Predict based on the model
Arguments
- model
The trained model.
- pool
The dataset to predict on.
- prediction_type
Prediction type.
- ...
Additional parameters.
Examples
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)
#> [1] 0.6489797 0.6749884 0.3331358 0.5018237 0.7541328 0.3806107 0.2357996
#> [8] 0.4752205 0.7186610 0.4027660 0.2351567 0.7293696 0.5726987 0.3565523
#> [15] 0.2649454 0.5513967 0.3885718 0.6271439 0.3863992 0.5583793