Checkpointed Model Evaluation Across Resamples (`fit_resamples_checkpointed`)

A fingerprinted, resumable wrapper around [tune::fit_resamples()] that evaluates each resample split independently and saves fingerprinted checkpoints into `checkpoint_dir`.

Usage

fit_resamples_checkpointed(
  object,
  resamples,
  metrics = NULL,
  control = tune::control_resamples(),
  checkpoint_dir = NULL,
  shard = NULL,
  checkpoint_mismatch = c("recompute", "overwrite", "error"),
  progress = FALSE,
  ...
)

Arguments

object

A `workflow` or model specification.

resamples

An `rset` resampling object.

metrics

A `yardstick::metric_set()`.

control

A `tune::control_resamples()` object.

checkpoint_dir

Directory to store per-resample fingerprinted checkpoints.

shard

Optional integer vector `c(i, n)` for distributed evaluation.

checkpoint_mismatch

Action when a checkpoint fingerprint mismatches: `"recompute"` (default), `"overwrite"`, or `"error"`.

progress

Logical (default `FALSE`).

...

Additional arguments passed to [tune::fit_resamples()].

Value

A standard `resample_results` object.

See also

[tune_grid_checkpointed()], [checkpoint_map()]