Returns either the Evals object itself or a reference to it.
Arguments
- sim
a simulation object
- ...
logical conditions to specify a subset of models. Conditions can only involve params of model that have length 1 and are of class numeric or character.
- subset
a vector of integers indexing the models or a vector of model names. To select models based on parameter values, use
...
. However, using...
is slower than using subset.- index
a vector of positive integers specifying which draws' objects are desired. If missing, then all draws' evals are returned.
- methods
character vector of method names of interest. If missing, then all methods' evals are returned
- reference
whether to return the ModelRef or the Model object itself
Examples
if (FALSE) {
# suppose previously we had run the following:
sim <- new_simulation(name = "normal-example",
label = "Normal Mean Estimation",
dir = tempdir()) %>%
generate_model(make_my_example_model, n = 20) %>%
simulate_from_model(nsim = 50, index = 1:3) %>%
run_method(my_example_method) %>%
evaluate(my_example_loss)
# then we could get the metric evaluated on the method's output:
e <- evals(sim)
# we can export it as a data.frame
as.data.frame(e)
# or we can get at a particular draw-method-metric triplet
e@evals$`my-method`$r1.1$myloss
}