Skip to main content
def AgentsPredictorWorker.transform(self, data, agents=tuple(AGENTS)) [source]
Predict with one or more agents on a new dataset. Produces synchronized probability predictions for the requested agents on any data sample. Use this to:
  • Score a new OOT window.
  • Predict on a subset of agents (e.g., the top 2).
  • Generate predictions for a different segment.

​
Parameters

data : pDF
The dataset to be scored (must contain the features
required by the agents).
 agents : tuple
Subset of agents to engage. Defaults to all trained
agents.

​
Returns

AgentsPredictorResult
A unified result object containing synchronized
predictions for all requested agents.
No State Change: This method does not modify the fitted state. It produces a fresh prediction on the new data. The original fit results remain accessible via .fitted.