performance (version 0.8.0)

performance_logloss: Log Loss

Description

Compute the log loss for models with binary outcome.

Usage

performance_logloss(model, verbose = TRUE, ...)

Arguments

model

Model with binary outcome.

verbose

Toggle off warnings.

...

Currently not used.

Value

Numeric, the log loss of model.

Details

Logistic regression models predict the probability of an outcome of being a "success" or "failure" (or 1 and 0 etc.). performance_logloss() evaluates how good or bad the predicted probabilities are. High values indicate bad predictions, while low values indicate good predictions. The lower the log-loss, the better the model predicts the outcome.

See Also

performance_score()

Examples

# NOT RUN {
data(mtcars)
m <- glm(formula = vs ~ hp + wt, family = binomial, data = mtcars)
performance_logloss(m)
# }