h SrSSKJr SSKrSSKJr SSKrSSKJ r J r SSK J r SSK JrJrJrJr S/r"S S\5rg) z?Wrapper that tracks the cumulative rewards and episode lengths.) annotationsN)deque)ActTypeObsType)warn) ArrayType AutoresetMode VectorEnv VectorWrapperRecordEpisodeStatisticscr^\rSrSrSrSSU4SjjjrS S U4SjjjrS SjrSrU=r $) r a`This wrapper will keep track of cumulative rewards and episode lengths. At the end of any episode within the vectorized env, the statistics of the episode will be added to ``info`` using the key ``episode``, and the ``_episode`` key is used to indicate the environment index which has a terminated or truncated episode. >>> infos = { # doctest: +SKIP ... ... ... "episode": { ... "r": "", ... "l": "", ... "t": "" ... }, ... "_episode": "" ... } Moreover, the most recent rewards and episode lengths are stored in buffers that can be accessed via :attr:`wrapped_env.return_queue` and :attr:`wrapped_env.length_queue` respectively. Attributes: return_queue: The cumulative rewards of the last ``deque_size``-many episodes length_queue: The lengths of the last ``deque_size``-many episodes Example: >>> from pprint import pprint >>> import gymnasium as gym >>> envs = gym.make_vec("CartPole-v1", num_envs=3) >>> envs = RecordEpisodeStatistics(envs) >>> obs, info = envs.reset(123) >>> _ = envs.action_space.seed(123) >>> end = False >>> while not end: ... obs, rew, term, trunc, info = envs.step(envs.action_space.sample()) ... end = term.any() or trunc.any() ... >>> envs.close() >>> pprint(info) # doctest: +SKIP {'_episode': array([ True, False, False]), '_final_info': array([ True, False, False]), '_final_observation': array([ True, False, False]), 'episode': {'l': array([11, 0, 0], dtype=int32), 'r': array([11., 0., 0.], dtype=float32), 't': array([0.007812, 0. , 0. ], dtype=float32)}, 'final_info': array([{}, None, None], dtype=object), 'final_observation': array([array([ 0.11448676, 0.9416149 , -0.20946532, -1.7619033 ], dtype=float32), None, None], dtype=object)} c>[TU]U5 X0lSURR;a$[ US35 [ RUlOL[URRS[ 5(deURRSUlSUl [R"UR45Ul[R"UR45Ul[R"UR4[ S9Ul[R"UR4[$S9Ul[)US9Ul[)US9Ul[)US9Ulg)aThis wrapper will keep track of cumulative rewards and episode lengths. Args: env (Env): The environment to apply the wrapper buffer_length: The size of the buffers :attr:`return_queue`, :attr:`length_queue` and :attr:`time_queue` stats_key: The info key to save the data autoreset_modez is missing `autoreset_mode` tag in its metadata, therefore, `RecordEpisodeStatistics` is assuming that the environment uses `AutoresetMode.NEXT_STEP`. See `https://farama.org/Vector-Autoreset-Mode` for more information on autoreset modes.rdtype)maxlenN)super__init__ _stats_keyenvmetadatarr NEXT_STEP_autoreset_mode isinstance episode_countnpzerosnum_envsepisode_start_timesepisode_returnsintepisode_lengthsbool prev_donesr time_queue return_queue length_queue)selfr buffer_length stats_key __class__s Z/home/james-whalen/.local/lib/python3.13/site-packages/gymnasium/wrappers/vector/common.pyr RecordEpisodeStatistics.__init__Hs # 488#4#4 4 &HI $1#:#:D dhh//0@A=QQ QQ#'88#4#45E#FD /1xx8H/I +-88T]]4D+E+-88T]]4DC+P&(hh /?t&L}5!7!7c>[TU]XS9up4UGb/SU;Ga(URS5n[U[R 5(dS[ U535eURUR4:Xd!SURSUR35eUR[R:XdSUR35e[R"U5(d SU35e[R"5URU'SURU'SUR U'S UR"U'X44$[R$"UR[R"55Ul[R&"UR5Ul[R&"UR[(S 9Ul[R&"UR[*S 9UlX44$) zOResets the environment using kwargs and resets the episode returns and lengths.)seedoptions reset_maskz9`options['reset_mask': mask]` must be a numpy array, got z0`options['reset_mask': mask]` must have shape `(z ,)`, got z>`options['reset_mask': mask]` must have `dtype=np.bool_`, got zK`options['reset_mask': mask]` must contain a boolean array, got reset_mask=rFr)rresetpoprrndarraytypeshaperrbool_anytime perf_counterr r!r#r%fullrr"r$)r)r1r2obsinfor3r,s r-r4RecordEpisodeStatistics.resetks GMtM=   <7#: \2JBJJ ^J4PZK[J\] ^## ( mA$--PYZdZjZjYkl m  BHH, cOPZP`P`Oab c,66 j\]g\hi j483D3D3FD $ $Z 0/0D  ,/0D  ,*/DOOJ 'y (*wwt}}d>O>O>Q'RD $#%88DMM#:D #%88DMM#ED  hht}}DADOyr/c URRU5unnnnn[U[5(dS[ U5S35eSUR UR 'UR [R"UR 5==U[R"UR 5- ss'SURUR 'URUR )==S- ss'[R"5URUR '[R"XE5=Uln[R"U5nU(GaURU;dSUR3U;a2[!SURS[#UR%5535e[R&"[R"5UR- S5n [R("XpR S 5[R("XpRS5[R("XyS 5S .X`R'XvSUR3'U=R*U- sl[R("U5Hpn UR,R/X5 UR0R/UR U 5 UR2R/URU 5 Mr UUUUU4$) z@Steps through the environment, recording the episode statistics.zW`vector.RecordEpisodeStatistics` requires `info` type to be `dict`, its actual type is z@. This may be due to usage of other wrappers in the wrong order.r_z)Attempted to add episode stats with key 'z'' but this key already exists in info: g)rlt)rsteprdictr7r!r%r logical_notr#r;r<r logical_orsumr ValueErrorlistkeysroundwhererr&extendr'r() r)actions observationsrewards terminations truncationsinfosdones num_donesepisode_time_lengthis r-rHRecordEpisodeStatistics.steps HHMM' "       4   s deijoepdqrr s s 12T__- R^^DOO<= NN4?? +B  =12T__- doo-.!3.484E4E4G  1"$-- "JJ%FF5M %'Qt.?+@E+I ??PPwx|~C~H~H~JyKxLM')hh%%'$*B*BBA'#%)=)=sC%)=)=qA%cB*oo& 05$//*+,   ) + XXe_&&':'=>!!(()=)=a)@A!!(()=)=a)@A%        r/) rrrr#r!r r(r%r'r&)depisode)rr r*r"r+str)NN)r1zint | list[int] | Noner2z dict | None)rSrreturnz5tuple[ObsType, ArrayType, ArrayType, ArrayType, dict]) __name__ __module__ __qualname____firstlineno____doc__rr4rH__static_attributes__ __classcell__)r,s@r-r r sw.f!" !8 !8!8 !8!8J(,#!$!!!F: : >: : r/)rf __future__rr; collectionsrnumpyrgymnasium.corerrgymnasium.loggerrgymnasium.vector.vector_envrr r r __all__r r/r-rqs<E" +! % %q mq r/