed("catboost", reason = "to train the model")
cl <- rlang::call2(
"catboost.train",
.ns = "catboost",
learn_pool = learn_pool,
test_pool = test_pool,
params = params
)
rlang::eval_tidy(cl)
}
#' Predict based on the model
#'
#' @param model The trained model.
#' @param pool The dataset to predict on.
#' @param prediction_type Prediction type.
#' @param ... Additional parameters.
#'
#' @return Predicted values.
#'
#' @export
#'
#' @examplesIf is_installed_catboost()
#' sim_data <- msaenet::msaenet.sim.binomial(
#' n = 100,
#' p = 10,
#' rho = 0.6,
#' coef = rnorm(5, mean = 0, sd = 10),
#' snr = 1,
#' p.train = 0.8,
#' seed = 42
#' )
#'
#' x_train <- catboost_load_pool(data = sim_data$x.tr, label = sim_data$y.tr)
#' x_test <- catboost_load_pool(data = sim_data$x.te, label = NULL)
#'
#' fit <- catboost_train(
#' x_train,
#' NULL,
#' params = list(
#' loss_function = "Logloss",
#' iterations = 100,
#' depth = 3,
#' logging_level = "Silent"
#' )
#' )
#'
#' catboost_predict(fit, x_test)
catboost_predict <- function(model, pool, prediction_type = "Probability", ...) {
rlang::check_installed("catboost", reason = "to predict based on the model")
cl <- rlang::call2(
"catboost.predict",
.ns = "catboost",
model = model,
pool = pool,
prediction_type = prediction_type,
...
)
rlang::eval_tidy(cl)
}