Skip to main content
def AgentsComparatorWorker.fit(self, inputs, fx_choice=REG.AIG, **params) [source]
Execute the full multi-agent comparison and scoring. Predicts with all agents, computes per-metric scores (performance, calibration, cross-audit), standardizes to a 0–1 scale, applies weights, and ranks agents by composite score.

​
Parameters

inputs : AgentsComparatorInput
Container for target evaluation data and configuration.
 fx_choice : str
The comparison strategy identifier. Defaults to REG.AIG.
 **params : dict
Runtime parameter overrides, passed to the internal fitter
(select_alpha, select_bins).

​
Returns

AgentsComparator
The fitted model instance containing the calculated performance matrices.
Agent Ranking: The ranked list is available via .stats.ordered — the first element is the recommended agent.