get_auto_filled_metrics#

Trainable.get_auto_filled_metrics(now: datetime | None = None, time_this_iter: float | None = None, timestamp: int | None = None, debug_metrics_only: bool = False) → dict[source]#

Return a dict with metrics auto-filled by the trainable.

If debug_metrics_only is True, only metrics that don’t require at least one iteration will be returned (ray.tune.result.DEBUG_METRICS).

Parameters:
  • now (datetime | None) – The time to report in the date metric, and in timestamp when timestamp isn’t given. Defaults to the current time.

  • time_this_iter (float | None) – The value of the time_this_iter_s metric.

  • timestamp (int | None) – The value of the timestamp metric, in seconds since the epoch. Defaults to now.

  • debug_metrics_only (bool) – Whether to return only the debug metrics.

Returns:

A dict that maps each auto-filled metric name to its value.

Return type:

dict