lmflow.pipeline.utils.rm_trainer
================================

.. py:module:: lmflow.pipeline.utils.rm_trainer


Classes
-------

.. autoapisummary::

   lmflow.pipeline.utils.rm_trainer.RewardTrainer
   lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer


Functions
---------

.. autoapisummary::

   lmflow.pipeline.utils.rm_trainer.compute_metrics
   lmflow.pipeline.utils.rm_trainer.rm_loss


Module Contents
---------------

.. py:function:: compute_metrics(eval_pred)

.. py:function:: rm_loss(model, inputs, return_outputs=False)

.. py:class:: RewardTrainer

   Bases: :py:obj:`transformers.Trainer`


   .. py:method:: compute_loss(model, inputs, return_outputs=False)


.. py:class:: PeftRewardTrainer

   Bases: :py:obj:`lmflow.pipeline.utils.peft_trainer.PeftTrainer`


   .. py:method:: compute_loss(model, inputs, return_outputs=False)