# API reference

The public entry points below form the supported integration surface.
Underscore-prefixed helpers and internal DP cost tables may change between
releases. Use evaluator injection instead of replacing a planner's internal
engine or calling its private table builders.

Use `training_cost_fn` to customize only candidate training cost; use
`EvaluationProvider` to supply complete candidate evaluations. These options
are mutually exclusive on `RetrainingPlanner`. See
{ref}`cost accounting <custom-training-costs>` and
{ref}`historical state initialization <online-history>`.

```{eval-rst}
.. autoclass:: retrainopt.PlannerConfig
   :members:

.. autoclass:: retrainopt.RetrainingPlanner
   :members: plan

.. autoclass:: retrainopt.PlanResult
   :members: to_dict

.. autoclass:: retrainopt.OnlineState
   :members:

.. autoclass:: retrainopt.OnlineStepResult
   :members:

.. autoclass:: retrainopt.OnlineTrainingData
   :members:

.. autoclass:: retrainopt.OnlinePolicy
   :members:

.. autoclass:: retrainopt.BaseOnlinePolicy
   :members:

.. autoclass:: retrainopt.LearnedOnlinePolicy
   :members:

.. autoclass:: retrainopt.OnlineController
   :members:

.. autofunction:: retrainopt.collect_online_training_data

.. autoclass:: retrainopt.Decision

.. autoclass:: retrainopt.DecisionAction

.. autoclass:: retrainopt.CandidateEvaluation
   :members:

.. autoclass:: retrainopt.EvaluationProvider
   :members:

.. autoclass:: retrainopt.TrainingCostFunction
   :members: __call__

.. autoclass:: retrainopt.evaluation.EvaluationEngine
   :members: evaluate_initial_model, evaluate_candidate, evaluate_candidate_uncached, iter_cached_candidates

.. autoclass:: retrainopt.TimelineDataset
   :members:

.. autoclass:: retrainopt.ModelAdapter
   :members:

.. automodule:: retrainopt.adapters
   :members:
```