Classification accuracy (acc)

Computes pointwise and average classification accuracy for binary or multiclass outcomes. The class probabilities are first averaged over the draws, using log_weights if given. For binary outcomes, the averaged probability is thresholded at 0.5. For multiclass outcomes, the predicted class is the most likely category of the averaged probabilities, via which.max().

Usage

measure_acc(y, mupred, log_weights = NULL, pointwise = NULL)

Arguments

y

An integer vector of observed class labels. Code it 0 or 1 when mupred is a matrix, and 1 to K when mupred is a 3-D array.

mupred

A numeric array of posterior predictive means. For binary outcomes use a draws x observations matrix. For multiclass outcomes use a draws x observations x categories array.

log_weights

Optional numeric matrix of unnormalized log-importance weights with dimensions \(S \times n\). Weights are column-normalized before computing each per-observation contribution.

pointwise

Optional numeric vector of precomputed pointwise accuracy contributions. If provided, y, mupred, and log_weights are ignored.

Value

An object of class "measure": a list with:

estimates

A 1 \(\times\) 2 matrix. The row is named after the measure, the columns are Estimate and SE (standard error).

pointwise

An \(n \times 1\) matrix of per-observation values. The column is named after the measure.

extra

Optional. Auxiliary data for the standard error of a difference. Most measures do not set it.

Attributes measure (i.e., measure name) and dims (draws \(\times\) observations) are also set. Use print() for a readable summary.

Examples

y <- c(1L, 0L, 1L)
mupred <- matrix(c(0.8, 0.3, 0.7, 0.6, 0.4, 0.9), nrow = 2)
measure_acc(y, mupred)
#> 
#> Computed from 2 draws by 3 observations.
#> 
#>     Estimate    SE
#> acc    0.667 0.272