AutoMLRunSummary

Models & Types · AiDotNet.Models.Results

Represents a redacted (safe-to-share) summary of an AutoML run.

For Beginners

AutoML is an automatic process that tries multiple model attempts (“trials”) and keeps the best one. This class tells you how that search went, including:

  • How many trials ran
  • The best score found
  • A per-trial outcome history (without exposing secret settings)

How It Works

This summary is intended for facade outputs (for example, AiModelResult) and avoids exposing hyperparameters or sensitive model details. It provides transparency into the AutoML process while protecting proprietary implementation choices.

Constructors

Constructor Summary
AutoMLRunSummary()

Properties

Property Summary
AutoMLSearchStrategy? SearchStrategy { get; set; } Gets or sets the search strategy used for the AutoML run, if known.
DateTimeOffset SearchEndedUtc { get; set; } Gets or sets the UTC timestamp when the AutoML search ended.
DateTimeOffset SearchStartedUtc { get; set; } Gets or sets the UTC timestamp when the AutoML search started.
List<AutoMLTrialSummary> Trials { get; set; } Gets or sets a redacted list of trial summaries.
MetricType OptimizationMetric { get; set; } Gets or sets the optimization metric used to rank trials.
NASResultSummary NASResult { get; set; } Gets or sets NAS-specific result information when a NAS strategy was used.
TimeSpan TimeLimit { get; set; } Gets or sets the time limit used for the AutoML search.
bool MaximizeMetric { get; set; } Gets or sets a value indicating whether higher metric values are better.
bool UsedEnsemble { get; set; } Gets or sets a value indicating whether AutoML selected an ensemble as the final model.
double BestScore { get; set; } Gets or sets the best score achieved during the AutoML search.
int TrialLimit { get; set; } Gets or sets the maximum number of trials allowed for the AutoML search.
int? EnsembleSize { get; set; } Gets or sets the number of models in the selected ensemble, when applicable.