torchdrivesim.behavior.replay
=============================

.. py:module:: torchdrivesim.behavior.replay


Classes
-------

.. autoapisummary::

   torchdrivesim.behavior.replay.ReplayController


Functions
---------

.. autoapisummary::

   torchdrivesim.behavior.replay.interaction_replay


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

.. py:function:: interaction_replay(location, dataset_path, initial_frame=1, segment_length=40, recording=0)

.. py:class:: ReplayController(npc_size, npc_states, npc_present_masks: Optional[torch.Tensor] = None, time: int = 0, npc_types: Optional[torch.Tensor] = None, agent_type_names: Optional[List[str]] = None, spawn_controller: Optional[torchdrivesim.simulator.SpawnController] = None)

   Bases: :py:obj:`torchdrivesim.simulator.NPCController`


   Base class for non-playable character controllers. It leaves the state unchanged on each step.


   .. py:attribute:: time
      :value: 0



   .. py:attribute:: npc_states


   .. py:attribute:: npc_present_masks
      :value: None



   .. py:method:: advance_npcs(simulator: torchdrivesim.simulator.Simulator) -> None


   .. py:method:: to(device)


   .. py:method:: copy()


   .. py:method:: extend(n, in_place=True)


   .. py:method:: select_batch_elements(idx, in_place=True)