§
    4P„j™G  ã                   ó  — d dl Z d dlZd dlZd dlmZmZmZmZm	Z	m
Z
mZmZ d dlZddlmZmZmZ e j        dk    Z ej        d¦  «        Z G d„ d¦  «        Zd	ej        d
eeef         fd„Z	 d de	e	e
eef                           deded
eee	e         f         fd„Zdej        ded
efd„Z	 	 	 d!deeeej        ef         f         dedee         deeee                           deeef         dededee         d
eeeeef         f         fd„Z	 	 	 d"deeeej        ef         f         dedee         deeee                           deeef         dededee         d
eeeeef         f         fd„Z  G d„ d¦  «        Z!dS )#é    N)ÚSequenceÚDictÚOptionalÚTupleÚListÚUnionÚAnyÚMappingé   )ÚMetricÚScoreÚ	SignatureÚntÚ	sacrebleuc            
       óZ   — e Zd ZdZ	 	 d	dedee         dee         dee         fd„Zd„ ZdS )
ÚResulta=  A container to represent results from a particular statistical
    significance test.
    :param score: The floating point score for the system at hand.
    :param p_value: If exists, represents the p-value when the system at
    hand is compared to a baseline using a paired test.
    :param mean: When paired bootstrap test is applied, this represents
    the true mean score estimated from bootstrap resamples of the system.
    :param ci: When paired bootstrap test is applied, this represents
    the 95% confidence interval around the true mean score `sys_mean`.
    NÚscoreÚp_valueÚmeanÚcic                 ó>   — || _         || _        || _        || _        d S )N)r   r   r   r   )Úselfr   r   r   r   s        ú[/root/disk_3tb/conda_envs/Huy-visual/lib/python3.11/site-packages/sacrebleu/significance.pyÚ__init__zResult.__init__   s"   € àˆŒ
ØˆŒØˆŒ	ØˆŒˆˆó    c                 ón   — d                      d„ | j                             ¦   «         D ¦   «         ¦  «        S )Nú,c                 ó<   — g | ]\  }}|› d t          |¦  «        › �‘ŒS )ú=)Ústr)Ú.0ÚkÚvs      r   ú
<listcomp>z#Result.__repr__.<locals>.<listcomp>#   s,   € ÐKÐKÐK©T¨Q°˜A˜˜¥ A¡¤˜˜ÐKÐKÐKr   )ÚjoinÚ__dict__Úitems)r   s    r   Ú__repr__zResult.__repr__"   s1   € Ø�xŠxÐKÐK°T´]×5HÒ5HÑ5JÔ5JÐKÑKÔKÑLÔLÐLr   )NNN)Ú__name__Ú
__module__Ú__qualname__Ú__doc__Úfloatr   r   r(   © r   r   r   r      s   € € € € € ð	ð 	ð AEØEIðð ˜eð ¨h°u¬oð Ø ”ðØ3;¸E´?ðð ð ð ðMð Mð Mð Mð Mr   r   ÚscoresÚreturnc                 ó¾   — t          j        | ¦  «        } t          | ¦  «        }|dz  }||z
  dz
  }| |         | |         }}d||z
  z  }|                      ¦   «         |fS )z¿Takes a list of scores and returns mean and 95% confidence
    interval around the mean.

    :param scores: A list of floating point scores.
    :return: A tuple of mean and the 95% CI.
    é(   r   g      à?)ÚnpÚsortÚlenr   )r/   ÚnÚ	lower_idxÚ	upper_idxÚlowerÚupperr   s          r   Úestimate_cir;   &   si   € õ ŒW�V‰_Œ_€FÝˆF‰Œ€Að �R‘€IØ�I‘ Ñ!€IØ˜)Ô$ f¨YÔ&7ˆ5€EØ	�˜‘Ñ	€BØ�KŠK‰MŒM˜2ÐÐr   éè  ÚstatsÚmetricÚ	n_samplesc                 óÖ  ‡— t           j                             dd¦  «        }|                     ¦   «         dk    rdnt	          |¦  «        }t
          j                             |¦  «        }|                     t          | ¦  «        |t          | ¦  «        fd¬¦  «        }t          j
        | d¬¦  «        }ˆfd	„||         D ¦   «         }t          |¦  «                             ¦   «         |fS )
a¸  Performs bootstrap resampling for a single system to estimate
    a confidence interval around the true mean.
    :param stats: A list of statistics extracted from the system's hypotheses.
    :param metric: The `Metric` instance to be used for score computation.
    :n_samples: Number of bootstrap resamples to use.

    :return: A tuple of the seed choice as string and the list of `Score`
    instances for all bootstrap resamples.
    ÚSACREBLEU_SEEDÚ12345ÚnoneNT©ÚsizeÚreplaceÚfloat32©Údtypec                 ó`   •— g | ]*}‰                      |                     d ¦  «        ¦  «        ‘Œ+S ©r   )Ú_compute_score_from_statsÚsum©r!   Ú_sr>   s     €r   r$   z'_bootstrap_resample.<locals>.<listcomp>S   sD   ø€ ð Nð Nð NØ8:ˆ×(Ò(¨¯ª°©¬Ñ3Ô3ðNð Nð Nr   )ÚosÚenvironÚgetr9   Úintr3   ÚrandomÚdefault_rngÚchoicer5   Úarrayr    )	r=   r>   r?   ÚseedÚ_seedÚrngÚidxsÚstats_npr/   s	    `       r   Ú_bootstrap_resampler]   9   sá   ø€ õ Œ:�>Š>Ð*¨GÑ4Ô4€DØ—J’J‘L”L FÒ*Ð*ˆDˆDµ°D±	´	€EÝ
Œ)×
Ò
 Ñ
&Ô
&€Cð �:Š:•c˜%‘j”j¨	µ3°u±:´:Ð'>Èˆ:ÑMÔM€Dõ Œx˜ YÐ/Ñ/Ô/€HðNð Nð Nð NØ>FÀt¼nðNñ Nô N€Fõ ˆt‰9Œ9�?Š?ÑÔ˜fÐ$Ð$r   Úreal_differencec                 óŠ   — t          j        | |k    ¦  «                             ¦   «         }|dz   t          | ¦  «        dz   z  }|S )zÍComputes the p-value given the sample statistics and the real statistic.
    :param stats: A numpy array with the sample statistics.
    :real_difference: The real statistic.
    :return: The p-value.
    r   )r3   rM   Úitemr5   )r=   r^   ÚcÚps       r   Ú_compute_p_valuerc   Y   sC   € õ 	Œˆu�Ò&Ñ'Ô'×,Ò,Ñ.Ô.€Að 
ˆQ‰•3�u‘:”: ‘>Ñ"€Aà€Hr   é'  éÿÿÿÿÚbaseline_infoÚsys_nameÚ
hypothesesÚ
referencesÚmetricsÚn_ar_confidencerX   c           	      ó  ‡— t           j                             |¦  «        }|                     d|t	          |¦  «        ft
          ¬¦  «        }	|	 }
|dk    r4|                     t	          |¦  «        |t	          |¦  «        fd¬¦  «        }i }|                     ¦   «         D �]à\  }Š| |         \  }}t           	                    d|› d|›d�¦  «         ‰ 
                    ||¦  «        }‰                     |¦  «        }t          |j        |j        z
  ¦  «        }t           	                    d	|› d
�¦  «         |	|z  |
|z  z   }|
|z  |	|z  z   }t          j        ˆfd„|dd…df         D ¦   «         ¦  «        }t          j        ˆfd„|dd…df         D ¦   «         ¦  «        }t          t          j        t          j        |¦  «        t          j        |¦  «        z
  ¦  «        |¦  «        }t!          |j        |¦  «        }|dk    rvt           	                    d|› d
�¦  «         t          j        |d¬¦  «        }t          j        ˆfd„||         D ¦   «         ¦  «        }t#          |¦  «        \  |_        |_        |||<   �Œâ||fS )a>  Paired two-sided approximate randomization (AR) test for MT evaluation.

    :param baseline_info: A dictionary with `Metric` instances as the keys,
    that contains sufficient statistics and a `Result` instance for the baseline system.
    :param sys_name: The name of the system to be evaluated.
    :param hypotheses: A sequence of string hypotheses for the system.
    :param references: A sequence of reference documents with document being
    defined as a sequence of reference strings. If `None`, references
    will be used through each metric's internal cache.
    :param metrics: A dictionary of `Metric` instances that will be computed
    for each system.
    :param n_samples: The number of AR trials.
    :param n_ar_confidence: The number of bootstrap resamples to use for
    confidence estimation. A value of -1 disables confidence estimation.
    :param seed: The seed value for the RNG. If `None`, the RNG will not be
    fixed to a particular seed.

    :return: A tuple with first element being the system name and the second
    being a `Result` namedtuple.
    é   )rE   rI   r   TrD   ú
Computing ú for ú% and extracting sufficient statisticsz8 > Performing approximate randomization test (# trials: ú)c                 óD   •— g | ]}‰                      |¦  «        j        ‘ŒS r.   ©Ú_aggregate_and_computer   ©r!   Úxr>   s     €r   r$   z#_paired_ar_test.<locals>.<listcomp>¯   ó*   ø€ ÐMÐMÐM¸ˆV×*Ò*¨1Ñ-Ô-Ô3ÐMÐMÐMr   Nc                 óD   •— g | ]}‰                      |¦  «        j        ‘ŒS r.   rs   ru   s     €r   r$   z#_paired_ar_test.<locals>.<listcomp>±   rw   r   zI > Performing bootstrap resampling for confidence interval (# resamples: rG   rH   c                 ój   •— g | ]/}‰                      |                     d ¦  «        ¦  «        j        ‘Œ0S rK   ©rL   rM   r   rN   s     €r   r$   z#_paired_ar_test.<locals>.<listcomp>½   sA   ø€ ð #ð #ð #ØFH�×0Ò0°·²¸±´Ñ;Ô;ÔAð#ð #ð #r   )r3   rT   rU   Úintegersr5   ÚboolrV   r'   ÚsacreloggerÚinfoÚ_extract_corpus_statisticsrt   Úabsr   rW   rc   r   r;   r   r   )rf   rg   rh   ri   rj   r?   rk   rX   rZ   Úpos_selÚneg_selÚbs_idxsÚresultsÚnameÚbl_statsÚ	bl_resultÚ	sys_statsÚ	sys_scoreÚdiffÚshuf_aÚshuf_bÚscores_aÚscores_brb   ÚresÚ
sys_scoresr>   s                             @r   Ú_paired_ar_testr‘   p   sÑ  ø€ õ: Œ)×
Ò
 Ñ
%Ô
%€Cð �lŠl˜1 I­s°:©¬Ð#?ÅtˆlÑLÔL€Gð ˆh€Gà˜ÒÐà—*’*Ý�
‰OŒO ?µC¸
±O´OÐ"DÈdð ñ Tô Tˆð €GàŸš™œð 'ñ '‰ˆˆfà+¨DÔ1Ñˆ�)õ 	×ÒÐb dÐbÐb°ÐbÐbÐbÑcÔcÐcØ×5Ò5°jÀ*ÑMÔMˆ	Ø×1Ò1°)Ñ<Ô<ˆ	õ �9”? Y¤_Ñ4Ñ5Ô5ˆå×ÒÐ`ÐT]Ð`Ð`Ð`ÑaÔaÐaà˜8Ñ# g°	Ñ&9Ñ9ˆØ˜8Ñ# g°	Ñ&9Ñ9ˆõ ”8ØMÐMÐMÐM¸VÀAÀAÀAÀtÀG¼_ÐMÑMÔMñOô Oˆå”8ØMÐMÐMÐM¸VÀAÀAÀAÀtÀG¼_ÐMÑMÔMñOô Oˆõ ÝŒF•2”8˜HÑ%Ô%­¬°Ñ(:Ô(:Ñ:Ñ;Ô;¸TñCô Cˆõ �Y”_ aÑ(Ô(ˆà˜QÒÐÝ×ÒÐ{ÐixÐ{Ð{Ð{Ñ|Ô|Ð|Ýœ °)Ð<Ñ<Ô<ˆIåœð #ð #ð #ð #ØLUÐV]ÔL^ð#ñ #ô #ñ ô ˆJõ  +¨:Ñ6Ô6ÑˆCŒH�c”fð ˆ�‰‰à�WÐÐr   c                 óÔ  ‡— t           j                             |¦  «        }i }	|                     t	          |¦  «        |t	          |¦  «        fd¬¦  «        }
|                     ¦   «         D �]y\  }Š| |         \  }}t                               d|› d|›d�¦  «         ‰                     ||¦  «        }‰ 	                    |¦  «        }t          j
        |d¬¦  «        }t          j
        |d¬¦  «        }t          |j        |j        z
  ¦  «        }t                               d|› d	�¦  «         t          j
        ˆfd
„||
         D ¦   «         ¦  «        }t          j
        ˆfd„||
         D ¦   «         ¦  «        }t          |¦  «        \  }}t          j        ||z
  ¦  «        }||                     ¦   «         z
  }t          ||¦  «        }t!          |j        |||¦  «        |	|<   �Œ{||	fS )a   Paired bootstrap resampling test for MT evaluation. This function
    replicates the behavior of the Moses script called
    `bootstrap-hypothesis-difference-significance.pl`.

    :param baseline_info: A dictionary with `Metric` instances as the keys,
    that contains sufficient statistics and a `Result` instance for the baseline system.
    :param sys_name: The name of the system to be evaluated.
    :param hypotheses: A sequence of string hypotheses for the system.
    :param references: A sequence of reference documents with document being
    defined as a sequence of reference strings. If `None`, references
    will be used through each metric's internal cache.
    :param metrics: A dictionary of `Metric` instances that will be computed
    for each system.
    :param n_samples: The number of bootstrap resamples.
    :param n_ar_confidence: This parameter is not used for this function but
    is there for signature compatibility in the API.
    :param seed: The seed value for the RNG. If `None`, the RNG will not be
    fixed to a particular seed.

    :return: A tuple with first element being the system name and the second
    being a `Result` namedtuple.
    TrD   rn   ro   rp   rG   rH   z= > Performing paired bootstrap resampling test (# resamples: rq   c                 ój   •— g | ]/}‰                      |                     d ¦  «        ¦  «        j        ‘Œ0S rK   rz   rN   s     €r   r$   z#_paired_bs_test.<locals>.<listcomp>  s4   ø€ ÐXÐXÐXÀ2ˆV×-Ò-¨b¯fªf°Q©i¬iÑ8Ô8Ô>ÐXÐXÐXr   c                 ój   •— g | ]/}‰                      |                     d ¦  «        ¦  «        j        ‘Œ0S rK   rz   rN   s     €r   r$   z#_paired_bs_test.<locals>.<listcomp>  s4   ø€ ÐYÐYÐYÀ2ˆV×-Ò-¨b¯fªf°Q©i¬iÑ8Ô8Ô>ÐYÐYÐYr   )r3   rT   rU   rV   r5   r'   r}   r~   r   rt   rW   r€   r   r;   r   rc   r   )rf   rg   rh   ri   rj   r?   rk   rX   rZ   r„   r[   r…   r†   r‡   rˆ   r‰   rŠ   Ú	scores_blÚ
scores_sysÚsys_meanÚsys_ciÚsample_diffsr=   rb   r>   s                           @r   Ú_paired_bs_testrš   È   s  ø€ õ> Œ)×
Ò
 Ñ
%Ô
%€Cà€Gð �:Š:ÝˆJ‰Œ˜y­#¨j©/¬/Ð:ÀDð ñ Jô J€Dð  Ÿš™œð  Eñ  E‰ˆˆfà+¨DÔ1Ñˆ�)õ 	×ÒÐb dÐbÐb°ÐbÐbÐbÑcÔcÐcØ×5Ò5°jÀ*ÑMÔMˆ	Ø×1Ò1°)Ñ<Ô<ˆ	õ ”H˜Y¨iÐ8Ñ8Ô8ˆ	Ý”8˜H¨IÐ6Ñ6Ô6ˆõ �9”? Y¤_Ñ4Ñ5Ô5ˆå×ÒÐeÐYbÐeÐeÐeÑfÔfÐfÝ”HØXÐXÐXÐXÈÐRVÌÐXÑXÔXñZô Zˆ	å”XØYÐYÐYÐYÈÐSWÌÐYÑYÔYñ[ô [ˆ
õ ' zÑ2Ô2Ñˆ�&õ ”v˜j¨9Ñ4Ñ5Ô5ˆØ˜|×0Ò0Ñ2Ô2Ñ2ˆõ ˜U DÑ)Ô)ˆå˜yœ°°8¸VÑDÔDˆ�‰‰à�WÐÐr   c                   óþ   — e Zd ZdZdddœZ	 	 	 	 dd	eeeee         f                  d
e	ee
f         deeee                           dedededefd„Zdeeeef         eeeeeef                  f         f         fd„ZdS )Ú
PairedTesta
  This is the manager class that will call the actual standalone implementation
    for approximate randomization or paired bootstrap resampling, based on the
    `test_type` argument.

    :param named_systems: A lisf of (system_name, system_hypotheses) tuples on
    which the test will be applied.
    :param metrics: A dictionary of `Metric` instances that will be computed
    for each system.
    :param references: A sequence of reference documents with document being
    defined as a sequence of reference strings. If `None`, already cached references
    will be used through each metric's internal cache.
    :param test_type: `ar` for approximate randomization, `bs` for paired bootstrap.
    :param n_samples: The number of AR trials (for `ar`) or bootstrap resamples (for `bs`).
    The defaults (10000 or 1000 respectively) will be used if 0 is passed.
    :param n_ar_confidence: If `approximate randomization` is selected, the number
    of bootstrap resamples to use for confidence estimation. A value of -1 disables
    confidence estimation. 0 will use the default of 1000.
    :param n_jobs: If 0, a worker process will be spawned for each system variant.
    If > 0, the number of workers will be set accordingly. The default of 1
    does not use multi-processing.
    rd   r<   ©ÚarÚbsrž   r   re   r   Únamed_systemsrj   ri   Ú	test_typer?   rk   Ún_jobsc                 ó€  — |dv sJ d|›�¦   «         ‚|| _         | j         dk    rt          | _        n| j         dk    rt          | _        t          j                             dd¦  «        }|                     ¦   «         dk    rd nt          |¦  «        | _	        || _
        || _        || _        |dk    r|n| j        d         | _        |dk    r|n| j        | j                  | _        t!          |¦  «        d	z
  | _        t$          r"t&                               d
¦  «         d	| _
        nI| j
        dk    r>t+          j        ¦   «         dz  }	|	dk    rd	| _
        nt/          |	| j        ¦  «        | _
        i | _        i | _        i | _        | j        d         \  }
}|                     ¦   «         D �]–\  }}t&                               d|› d|
›�¦  «         |                     || j        ¦  «        }|                     |¦  «        }| j         dk    r| j        n| j        }d\  }}|dk    rBt?          |||¦  «        \  }}tA          tC          j"        d„ |D ¦   «         ¦  «        ¦  «        \  }}tG          |j$        ||¬¦  «        }||f| j        |j%        <   || j        |j%        <   | &                    ¦   «         }| '                    dtQ          | j	        ¦  «                             ¦   «         ¦  «         | '                    | j         | j        ¦  «         | j        dk    r| '                    d| j        ¦  «         || j        |j%        <   �Œ˜d S )Nr�   zUnknown test type rž   rŸ   rA   rB   rC   r   r   z,Parallel tests are not supported on Windows.rm   zPre-computing z statistics for )NNc                 ó   — g | ]	}|j         ‘Œ
S r.   )r   )r!   rv   s     r   r$   z'PairedTest.__init__.<locals>.<listcomp>v  s   € Ð6RÐ6RÐ6RÀ1°q´wÐ6RÐ6RÐ6Rr   )r   r   rX   ))r¡   r‘   Ú_fnrš   rP   rQ   rR   r9   rS   rY   r¢   ri   r    Ú_DEFAULT_SAMPLESrk   r?   r5   Ú	n_systemsÚ
IS_WINDOWSr}   ÚwarningÚmpÚ	cpu_countÚminÚ_signaturesÚ_baseline_inforj   r'   r~   r   rt   r]   r;   r3   rW   r   r   r…   Úget_signatureÚupdater    )r   r    rj   ri   r¡   r?   rk   r¢   rX   Ú
n_max_jobsÚbl_nameÚbl_hypsr…   r>   r†   Úbl_scoreÚconfidence_nÚbl_meanÚbl_ciÚ_Ú	bl_scoresÚresultÚsigs                          r   r   zPairedTest.__init__/  sN  € ð ˜LÐ(Ð(Ð(Ð*L¸yÐ*LÐ*LÑ(Ô(Ð(Ø"ˆŒð Œ>˜TÒ!Ð!Ý&ˆDŒHˆHØŒ^˜tÒ#Ð#Ý&ˆDŒHõ
 Œz�~Š~Ð.°Ñ8Ô8ˆØ!ŸZšZ™\œ\¨VÒ3Ð3�T�T½¸T¹¼ˆŒ
ØˆŒØ$ˆŒØ*ˆÔð 3BÀQÒ2FÐ2F˜˜ØÔ! $Ô'ð 	Ôð '0°!¢m m˜˜ØÔ! $¤.Ô1ð 	Œõ ˜]Ñ+Ô+¨aÑ/ˆŒõ ð 	>Ý×ÒÐ NÑOÔOÐOØˆDŒKˆKØŒ[˜AÒÐõ œ™œ¨1Ñ,ˆJØ˜QŠˆØ�”�õ " *¨d¬nÑ=Ô=�”à13ˆÔØ=?ˆÔð
 ˆŒàÔ-¨aÔ0Ñˆ�à#ŸMšM™OœOð 	2ñ 	2‰LˆD�&Ý×ÒÐO¨dÐOÐOÀGÐOÐOÑPÔPÐPØ×8Ò8¸À$Ä/ÑRÔRˆHØ×4Ò4°XÑ>Ô>ˆHð .2¬^¸tÒ-CÐ-C˜4œ>˜>ØÔ)ð ð (‰NˆG�UØ˜aÒÐÝ2°8¸VÀ\ÑRÔR‘��9Ý!,­R¬XÐ6RÐ6RÈ	Ð6RÑ6RÔ6RÑ-SÔ-SÑ!TÔ!T‘�˜å˜HœN°¸UÐCÑCÔCˆFà2:¸FÐ1CˆDÔ ¤Ñ.Ø*0ˆDŒL˜œÑ'ð ×&Ò&Ñ(Ô(ˆCØ�JŠJ�v�s 4¤:™œ×4Ò4Ñ6Ô6Ñ7Ô7Ð7ð �JŠJ�t”~ t¤~Ñ6Ô6Ð6ØÔ# aÒ'Ð'à—
’
˜4 Ô!5Ñ6Ô6Ð6Ø.1ˆDÔ˜Xœ]Ñ+Ñ+ð9	2ð 	2r   r0   c                 ó   ‡ ‡— g }i }d„ ‰ j         D ¦   «         |d<   ‰ j                             ¦   «         D ]\  }\  }}|g||<   Œt          ‰ j         dd…         ¦  «        D ]N\  }\  }}‰ j        r‰ j        nd}	|                     ‰ j        ||‰ j        ‰ j        ‰ j        ‰ j	        |	f¦  «         ŒO‰ j
        dk    rˆ fd„|D ¦   «         }
n‚t                               d‰ j
        › d�¦  «         t          j        d¦  «                             ‰ j
        ¦  «        5 Šˆˆ fd	„|D ¦   «         }d
„ |D ¦   «         }
ddd¦  «         n# 1 swxY w Y   |
D ]:\  }}|                     ¦   «         D ] \  }}||                              |¦  «         Œ!Œ;‰ j        |fS )zCRuns the paired test either on single or multiple worker processes.c                 ó   — g | ]
}|d          ‘ŒS rK   r.   )r!   Únss     r   r$   z'PairedTest.__call__.<locals>.<listcomp>Ž  s   € Ð?Ð?Ð? b˜B˜qœEÐ?Ð?Ð?r   ÚSystemr   Nc                 ó$   •— g | ]} ‰j         |Ž ‘ŒS r.   )r¥   )r!   Úargsr   s     €r   r$   z'PairedTest.__call__.<locals>.<listcomp>Ÿ  s!   ø€ Ð9Ð9Ð9¨4�x�t”x �Ð9Ð9Ð9r   z
Launching z parallel workers.Úforkc                 óF   •— g | ]}‰                      ‰j        |¦  «        ‘ŒS r.   )Úapply_asyncr¥   )r!   rÁ   Úpoolr   s     €€r   r$   z'PairedTest.__call__.<locals>.<listcomp>©  s+   ø€ ÐKÐKÐK¸T˜×(Ò(¨¬°4Ñ8Ô8ÐKÐKÐKr   c                 ó6   — g | ]}|                      ¦   «         ‘ŒS r.   )rR   )r!   Újs     r   r$   z'PairedTest.__call__.<locals>.<listcomp>¬  s    € Ð1Ð1Ð1 q˜1Ÿ5š5™7œ7Ð1Ð1Ð1r   )r    r®   r'   Ú	enumeraterY   Úappendri   rj   r?   rk   r¢   r}   r~   rª   Úget_contextÚPoolr­   )r   Útasksr/   r>   r¸   rº   Úidxr…   ÚhypsrX   r„   Újobsrg   Úsys_resultsÚ_resultrÅ   s   `              @r   Ú__call__zPairedTest.__call__ˆ  s=  øø€ àˆØ68ˆð @Ð?¨DÔ,>Ð?Ñ?Ô?ˆˆxÑð $(Ô#6×#<Ò#<Ñ#>Ô#>ð 	&ð 	&ÑˆF‘K�Q˜Ø$˜XˆF�6‰NˆNõ "+¨4Ô+=¸a¸b¸bÔ+AÑ!BÔ!Bð 	Lð 	LÑˆC‘�$˜Ø!%¤Ð5�4”:�:°ˆDà�LŠLØÔ$ d¨D°$´/Ø”˜tœ~¨tÔ/CÀTðKñLô Lð Lð Lð
 Œ;˜!ÒÐØ9Ð9Ð9Ð9°5Ð9Ñ9Ô9ˆGˆGõ ×ÒÐI¨$¬+ÐIÐIÐIÑJÔJÐJÝ” Ñ'Ô'×,Ò,¨T¬[Ñ9Ô9ð 2¸TØKÐKÐKÐKÐKÀUÐKÑKÔK�ð 2Ð1¨DÐ1Ñ1Ô1�ð	2ð 2ð 2ñ 2ô 2ð 2ð 2ð 2ð 2ð 2ð 2øøøð 2ð 2ð 2ð 2ð &-ð 	/ð 	/Ñ!ˆH�kØ#.×#4Ò#4Ñ#6Ô#6ð /ð /‘�˜Ø�v”×%Ò% gÑ.Ô.Ð.Ð.ð/ð Ô Ð'Ð's   ÄD?Ä?EÅEN)rž   r   re   r   )r)   r*   r+   r,   r¦   r   r   r    r   r
   r   r   rS   r   r   r   r   r   rÒ   r.   r   r   rœ   rœ     s  € € € € € ðð ð, Øðð Ðð #'Ø"#Ø(*Ø ðW2ð W2 d¨5°°h¸s´mÐ1CÔ+DÔ&Eð W2Ø! # v +Ô.ðW2à% h¨x¸¬}Ô&=Ô>ðW2ð  ðW2ð  ð	W2ð
 #&ðW2ð ðW2ð W2ð W2ð W2ðr+(˜%  S¨) ^Ô 4°d¸3ÀÀUÈ3ÐPVÈ;ÔEWÔ@XÐ;XÔ6YÐ YÔZð +(ð +(ð +(ð +(ð +(ð +(r   rœ   )r<   )rd   re   N)r<   re   N)"rP   ÚloggingÚmultiprocessingrª   Útypingr   r   r   r   r   r   r	   r
   Únumpyr3   Úmetrics.baser   r   r   r…   r¨   Ú	getLoggerr}   r   Úndarrayr-   r;   rS   r    r]   rc   r‘   rš   rœ   r.   r   r   ú<module>rÚ      sh  ðØ 	€	€	€	Ø €€€Ø Ð Ð Ð Ø MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ MÐ Mà Ð Ð Ð à 2Ð 2Ð 2Ð 2Ð 2Ð 2Ð 2Ð 2Ð 2Ð 2àŒW˜Š_€
ð  ˆgÔ Ñ,Ô,€ðMð Mð Mð Mð Mñ Mô Mð Mð,˜œ
ð  u¨U°E¨\Ô':ð ð ð ð ð( :>ð%ð %˜t D¨¨s°E¨zÔ):Ô$;Ô<ð %Ø &ð%Ø36ð%ØBGÈÈTÐRWÌ[ÐHXÔBYð%ð %ð %ð %ð@˜BœJð ¸ð À5ð ð ð ð ð8 &+Ø+-Ø*.ðUð U 4¨¨U°2´:¸vÐ3EÔ-FÐ(FÔ#Gð UØ!ðUà (¨¤ðUð !)¨°(¸3´-Ô)@Ô AðUð " # v +Ô.ð	Uð
  #ðUð &)ðUð # 3œ-ðUð 49¸¸dÀ3ÈÀ;Ô>OÐ9OÔ3PðUð Uð Uð Uðz &*Ø+-Ø*.ðIð I 4¨¨U°2´:¸vÐ3EÔ-FÐ(FÔ#Gð IØ!ðIà (¨¤ðIð !)¨°(¸3´-Ô)@Ô AðIð " # v +Ô.ð	Ið
  #ðIð &)ðIð # 3œ-ðIð 49¸¸dÀ3ÈÀ;Ô>OÐ9OÔ3PðIð Ið Ið IðX_(ð _(ð _(ð _(ð _(ñ _(ô _(ð _(ð _(ð _(r   