Nothing
R/summary.R
Defines functions summary.fair.model
Documented in summary.fair.model
# create a summary with key information from a fair model.
summary.fair.model = function(object, ...) {
if (!inherits(object, "fair.model"))
stop("'object' must be a 'fair.model' object.")
check.unused.args(list(...), character(0))
# find out what type of model this is.
model = class(object)[1]
# find out how fairness was enforced.
fair = object$fairness
# extract optional argument names and values.
args = object$main$arguments
argnames = names(args)
args.summary = list()
# linear regresssion models.
if ((model %in% fair.regressions) ||
(model %in% fair.family) && (object$main$family == "gaussian")) {
r2 = 1 - var(residuals(object)) / var(fitted(object) + residuals(object))
if ("lr" %in% argnames) {
args.summary = c(args.summary,
list(c("Ridge penalty (sensitive attributes)" = args$lr,
"(predictors)" = args$lambda)))
}#THEN
else if ("lambda" %in% argnames) {
args.summary = c(args.summary, list("Ridge penalty" = args$lambda))
}#THEN
if ("covfun" %in% argnames)
args.summary = c(args.summary,
list("Custom covariance matrix" = !identical(args$covfun, cov)))
perf.summary = list(
"Log-likelihood" = logLik(object),
"Residual standard error" = sigma(object),
"Multiple R^2" = r2
)
}#THEN
else if ((model %in% fair.classifiers) || (model %in% fair.family)) {
if ("lr" %in% argnames) {
args.summary = c(args.summary,
list(c("Ridge penalty (sensitive attributes)" = args$lr,
"(predictors)" = args$lambda)))
}#THEN
else if ("lambda" %in% argnames) {
args.summary = c(args.summary, list("Ridge penalty" = args$lambda))
}#THEN
perf.summary = list("Log-likelihood" = logLik(object))
}#THEN
else {
stop("no summary() method for model ", q(model), ".")
}#ELSE
# use an ad-hoc label for custom functions, look it up for built-ins.
if (is.function(fair$definition))
fairness.label = "Custom fairness function"
else
fairness.label = fairness.definitions.labels[fair$definition]
if (length(fair$value) == 1) {
fair.summary = list(
structure(c(fair$value, fair$bound),
names = c(fairness.label, "with bound"))
)
}#THEN
else {
fair.summary = structure(list(fair$value, fair$bound),
names = c(fairness.label, "with bound")
)
}#ELSE
structure(list(model = object,
banner = paste("Method:", fair.models.labels[model]),
info = c(args.summary, perf.summary, fair.summary)),
class = "summary.fair.model")
}#SUMMARY.FAIR.MODEL
Try the fairml package in your browser
Any scripts or data that you put into this service are public.
