§
    >²¸j* ã                  óx  — U d dl mZ d dlZd dlZd dlZd dlmZ d dlmZ d dlm	Z	 d dl
mZ d dlmZmZmZmZ d dlZd dlmZ d dlmZmZ d d	lmZ d d
lmZmZmZ d dlmZ d dl m!Z!m"Z" d dl#m$Z$ d dl%m&Z&  e&e'¦  «        Z(erd dl)Z* G d„ ded¬¦  «        Z+e,e-e.e+         f         Z/de0d<    G d„ de¦  «        Z1e	 G d„ d¦  «        ¦   «         Z2 G d„ d¦  «        Z3dZ4dZ5dZ6 ej7        ej8        ¦  «        j9        Z:d.d%„Z; G d&„ d'e¦  «        Z< G d(„ d)¦  «        Z= G d*„ d+¦  «        Z> G d,„ d-e!e2         ¦  «        Z?dS )/é    )ÚannotationsN)Údefaultdict)Údeepcopy)Ú	dataclass)ÚEnum)ÚTYPE_CHECKINGÚAnyÚ	TypeAliasÚ	TypedDict)ÚAREA_DATA_FIELDÚORIENTED_BOX_COORDINATES)Ú
Detections)Úbox_iou_batch_with_jaccardÚmask_iou_batchÚoriented_box_iou_batch)ÚLEGACY_COLOR_PALETTE)ÚMetricÚMetricTarget)Úensure_pandas_installed)Ú_get_loggerc                  ó¦   — e Zd ZU ded<   ded<   ded<   ded<   ded<   ded	<   ded
<   ded<   ded<   ded<   ded<   ded<   ded<   ded<   ded<   dS )Ú_TypeCocoDictÚintÚidÚimage_idÚcategory_idúlist[float]ÚbboxÚfloatÚareaÚiscrowdÚignoreÚ_ignoreÚscorezlist[list[float]]ÚsegmentationÚstrÚnameÚsupercategoryÚcaptionÚ	keypointsznpt.NDArray[Any]ÚcontentN©Ú__name__Ú
__module__Ú__qualname__Ú__annotations__© ó    úz/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/supervision/metrics/mean_average_precision.pyr   r   !   s­   € € € € € € Ø€G€G�GØ€M€M�MØÐÐÑØÐÐÑØ€K€K�KØ€L€L�LØ€K€K�KØ€L€L�LØ€L€L�LØ#Ð#Ð#Ñ#Ø€I€I�IØÐÐÑØ€L€L�LØÐÐÑð
 ÐÐÑÐÐr2   r   F)Útotalr
   Ú_TypeCocoDatasetc                  ó~   — e Zd ZU ded<   ded<   ded<   ded<   ded<   ded	<   d
ed<   d
ed<   ded<   d
ed<   ded<   dS )Ú_TypeEvaluationImageResultr   r   r   ú!list[float] | tuple[float, float]Ú
area_rangeÚmax_detú	list[int]Údt_idsÚgt_idsznpt.NDArray[np.int64]Ú	dtMatchesÚ	gtMatchesr   ÚdtScoresÚgtIgnoreznpt.NDArray[np.bool_]ÚdtIgnoreNr,   r1   r2   r3   r7   r7   :   s�   € € € € € € Ø€M€M�MØÐÐÑØ1Ð1Ð1Ñ1Ø€L€L�LØÐÐÑØÐÐÑØ$Ð$Ð$Ñ$Ø$Ð$Ð$Ñ$ØÐÐÑØ#Ð#Ð#Ñ#Ø#Ð#Ð#Ñ#Ð#Ð#r2   r7   c                  óÚ   — e Zd ZU dZded<   ded<   edd„¦   «         Zedd	„¦   «         Zedd
„¦   «         Zded<   ded<   ded<   ded<   dZ	ded<   dZ
ded<   dZded<   dd„Zdd„Zdd„ZdS ) ÚMeanAveragePrecisionResultu.  
    The result of the Mean Average Precision calculation.

    Returns `-1` sentinel scores when no detections or targets are present.

    Attributes:
        metric_target: the type of data used for the metric -
            boxes, masks or oriented bounding boxes.
        is_class_agnostic: When computing class-agnostic results, class ID
            is set to `-1`.
        mAP_scores: the mAP scores at each IoU threshold.
            Shape: `(num_iou_thresholds,)`
        ap_per_class: the average precision scores per
            class and IoU threshold. Shape: `(num_target_classes, num_iou_thresholds)`
        iou_thresholds: the IoU thresholds used in the calculations.
        matched_classes: the class IDs of all matched classes.
            Corresponds to the rows of `ap_per_class`.
        small_objects: the mAP results
            for small objects (area < 32Â²).
        medium_objects: the mAP results
            for medium objects (32Â² â‰¤ area < 96Â²).
        large_objects: the mAP results
            for large objects (area â‰¥ 96Â²).
    r   Úmetric_targetÚboolÚis_class_agnosticÚreturnr   c                óš   — | j         | j         dk             }t          |¦  «        dk    r!t          |                     ¦   «         ¦  «        S dS )z5the mAP score at IoU thresholds from `0.5` to `0.95`.éÿÿÿÿr   )Ú
mAP_scoresÚlenr   Úmean)ÚselfÚvalid_scoress     r3   Úmap50_95z#MeanAveragePrecisionResult.map50_95f   sI   € ð ” t¤¸Ò';Ô<ˆÝˆ|ÑÔ˜qÒ Ð Ý˜×*Ò*Ñ,Ô,Ñ-Ô-Ð-à�2r2   c                ó6   — t          | j        d         ¦  «        S )z(the mAP score at IoU threshold of `0.5`.r   ©r   rK   ©rN   s    r3   Úmap50z MeanAveragePrecisionResult.map50o   ó   € õ �T”_ QÔ'Ñ(Ô(Ð(r2   c                ó6   — t          | j        d         ¦  «        S )z)the mAP score at IoU threshold of `0.75`.é   rR   rS   s    r3   Úmap75z MeanAveragePrecisionResult.map75t   rU   r2   únpt.NDArray[np.float64]rK   Úap_per_classÚiou_thresholdsznpt.NDArray[np.int32]Úmatched_classesNz!MeanAveragePrecisionResult | NoneÚsmall_objectsÚmedium_objectsÚlarge_objectsr&   c                óô   — | j         �| j        �| j        €d| j        d›d| j        d›d| j        d›�S d| j        d›d| j        d›d| j        d›d| j         j        d›d| j        j        d›d| j        j        d›�S )	aI  
        Formats the evaluation output metrics to match the structure used by pycocotools

        Example:
           ```pycon
           >>> import numpy as np
           >>> import supervision as sv
           >>> from supervision.metrics import MeanAveragePrecision
           >>> predictions = sv.Detections(
           ...     xyxy=np.array([[0, 0, 10, 10]]),
           ...     class_id=np.array([0]),
           ...     confidence=np.array([0.9])
           ... )
           >>> targets = sv.Detections(
           ...     xyxy=np.array([[0, 0, 10, 10]]),
           ...     class_id=np.array([0])
           ... )
           >>> map_metric = MeanAveragePrecision()
           >>> map_result = map_metric.update(predictions, targets).compute()
           >>> print(map_result)  # doctest: +ELLIPSIS
           Average Precision (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = ...
           Average Precision (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = ...
           Average Precision (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = ...
           Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = ...
           Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = ...
           Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = ...

           ```
        NzHAverage Precision (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = z.3fzI
Average Precision (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = zI
Average Precision (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = zI
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = zI
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = zI
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = )r]   r^   r_   rP   rT   rX   rS   s    r3   Ú__str__z"MeanAveragePrecisionResult.__str__�   sü   € ð> ÔÐ&ØÔ"Ð*ØÔ!Ð)ð4Ø#'¤=Ð8ð4ð 4ð $(¤:Ð5ð4ð 4ð
 $(¤:Ð3ð4ð 4ððAØ#œ}Ð4ðAð Að  $œzÐ1ðAð Að
  $œzÐ1ðAð Að  $Ô1Ô:ÐBðAð Að  $Ô2Ô;ÐCðAð Að  $Ô1Ô:Ð@ðAð Að	
r2   úpd.DataFramec                ó  — t          ¦   «          ddl}| j        | j        | j        dœ}| j        �;| j                             ¦   «         }|                     ¦   «         D ]\  }}||d|› �<   Œ| j        �;| j                             ¦   «         }|                     ¦   «         D ]\  }}||d|› �<   Œ| j	        �;| j	                             ¦   «         }|                     ¦   «         D ]\  }}||d|› �<   Œ |j
        |dg¬¦  «        S )zt
        Convert the result to a pandas DataFrame.

        Returns:
            The result as a DataFrame.
        r   N©z	mAP@50:95zmAP@50zmAP@75Úsmall_objects_Úmedium_objects_Úlarge_objects_)Úindex)r   ÚpandasrP   rT   rX   r]   Ú	to_pandasÚitemsr^   r_   Ú	DataFrame)rN   ÚpdÚpandas_dataÚsmall_objects_dfÚkeyÚvalueÚmedium_objects_dfÚlarge_objects_dfs           r3   rj   z$MeanAveragePrecisionResult.to_pandas¼   sY  € õ 	 Ñ!Ô!Ð!ØÐÐÐð œØ”jØ”jð*
ð *
ˆð ÔÐ)Ø#Ô1×;Ò;Ñ=Ô=ÐØ.×4Ò4Ñ6Ô6ð <ð <‘
��UØ6;�Ð2¨SÐ2Ð2Ñ3Ð3ØÔÐ*Ø $Ô 3× =Ò =Ñ ?Ô ?ÐØ/×5Ò5Ñ7Ô7ð =ð =‘
��UØ7<�Ð3¨cÐ3Ð3Ñ4Ð4ØÔÐ)Ø#Ô1×;Ò;Ñ=Ô=ÐØ.×4Ò4Ñ6Ô6ð <ð <‘
��UØ6;�Ð2¨SÐ2Ð2Ñ3Ð3ð ˆrŒ|ØØ�#ð
ñ 
ô 
ð 	
r2   ÚNonec                óÎ  — ddl m} g d¢}| j        | j        | j        g}t
          d         gdz  }| j        �A|g d¢z  }|| j        j        | j        j        | j        j        gz  }|t
          d         gdz  z  }| j        �A|g d¢z  }|| j        j        | j        j        | j        j        gz  }|t
          d         gdz  z  }| j        �A|g d	¢z  }|| j        j        | j        j        | j        j        gz  }|t
          d
         gdz  z  }d|j	        d<   | 
                    d¬¦  «        \  }}|                     dd¦  «         |                     dd¬¦  «         |                     dd¬¦  «         t          t          |¦  «        ¦  «        }|                     |||d¬¦  «        }|                     |¦  «         |                     |dd¬¦  «         |D ]_}	|	                     ¦   «         }
|                     |	                     ¦   «         |	                     ¦   «         dz  z   |
dz   |
d›dd¬¦  «         Œ`d|j	        d<   |                     ¦   «          |                     ¦   «          dS )z¼
        Plot the mAP results.

        ![example_plot](
            https://media.roboflow.com/supervision-docs/metrics/mAP_plot_example.png
        ){ align=center width="800" }
        r   )Úpyplotrd   é   N)zSmall: mAP@50:95zSmall: mAP@50zSmall: mAP@75)zMedium: mAP@50:95zMedium: mAP@50zMedium: mAP@75é   )zLarge: mAP@50:95zLarge: mAP@50zLarge: mAP@75é   Ú	monospacezfont.family)é
   é   )Úfigsizeé   ÚValueÚbold)Ú
fontweightzMean Average PrecisionÚcenter)ÚcolorÚaligné-   Úright)ÚrotationÚhag{®Gáz”?z.2fÚbottom)rˆ   Úvaz
sans-serif)Ú
matplotlibrv   rP   rT   rX   r   r]   r^   r_   ÚrcParamsÚsubplotsÚset_ylimÚ
set_ylabelÚ	set_titleÚrangerL   ÚbarÚ
set_xticksÚset_xticklabelsÚ
get_heightÚtextÚget_xÚ	get_widthÚtight_layoutÚshow)rN   ÚpltÚlabelsÚvaluesÚcolorsÚ_ÚaxÚx_positionsÚbarsr’   Úy_values              r3   ÚplotzMeanAveragePrecisionResult.plotß   s¾  € ð 	-Ð,Ð,Ð,Ð,Ð,à2Ð2Ð2ˆØ”- ¤¨T¬ZÐ8ˆÝ& qÔ)Ð*¨QÑ.ˆàÔÐ)ØÐLÐLÐLÑLˆFØØÔ"Ô+ØÔ"Ô(ØÔ"Ô(ðñ ˆFð
 Õ+¨AÔ.Ð/°!Ñ3Ñ3ˆFàÔÐ*ØÐOÐOÐOÑOˆFØØÔ#Ô,ØÔ#Ô)ØÔ#Ô)ðñ ˆFð
 Õ+¨AÔ.Ð/°!Ñ3Ñ3ˆFàÔÐ)ØÐLÐLÐLÑLˆFØØÔ"Ô+ØÔ"Ô(ØÔ"Ô(ðñ ˆFð
 Õ+¨AÔ.Ð/°!Ñ3Ñ3ˆFà&1ˆŒ�]Ñ#à—’ W�Ñ-Ô-‰ˆˆ2Ø
�Š�A�qÑÔÐØ
�Š�g¨&ˆÑ1Ô1Ð1Ø
�ŠÐ-¸&ˆÑAÔAÐAå�C ™KœKÑ(Ô(ˆØ�vŠv�k 6°¸xˆvÑHÔHˆà
�Š�kÑ"Ô"Ð"Ø
×Ò˜6¨B°7ÐÑ;Ô;Ð;àð 	ð 	ˆCØ—n’nÑ&Ô&ˆGØ�GŠGØ—	’	‘”˜cŸmšm™oœo°Ñ1Ñ1Ø˜$‘ØÐ Ð ØØð ñ ô ð ð ð '3ˆŒ�]Ñ#à×ÒÑÔÐØ�Š‰
Œ
ˆ
ˆ
ˆ
r2   )rH   r   )rH   r&   )rH   rb   ©rH   rt   )r-   r.   r/   Ú__doc__r0   ÚpropertyrP   rT   rX   r]   r^   r_   ra   rj   r¤   r1   r2   r3   rD   rD   H   sA  € € € € € € ðð ð2  ÐÐÑØÐÐÑàðð ð ñ „Xðð ð)ð )ð )ñ „Xð)ð ð)ð )ð )ñ „Xð)ð (Ð'Ð'Ñ'Ø)Ð)Ð)Ñ)Ø+Ð+Ð+Ñ+Ø*Ð*Ð*Ñ*Ø7;€MÐ;Ð;Ð;Ñ;Ø8<€NÐ<Ð<Ð<Ñ<Ø7;€MÐ;Ð;Ð;Ñ;ð9
ð 9
ð 9
ð 9
ðv!
ð !
ð !
ð !
ðFCð Cð Cð Cð Cð Cr2   rD   c                  ó~   — e Zd ZdZddd„Zed d„¦   «         Zd!d	„Z	 	 	 	 d"d#d„Z	 	 	 d$d%d„Z		 	 d&d'd„Z
dd(d„Zd)d„ZdS )*ÚEvaluationDatasetzh
    Class used representing a dataset in the right format needed by the
    `COCOEvaluator` class.
    NÚtargetsú_TypeCocoDataset | NonerH   rt   c                óà   — i | _         i | _        i | _        i | _        t	          t
          ¦  «        | _        t	          t
          ¦  «        | _        |€dS || _         |                      ¦   «          dS )zê
        Constructor of EvaluationDataset object used to evaluate models with
        Mean Average Precision.

        Args:
            targets: The targets (ground truth) of the dataset in a the
                COCO format.
        N)	ÚdatasetÚannsÚcatsÚimgsr   ÚlistÚimg_to_annsÚcat_to_imgsÚcreate_class_members)rN   rª   s     r3   Ú__init__zEvaluationDataset.__init__+  sk   € ð *,ˆŒØ.0ˆŒ	Ø.0ˆŒ	Ø.0ˆŒ	Ý;FÅtÑ;LÔ;LˆÔÝ1<½TÑ1BÔ1BˆÔàˆ?ØˆFð ˆŒØ×!Ò!Ñ#Ô#Ð#Ð#Ð#r2   c                ó   —  | d ¬¦  «        S )N©rª   r1   )Úclss    r3   ÚemptyzEvaluationDataset.emptyD  s   € àˆs˜4Ð Ñ Ô Ð r2   c                óT  — i }i }i }t          t          ¦  «        }t          t          ¦  «        }d| j        v r<| j        d         D ].}||d                                       |¦  «         |||d         <   Œ/d| j        v r| j        d         D ]}|||d         <   Œd| j        v r| j        d         D ]}|||d         <   Œd| j        v r@d| j        v r7| j        d         D ])}||d                                       |d         ¦  «         Œ*|| _        || _        || _        || _        || _        dS )z8
        Create index elements for the dataset.
        r   r   r   ÚimagesÚ
categoriesr   N)	r   r±   r­   Úappendr®   r²   r³   r°   r¯   )	rN   r®   r¯   r°   r²   r³   ÚannÚimgÚcats	            r3   r´   z&EvaluationDataset.create_class_membersH  sa  € ð *,ˆØ)+ˆØ)+ˆÝ6AÅ$Ñ6GÔ6GˆÝ,7½Ñ,=Ô,=ˆØ˜DœLÐ(Ð(Ø”| MÔ2ð &ð &�Ø˜C 
œOÔ,×3Ò3°CÑ8Ô8Ð8Ø"%��S˜”Y‘�à�t”|Ð#Ð#Ø”| HÔ-ð &ð &�Ø"%��S˜”Y‘�à˜4œ<Ð'Ð'Ø”| LÔ1ð &ð &�Ø"%��S˜”Y‘�à˜DœLÐ(Ð(¨\¸T¼\Ð-IÐ-IØ”| MÔ2ð Hð H�Ø˜C Ô.Ô/×6Ò6°s¸:´ÑGÔGÐGÐGð ˆŒ	Ø&ˆÔØ&ˆÔØˆŒ	ØˆŒ	ˆ	ˆ	r2   FÚimg_idsúlist[int] | NoneÚcat_idsr9   útuple[float, float] | Noner!   rF   r;   c                óJ  ‡ ‡‡— |s‰s‰s‰ j         d         }nn|r;ˆ fd„|D ¦   «         }t          t          j                             |¦  «        ¦  «        }n‰ j         d         }‰s|nˆfd„|D ¦   «         }‰s|nˆfd„|D ¦   «         }|du rd„ |D ¦   «         }nd„ |D ¦   «         }|S )a¥  
        Get annotation ids that satisfy given filter conditions.

        Args:
            img_ids: ids of the images that we want to retrieve.
            cat_ids: ids of the categories that we want to retrieve.
            area_range: area range of the annotations that we want to retrieve
                in the format [min_area, max_area].
            iscrowd: if annotations to retrieve are `iscrowded=1`.
        r   c                ó<   •— g | ]}|‰j         v ¯‰j         |         ‘ŒS r1   )r²   )Ú.0Úimg_idrN   s     €r3   ú
<listcomp>z8EvaluationDataset.get_annotation_ids.<locals>.<listcomp>  s9   ø€ ð ð ð àØ Ô!1Ð1Ð1ð Ô$ VÔ,à1Ð1Ð1r2   c                ó(   •— g | ]}|d          ‰v ¯|‘ŒS ©r   r1   )rÇ   r¾   rÃ   s     €r3   rÉ   z8EvaluationDataset.get_annotation_ids.<locals>.<listcomp>Œ  s(   ø€ ÐKÐKÐK˜c¨S°Ô-?À7Ð-JÐ-J�cÐ-JÐ-JÐ-Jr2   c                óX   •— g | ]&}‰d          |d         cxk     r‰d         k     ¯!n n|‘Œ'S )r   r    r~   r1   )rÇ   r¾   r9   s     €r3   rÉ   z8EvaluationDataset.get_annotation_ids.<locals>.<listcomp>“  sN   ø€ ð ð ð Ø¨:°a¬=¸3¸v¼;Ð+VÐ+VÒ+VÐ+VÈÐTUÌÒ+VÐ+VÐ+VÐ+VÐ+V�CÐ+VÐ+VÐ+Vr2   Tc                ó6   — g | ]}|d          dk    ¯|d         ‘ŒS )r!   r~   r   r1   ©rÇ   r¾   s     r3   rÉ   z8EvaluationDataset.get_annotation_ids.<locals>.<listcomp>š  s*   € ÐDÐDÐD °°I´À!Ò0CÐ0C�3�t”9Ð0CÐ0CÐ0Cr2   c                ó   — g | ]
}|d          ‘ŒS ©r   r1   rÎ   s     r3   rÉ   z8EvaluationDataset.get_annotation_ids.<locals>.<listcomp>œ  s   € Ð-Ð-Ð- �3�t”9Ð-Ð-Ð-r2   )r­   r±   Ú	itertoolsÚchainÚfrom_iterable)rN   rÁ   rÃ   r9   r!   r®   ÚlistsÚidss   ` ``    r3   Úget_annotation_idsz$EvaluationDataset.get_annotation_idsi  s$  øøø€ ð$ ð 	˜wð 	¨zð 	Ø”< Ô.ˆDˆDàð 3ðð ð ð à")ðñ ô �õ
 �IœO×9Ò9¸%Ñ@Ô@ÑAÔA��à”| MÔ2�ð
 ðL��àKÐKÐKÐK TÐKÑKÔKð ð "ð��ðð ð ð Ø#'ðñ ô ð ð �dˆ?ˆ?ØDÐD¨ÐDÑDÔDˆCˆCà-Ð-¨Ð-Ñ-Ô-ˆCØˆ
r2   Ú	cat_namesúlist[str] | NoneÚsupercategory_namesc                óÒ   ‡‡‡— ‰s‰s‰s| j         d         }nC| j         d         }‰s|nˆfd„|D ¦   «         }‰s|nˆfd„|D ¦   «         }‰s|nˆfd„|D ¦   «         }d„ |D ¦   «         }|S )aX  
        Get category ids that satisfy given filter conditions.

        Args:
            cat_names: names of the categories to retrieve.
            supercategory_names: names of the supercategories to retrieve.
            cat_ids: ids of the categories to retrieve.

        Returns:
            ids: integer array of category ids.
        r¼   c                ó(   •— g | ]}|d          ‰v ¯|‘ŒS )r'   r1   )rÇ   rÀ   r×   s     €r3   rÉ   z6EvaluationDataset.get_category_ids.<locals>.<listcomp>º  s'   ø€ ÐFÐFÐF˜c¨S°¬[¸IÐ-EÐ-E�cÐ-EÐ-EÐ-Er2   c                ó(   •— g | ]}|d          ‰v ¯|‘ŒS )r(   r1   )rÇ   rÀ   rÙ   s     €r3   rÉ   z6EvaluationDataset.get_category_ids.<locals>.<listcomp>Á  s0   ø€ ð ð ð Ø¨3¨Ô+?ÐCVÐ+VÐ+V�CÐ+VÐ+VÐ+Vr2   c                ó(   •— g | ]}|d          ‰v ¯|‘ŒS rÐ   r1   )rÇ   rÀ   rÃ   s     €r3   rÉ   z6EvaluationDataset.get_category_ids.<locals>.<listcomp>È  s'   ø€ Ð)VÐ)VÐ)V°#ÀÀTÄÈgÐAUÐAU¨#ÐAUÐAUÐAUr2   c                ó   — g | ]
}|d          ‘ŒS rÐ   r1   )rÇ   rÀ   s     r3   rÉ   z6EvaluationDataset.get_category_ids.<locals>.<listcomp>Ê  s   € Ð)Ð)Ð)˜Sˆs�4ŒyÐ)Ð)Ð)r2   )r­   )rN   r×   rÙ   rÃ   r¯   rÕ   s    ```  r3   Úget_category_idsz"EvaluationDataset.get_category_idsŸ  sÞ   øøø€ ð$ ð 	Ð!4ð 	¸Wð 	Ø”< Ô-ˆDˆDà”< Ô-ˆDð
 !ðG��àFÐFÐFÐF TÐFÑFÔFð ð +ð��ðð ð ð Ø#'ðñ ô ð ð $ÐV��Ð)VÐ)VÐ)VÐ)V¸Ð)VÑ)VÔ)Vð ð *Ð) DÐ)Ñ)Ô)ˆØˆ
r2   c                ód  — |s*|s(| j                              ¦   «         }t          |¦  «        S |rt          |¦  «        nt          ¦   «         }|rUt	          |¦  «        D ]E\  }}|dk    r|st          | j        |         ¦  «        }Œ(|t          | j        |         ¦  «        z  }ŒFt          |¦  «        S )zÿ
        Get image ids that satisfy given filter conditions.

        Args:
            img_ids: ids of the images to retrieve.
            cat_ids: ids of the categories to retrieve.

        Returns:
            ids: integer array of image ids.
        r   )r°   Úkeysr±   ÚsetÚ	enumerater³   )rN   rÁ   rÃ   rÕ   Úids_setÚiÚcat_ids          r3   Úget_image_idszEvaluationDataset.get_image_idsÍ  s·   € ð  ð 	˜wð 	Ø”)—.’.Ñ"Ô"ˆCÝ˜‘9”9Ðà")Ð4•#�g‘,”,�,­s©u¬uˆàð 	=Ý& wÑ/Ô/ð =ð =‘	��6Ø˜’6�6 '�6Ý! $Ô"2°6Ô":Ñ;Ô;�G�Gà�s 4Ô#3°FÔ#;Ñ<Ô<Ñ<�G�Gå�G‰}Œ}Ðr2   rÕ   úlist[_TypeCocoDict]c                ó(   ‡ — |€g S ˆ fd„|D ¦   «         S )zµ
        Get annotations with the specified ids.

        Args:
            ids: integer ids specifying annotations.

        Returns:
            anns: loaded annotations.
        Nc                ó*   •— g | ]}‰j         |         ‘ŒS r1   )r®   )rÇ   ÚidxrN   s     €r3   rÉ   z5EvaluationDataset.get_annotations.<locals>.<listcomp>ø  s   ø€ Ð.Ð.Ð. 3�”	˜#”Ð.Ð.Ð.r2   r1   )rN   rÕ   s   ` r3   Úget_annotationsz!EvaluationDataset.get_annotationsì  s)   ø€ ð ˆ;ØˆIØ.Ð.Ð.Ð.¨#Ð.Ñ.Ô.Ð.r2   Úpredictionsc           	     ó°  — t                                ¦   «         }t          | j        d         ¦  «        |j        d<   t	          |t          ¦  «        st          d¦  «        ‚t          |¦  «        dk    rg |j        d<   |S d„ |D ¦   «         }t          |¦  «        t          |                      ¦   «         ¦  «        k    st          d¦  «        ‚d|d         v rt          d¦  «        ‚d	|d         v rt          d
¦  «        ‚d|d         v rt          d¦  «        ‚d|d         v r—|d         d         g k    s…t          j        | j        d         ¦  «        |j        d<   t          |¦  «        D ]N\  }}|d         \  }}}}	|||z   |||	z   g\  }
}}}d	|vr|
||
|||||gg|d	<   d|vr||	z  |d<   |dz   |d<   d|d<   ŒO||j        d<   |                     ¦   «          |S )zÚ
        Load prediction result into an EvaluationDataset object.

        Args:
            predictions: prediction result.

        Returns:
            EvaluationDataset object representing the predictions.
        r»   zresults must be a listr   r   c                ó   — g | ]
}|d          ‘ŒS )r   r1   )rÇ   Úpreds     r3   rÉ   z6EvaluationDataset.load_predictions.<locals>.<listcomp>  s   € Ð8Ð8Ð8 Dˆt�JÔÐ8Ð8Ð8r2   z-Results do not correspond to current coco setr)   z5Evaluating predictions with caption is not supported.r%   z:Evaluating predictions with segmentation is not supported.r*   z7Evaluating predictions with keypoints is not supported.r   r¼   r    r~   r   r!   )r©   r¹   r±   r­   Ú
isinstanceÚ
ValueErrorrL   râ   rç   ÚNotImplementedErrorÚcopyr   rã   r´   )rN   rí   Úpredictions_datasetrÕ   rë   rð   ÚxÚyÚwÚhÚx1Úx2Úy1Úy2s                 r3   Úload_predictionsz"EvaluationDataset.load_predictionsú  sM  € õ 0×5Ò5Ñ7Ô7ÐÝ04°T´\À(Ô5KÑ0LÔ0LÐÔ# HÑ-å˜+¥tÑ,Ô,ð 	7ÝÐ5Ñ6Ô6Ð6õ ˆ{ÑÔ˜qÒ Ð Ø9;ÐÔ'¨Ñ6Ø&Ð&à8Ð8¨KÐ8Ñ8Ô8ˆõ
 �3‰xŒx�3˜t×1Ò1Ñ3Ô3Ñ4Ô4Ò4Ð4ÝÐLÑMÔMÐMð ˜ AœÐ&Ð&Ý%ØGñô ð ð ˜{¨1œ~Ð-Ð-Ý%ØLñô ð ð ˜K¨œNÐ*Ð*Ý%ØIñô ð ð �{ 1”~Ð%Ð%¨k¸!¬n¸VÔ.DÈÒ.JÐ.JÝ8<¼Ø”˜\Ô*ñ9ô 9ÐÔ'¨Ñ5õ
 ' {Ñ3Ô3ð $ð $‘	��TØ! &œ\‘
��1�a˜Ø"# Q¨¡U¨A¨q°1©uÐ!5‘��B˜˜Bð "¨Ð-Ð-Ø-/°°R¸¸RÀÀRÈÐ,LÐ+M�D˜Ñ(à Ð%Ð%Ø#$ q¡5�D˜‘LØ  1™W��T‘
à"#��Y‘�Ø5@ÐÔ# MÑ2Ø×0Ò0Ñ2Ô2Ð2Ø"Ð"r2   ©N)rª   r«   rH   rt   )rH   r©   r¥   )NNNF)
rÁ   rÂ   rÃ   rÂ   r9   rÄ   r!   rF   rH   r;   )NNN)r×   rØ   rÙ   rØ   rÃ   rÂ   rH   r;   )NN)rÁ   rÂ   rÃ   rÂ   rH   r;   )rÕ   rÂ   rH   rè   )rí   rè   rH   r©   )r-   r.   r/   r¦   rµ   Úclassmethodr¹   r´   rÖ   rß   rç   rì   rþ   r1   r2   r3   r©   r©   %  s  € € € € € ðð ð
$ð $ð $ð $ð $ð2 ð!ð !ð !ñ „[ð!ðð ð ð ðF %)Ø$(Ø15Øð4ð 4ð 4ð 4ð 4ðp '+Ø04Ø$(ð	,ð ,ð ,ð ,ð ,ð` %)Ø$(ðð ð ð ð ð>/ð /ð /ð /ð /ðA#ð A#ð A#ð A#ð A#ð A#r2   r©   i   i $  g    _ BÚ
masks_trueúlist[npt.NDArray[np.bool_]]Úmasks_detectionÚis_crowdú
list[bool]rH   rY   c                óŠ  — t          |¦  «        dk    st          | ¦  «        dk    r<t          j        t          |¦  «        t          | ¦  «        ft          j        ¬¦  «        S t          j        | ¦  «                             t          ¦  «        }t          j        |¦  «                             t          ¦  «        }t          j        |t          ¬¦  «        }t          ||¦  «        j	                             t          j        ¦  «        }t          j
        |¦  «        s|S t          j        d¦  «        }t          j        |¦  «        d         }||                              t          |¦  «        d¦  «                             t          j        ¦  «        }	|                     |j        d         d¦  «                             t          j        ¦  «        }
|
|	j	        z                       t          j        ¦  «        }|
                     d¬¦  «                             t          j        ¦  «        }||dd…df         |z   z  |dd…|f<   |S )ak  
    Calculate the IoU between detection masks (dt) and ground-truth masks (gt),
    following the COCO convention: a detection may match any subregion of a
    crowd ground truth, so for crowd rows the union collapses to the detection
    area (mask counterpart of `box_iou_batch_with_jaccard`).

    Args:
        masks_true: List of ground-truth masks of shape `(H, W)`.
        masks_detection: List of detection masks of shape `(H, W)`.
        is_crowd: List indicating if each ground-truth mask is a crowd region.

    Returns:
        Array of IoU values of shape `(len(masks_detection), len(masks_true))`.
    r   ©Údtyper~   rJ   ©ÚaxisN)rL   Únpr¹   Úfloat64ÚstackÚastyperF   Úasarrayr   ÚTÚanyÚspacingÚwhereÚreshapeÚfloat32ÚshapeÚsum)r  r  r  Úgt_masksÚdt_masksÚcrowdÚiouÚepsÚ	crowd_idxÚgt_flatÚdt_flatÚ
area_interÚarea_dts                r3   Ú_mask_iou_with_jaccardr"  G  sÉ  € õ& ˆ?ÑÔ˜qÒ Ð ¥C¨
¡O¤O°qÒ$8Ð$8ÝŒx�˜_Ñ-Ô-­s°:©¬Ð?ÅrÄzÐRÑRÔRÐRåŒx˜
Ñ#Ô#×*Ò*­4Ñ0Ô0€HÝŒx˜Ñ(Ô(×/Ò/µÑ5Ô5€HÝŒJ�x¥tÐ,Ñ,Ô,€Eõ $2°(¸HÑ#EÔ#EÔ#G×#NÒ#NÝ
Œ
ñ$ô $€Cõ Œ6�%‰=Œ=ð Øˆ
õ
 Œ*�Q‰-Œ-€CÝ”˜‘” Ô"€IØ�yÔ!×)Ò)­#¨i©.¬.¸"Ñ=Ô=×DÒDÅRÄZÑPÔP€GØ×Ò˜xœ~¨aÔ0°"Ñ5Ô5×<Ò<½R¼ZÑHÔH€GØ˜GœIÑ%×-Ò-­b¬jÑ9Ô9€JØ�kŠk˜qˆkÑ!Ô!×(Ò(­¬Ñ4Ô4€GØ" g¨a¨a¨a°¨gÔ&6¸Ñ&<Ñ=€Cˆˆˆˆ9ˆÑØ€Jr2   c                  ó"   — e Zd ZdZdZdZdZdZdS )Ú
ObjectSizez
    Enum for object size.
    ÚallÚsmallÚmediumÚlargeN)r-   r.   r/   r¦   ÚALLÚSMALLÚMEDIUMÚLARGEr1   r2   r3   r$  r$  w  s.   € € € € € ðð ð €CØ€EØ€FØ€E€E€Er2   r$  c                  ó   — e Zd ZdZdd„ZdS )ÚCOCOEvaluatorParametersz&
    Parameters for COCOEvaluator
    rH   rt   c           	     ó¾  — g | _         g | _        t          j        ddt	          t          j        d¦  «        ¦  «        dz   dt          j        ¬¦  «        | _        t          j        ddt	          t          j        d	¦  «        ¦  «        dz   dt          j        ¬¦  «        | _        g d
¢| _	        dt          gdt          gt          t          gt          t          gg| _        dS )z(Initialize all parameters for evaluationç      à?gffffffî?gÿÿÿÿÿÿ!@r~   T)Úendpointr  ç        g      ð?g      Y@)r~   r{   éd   r   N)rÁ   rÃ   r  Úlinspacer   Úroundr  Úiou_thrsÚrec_thrsÚmax_detsÚMAX_ALL_OBJECT_AREAÚSMALL_OBJECT_AREAÚMEDIUM_OBJECT_AREAr9   rS   s    r3   rµ   z COCOEvaluatorParameters.__init__‡  sØ   € ð #%ˆŒØ"$ˆŒõ œØØÝ•”Ð,Ñ-Ô-Ñ.Ô.°Ñ2ØÝ”*ð
ñ 
ô 
ˆŒõ œØØÝ•”Ð,Ñ-Ô-Ñ.Ô.°Ñ2ØÝ”*ð
ñ 
ô 
ˆŒð %˜˜ˆŒð Õ#Ð$ØÕ!Ð"ÝÕ 2Ð3ÝÕ!4Ð5ð	.
ˆŒˆˆr2   Nr¥   )r-   r.   r/   r¦   rµ   r1   r2   r3   r.  r.  ‚  s2   € € € € € ðð ð
ð 
ð 
ð 
ð 
ð 
r2   r.  c                  óX   — e Zd ZdZej        fdd	„Zdd
„Zdd„Zdd„Z	dd„Z
dd„Zdd„ZdS )ÚCOCOEvaluatorz2
    Evaluator class to compute COCO metrics.
    Úcoco_targetsr©   Úcoco_predictionsrE   r   rH   rt   c                óö  — |€t          d¦  «        ‚|€t          d¦  «        ‚|| _        || _        || _        g | _        i | _        t          t          ¦  «        | _        t          t          ¦  «        | _	        t          ¦   «         | _        g | _        i | _        t          | j                             ¦   «         ¦  «        | j        _        t          | j                             ¦   «         ¦  «        | j        _        dS )aL  
        Constructor of COCOEvaluator object.

        Args:
            coco_targets: The dataset with the ground truths.
            coco_predictions: The dataset with the predictions.
            metric_target: The type of detection data used to compute the IoU -
                boxes, masks or oriented bounding boxes.
        Nzcoco_targets must be providedz!coco_predictions must be provided)rò   r>  r?  rE   Ú	eval_imgsÚresultsr   r±   Ú_targetsÚ_predictionsr.  ÚparamsÚstatsÚiousÚsortedrç   rÁ   rß   rÃ   )rN   r>  r?  rE   s       r3   rµ   zCOCOEvaluator.__init__­  sè   € ð ÐÝÐ<Ñ=Ô=Ð=ØÐ#ÝÐ@ÑAÔAÐAà(ˆÔØ 0ˆÔØ*ˆÔð CEˆŒà*,ˆŒåKVÝñL
ô L
ˆŒõ �ÑÔð 	Ôõ .Ñ/Ô/ˆŒà#%ˆŒ
àDFˆŒ	å$ TÔ%6×%DÒ%DÑ%FÔ%FÑGÔGˆŒÔÝ$ TÔ%6×%GÒ%GÑ%IÔ%IÑJÔJˆŒÔÐÐr2   c                óR  — | j                              | j        j        | j        j        ¬¦  «        }| j                              |¦  «        }| j                             | j        j        | j        j        ¬¦  «        }| j                             |¦  «        }|D ]i}t          |                     dd¦  «        ¦  «        }t          |                     dd¦  «        ¦  «        }t          t          |p|¦  «        ¦  «        |d<   Œjt          t          ¦  «        | _        |D ]0}| j        |d         |d         f                              |¦  «         Œ1t          t          ¦  «        | _        |D ]0}| j        |d         |d         f                              |¦  «         Œ1g | _        i | _        dS )zA
        Prepare targets and predictions for evaluation.
        )rÁ   rÃ   r"   r   r!   r   r   N)r>  rÖ   rE  rÁ   rÃ   rì   r?  r   ÚgetrF   r   r±   rC  r½   rD  rA  rB  )	rN   Úannotation_idsrª   Úprediction_idsrí   Úgtr"   r!   Údts	            r3   Ú _prepare_targets_and_predictionsz.COCOEvaluator._prepare_targets_and_predictionsÛ  s¦  € ð
 Ô*×=Ò=Ø”KÔ'°´Ô1Dð >ñ 
ô 
ˆð Ô#×3Ò3°NÑCÔCˆàÔ.×AÒAØ”KÔ'°´Ô1Dð Bñ 
ô 
ˆð Ô+×;Ò;¸NÑKÔKˆð ð 	8ð 	8ˆBÝ˜Ÿš ¨!Ñ,Ô,Ñ-Ô-ˆFÝ˜"Ÿ&š& ¨AÑ.Ô.Ñ/Ô/ˆGÝ�t FÐ$5¨gÑ6Ô6Ñ7Ô7ˆBˆx‰LˆLõ $¥DÑ)Ô)ˆŒØð 	Hð 	HˆBØŒM˜"˜Zœ.¨"¨]Ô*;Ð;Ô<×CÒCÀBÑGÔGÐGÐGõ (­Ñ-Ô-ˆÔØð 	Lð 	LˆBØÔ˜b œn¨b°Ô.?Ð?Ô@×GÒGÈÑKÔKÐKÐKð ˆŒØˆŒˆˆr2   rÈ   r   ræ   únpt.NDArray[np.float32]c                ó”  ‡— | j         ||f         }| j        ||f         Št          |¦  «        dk    r5t          ‰¦  «        dk    r"t          j        g t          j        ¬¦  «        }|S t          j        d„ ‰D ¦   «         d¬¦  «        }ˆfd„|D ¦   «         Št          ‰¦  «        | j        j        d         k    r‰d| j        j        d         …         Šd„ |D ¦   «         }| j	        t          j        k    r't          d	„ |D ¦   «         d
„ ‰D ¦   «         |¦  «        }�n| j	        t          j        k    rÒt          |¦  «        dk    st          ‰¦  «        dk    r=t          j        t          ‰¦  «        t          |¦  «        ft          j        ¬¦  «        }n˜t          j        d„ |D ¦   «         ¦  «        }t          j        d„ ‰D ¦   «         ¦  «        }	t#          ||	¦  «        j                             t          j        ¦  «        }n)d„ |D ¦   «         }
d„ ‰D ¦   «         }t)          |
||¦  «        }|                     t          j        ¦  «        S )ao  
        Compute the IoU between the targets and predictions for a given image and
        category, using boxes, masks or oriented bounding boxes depending on the
        configured metric target.

        Args:
            img_id: The image id.
            cat_id: The category id.

        Returns:
            The IoU between the targets and predictions.
        r   r  c                ó    — g | ]}|d           ‘ŒS ©r$   r1   ©rÇ   Úds     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>  s   € Ð3Ð3Ð3¨1˜A˜gœJ˜;Ð3Ð3Ð3r2   Ústable©Úkindc                ó    •— g | ]
}‰|         ‘ŒS r1   r1   ©rÇ   rå   rN  s     €r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>  s   ø€ Ð"Ð"Ð"˜ˆb�ŒeÐ"Ð"Ð"r2   rJ   c                ó8   — g | ]}t          |d          ¦  «        ‘ŒS )r!   )rF   )rÇ   Úos     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>  s$   € Ð3Ð3Ð3¨1•D˜˜9œÑ&Ô&Ð3Ð3Ð3r2   c                ó   — g | ]
}|d          ‘ŒS ©r+   r1   ©rÇ   Úgs     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>#  s   € Ð*Ð*Ð* !��9”Ð*Ð*Ð*r2   c                ó   — g | ]
}|d          ‘ŒS r^  r1   rT  s     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>#  s   € Ð,FÐ,FÐ,F¸a¨Q¨y¬\Ð,FÐ,FÐ,Fr2   c                ó   — g | ]
}|d          ‘ŒS r^  r1   r_  s     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>+  ó   € Ð"<Ð"<Ð"<°A 1 Y¤<Ð"<Ð"<Ð"<r2   c                ó   — g | ]
}|d          ‘ŒS r^  r1   rT  s     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>,  rc  r2   c                ó   — g | ]
}|d          ‘ŒS ©r   r1   r_  s     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>1  ó   € Ð.Ð.Ð. a˜˜&œ	Ð.Ð.Ð.r2   c                ó   — g | ]
}|d          ‘ŒS rf  r1   rT  s     r3   rÉ   z.COCOEvaluator._compute_iou.<locals>.<listcomp>2  rg  r2   )rC  rD  rL   r  Úarrayr  ÚargsortrE  r8  rE   r   ÚMASKSr"  ÚORIENTED_BOUNDING_BOXESr¹   r  r  r   r  r  r   )rN   rÈ   ræ   rM  Úempty_resultÚindsr  r  Úgt_obbÚdt_obbÚgt_boxesÚdt_boxesrN  s               @r3   Ú_compute_iouzCOCOEvaluator._compute_iouþ  s:  ø€ ð Œ]˜6 6˜>Ô*ˆØÔ˜v v˜~Ô.ˆõ ˆr‰7Œ7�aŠ<ˆ<�C ™GœG qšL˜LÝ46´H¸RÅrÄzÐ4RÑ4RÔ4RˆLØÐõ ŒzÐ3Ð3°Ð3Ñ3Ô3¸(ÐCÑCÔCˆØ"Ð"Ð"Ð"˜TÐ"Ñ"Ô"ˆõ ˆr‰7Œ7�T”[Ô)¨"Ô-Ò-Ð-Ø�A˜œÔ,¨RÔ0Ð0Ô1ˆBð 4Ð3°Ð3Ñ3Ô3ˆð Ô¥Ô!3Ò3Ð3Ý(Ø*Ð* rÐ*Ñ*Ô*Ð,FÐ,FÀ2Ð,FÑ,FÔ,FÈñô ˆC‰Cð Ô¥<Ô#GÒGÐGõ �2‰wŒw˜!Š|ˆ|�s 2™wœw¨!š|˜|Ý”h¥ B¡¤­¨R©¬Ð1½¼ÐDÑDÔD��åœÐ"<Ð"<¸Ð"<Ñ"<Ô"<Ñ=Ô=�ÝœÐ"<Ð"<¸Ð"<Ñ"<Ô"<Ñ=Ô=�õ -¨V°VÑ<Ô<Ô>×EÒEÅbÄjÑQÔQ��à.Ð.¨2Ð.Ñ.Ô.ˆHØ.Ð.¨2Ð.Ñ.Ô.ˆHÝ,¨X°xÀÑJÔJˆCØ�zŠz�"œ*Ñ%Ô%Ð%r2   r9   r8   r:   ú!_TypeEvaluationImageResult | Nonec                ó^  ‡‡‡‡— | j         ||f         Š| j        ||f         Št          ‰¦  «        dk    rt          ‰¦  «        dk    rdS |\  ŠŠ‰D ]*}|d         s‰|d         cxk    r‰k    sn d|d<   Œ%d|d<   Œ+t          j        d„ ‰D ¦   «         d¬	¦  «        }ˆfd
„|D ¦   «         Št          j        d„ ‰D ¦   «         d¬	¦  «        }ˆfd„|d|…         D ¦   «         Št          | j        ||f         ¦  «        dk    r| j        ||f         dd…|f         n| j        ||f         }t          | j        j        ¦  «        }	t          ‰¦  «        }
t          ‰¦  «        }t          j        |	|
ft          j	        ¬¦  «        }t          j        |	|ft          j	        ¬¦  «        }t          j
        d„ ‰D ¦   «         t          j	        ¬¦  «        }t          j        |	|ft          j        ¬¦  «        }t          |¦  «        dk    rýt          | j        j        ¦  «        D ]ã\  }}t          ‰¦  «        D ]Î\  }}t          |dg¦  «        }d}t          ‰¦  «        D ]s\  }}t          |                     dd¦  «        ¦  «        }|||f         dk    r|sŒ9|dk    r||         dk    r||         dk    r n|||f         |k    r|||f         }|}Œt|dk    r-||         |||f<   ‰|         d         |||f<   |d         |||f<   ŒÏŒät          j
        ˆˆfd„‰D ¦   «         ¦  «                             dt          ‰¦  «        f¦  «        }t          j        |t          j        |dk    t          j        ||	d¦  «        ¦  «        ¦  «        }||||d„ ‰D ¦   «         d„ ‰D ¦   «         ||d„ ‰D ¦   «         ||dœS )a-  
        Perform evaluation for single category and image.
        Args:
            img_id: The image id.
            cat_id: The category id.
            area_range: The area range.
            max_det: The maximum number of detections.

        Returns:
            The evaluation results.
        r   Nr"   r    r~   r#   c                ó   — g | ]
}|d          ‘ŒS ©r#   r1   r_  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>[  s   € Ð9Ð9Ð9°  )¤Ð9Ð9Ð9r2   rV  rW  c                ó    •— g | ]
}‰|         ‘ŒS r1   r1   )rÇ   rå   rM  s     €r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>\  s   ø€ Ð'Ð'Ð'˜ˆb�ŒeÐ'Ð'Ð'r2   c                ó    — g | ]}|d           ‘ŒS rS  r1   rT  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>_  s   € Ð8Ð8Ð8°  7¤ Ð8Ð8Ð8r2   c                ó    •— g | ]
}‰|         ‘ŒS r1   r1   rZ  s     €r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>`  s   ø€ Ð2Ð2Ð2˜ˆb�ŒeÐ2Ð2Ð2r2   r  c                ó   — g | ]
}|d          ‘ŒS rw  r1   r_  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>r  s   € Ð7Ð7Ð7¨q˜a 	œlÐ7Ð7Ð7r2   g�Aòÿÿÿï?rJ   r!   r   c                ó@   •— g | ]}|d          ‰k     p|d          ‰k    ‘ŒS )r    r1   )rÇ   rU  Úmax_areaÚmin_areas     €€r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>š  s0   ø€ ÐFÐFÐF¸aˆQˆvŒY˜Ò!Ð9 Q v¤Y°Ò%9ÐFÐFÐFr2   c                ó   — g | ]
}|d          ‘ŒS rÐ   r1   rT  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>ª  ó   € Ð+Ð+Ð+ 1�q˜”wÐ+Ð+Ð+r2   c                ó   — g | ]
}|d          ‘ŒS rÐ   r1   r_  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>«  r€  r2   c                ó   — g | ]
}|d          ‘ŒS rS  r1   rT  s     r3   rÉ   z1COCOEvaluator._evaluate_image.<locals>.<listcomp>®  s   € Ð0Ð0Ð0¨˜˜7œÐ0Ð0Ð0r2   )r   r   r9   r:   r<   r=   r>   r?   r@   rA   rB   )rC  rD  rL   r  rj  rG  rE  r6  ÚzerosÚint64ri  Úbool_rã   Úminr   rJ  r  Ú
logical_orÚlogical_andÚrepeat)rN   rÈ   ræ   r9   r:   r`  Ú	gt_sortedÚ	dt_sortedrG  Únum_thresholdsÚnum_ground_truthsÚnum_detectionsÚ
gt_matchesÚ
dt_matchesÚ	gt_ignoreÚ	dt_ignoreÚ	tresh_idxÚthreshÚdet_idxÚdetÚiou_best_matchÚbest_match_idxÚg_idxr!   Úarea_range_maskrN  rM  r}  r~  s                            @@@@r3   Ú_evaluate_imagezCOCOEvaluator._evaluate_image6  st  øøøø€ ð& #'¤-°¸°Ô"?ˆØ"&Ô"3°F¸F°NÔ"Cˆõ ˆr‰7Œ7�aŠ<ˆ<�C ™GœG qšL˜LØ�4à'Ñˆ�(ð ð 	!ð 	!ˆAØ�Œ{ð ! 8¨q°¬yÐ#DÐ#DÒ#DÐ#D¸HÒ#DÐ#DÐ#DÐ#DØ ��)‘�à ��)‘�õ ”JÐ9Ð9°bÐ9Ñ9Ô9ÀÐIÑIÔIˆ	Ø'Ð'Ð'Ð'˜YÐ'Ñ'Ô'ˆõ ”JÐ8Ð8°RÐ8Ñ8Ô8¸xÐHÑHÔHˆ	Ø2Ð2Ð2Ð2˜Y q¨ yÔ1Ð2Ñ2Ô2ˆõ
 �4”9˜V V˜^Ô,Ñ-Ô-°Ò1Ð1ð ŒI�f˜f�nÔ% a a a¨ lÔ3Ð3à”˜6 6˜>Ô*ð 	õ ˜Tœ[Ô1Ñ2Ô2ˆÝ ™GœGÐÝ˜R™œˆõ ”X˜~Ð/@ÐAÍÌÐRÑRÔRˆ
Ý”X˜~¨~Ð>ÅbÄhÐOÑOÔOˆ
å”HÐ7Ð7°BÐ7Ñ7Ô7½r¼xÐHÑHÔHˆ	Ý”H˜n¨nÐ=ÅRÄXÐNÑNÔNˆ	Ýˆt‰9Œ9˜Š>ˆ>å%.¨t¬{Ô/CÑ%DÔ%Dð  Jð  JÑ!�	˜6å$-¨b¡M¤Mð Jð J‘L�G˜Så%(¨&°)Ð)<Ñ%=Ô%=�Nà%'�Nå$-¨b¡M¤Mð 3ð 3™˜˜qõ #& a§e¢e¨I°qÑ&9Ô&9Ñ":Ô":˜Ø% i°Ð&6Ô7¸!Ò;Ð;ÀGÐ;Ø$ð +¨RÒ/Ð/Ø )¨.Ô 9Ø ò!!ð !!à )¨%Ô 0°AÒ 5Ð 5à!˜Eð   ¨ Ô/°>ÒAÐAØ-1°'¸5°.Ô-A˜NØ-2˜Nøð &¨Ò+Ð+Ø8AÀ.Ô8Q˜	 )¨WÐ"4Ñ5Ø9;¸NÔ9KÈDÔ9Q˜
 9¨gÐ#5Ñ6Ø@CÀDÄ	˜
 9¨nÐ#<Ñ=øð=JõB œ(ØFÐFÐFÐFÐFÀ2ÐFÑFÔFñ
ô 
ç
Š'�1•c˜"‘g”g�,Ñ
Ô
ð 	õ
 ”MØÝŒNØ˜a’¥¤¨?¸NÈAÑ!NÔ!Nñô ñ
ô 
ˆ	ð Ø!Ø$ØØ+Ð+¨Ð+Ñ+Ô+Ø+Ð+¨Ð+Ñ+Ô+Ø#Ø#Ø0Ð0¨RÐ0Ñ0Ô0Ø!Ø!ð
ð 
ð 	
r2   c                ó  ‡ ‡@‡A‡B‡C‡D‡E‡F— t          ‰ j        j        ¦  «        }t          ‰ j        j        ¦  «        }t          ‰ j        j        ¦  «        }t          ‰ j        j        ¦  «        }t          ‰ j        j        ¦  «        }t          ‰ j        j        ¦  «        }t          j	        |||||ft          j
        ¬¦  «         }t          j	        ||||ft          j
        ¬¦  «         }t          j	        |||||ft          j
        ¬¦  «         }	t          ‰ j        j        ¦  «        ŠDd„ ‰ j        j        D ¦   «         ŠCt          ‰ j        j        ¦  «        ŠFt          ‰ j        j        ¦  «        ŠEˆDfd„t          ‰ j        j        ¦  «        D ¦   «         }
ˆFfd„‰ j        j        D ¦   «         }ˆCfd„t          ‰ j        j        ¦  «        D ¦   «         }ˆEfd„t          ‰ j        j        ¦  «        D ¦   «         }t          |
¦  «        D �]ü\  }}||z  |z  ŠAt          |¦  «        D �]Ý\  }}||z  Š@t          |¦  «        D �]Á\  }ŠBˆ@ˆAˆ fd„|D ¦   «         }d„ |D ¦   «         }t          |¦  «        d	k    rŒ6t          j        ˆBfd
„|D ¦   «         ¦  «        }t          j        | d¬¦  «        }||         }t          j        ˆBfd„|D ¦   «         d¬¦  «        dd…|f         }t          j        ˆBfd„|D ¦   «         d¬¦  «        dd…|f         }t          j        d„ |D ¦   «         ¦  «        }t          j        |d	k    ¦  «        }|d	k    r�Œt          j        |t          j        |¦  «        ¦  «        }t          j        t          j        |¦  «        t          j        |¦  «        ¦  «        }t          j        |d¬¦  «                             t          j
        ¬¦  «        }t          j        |d¬¦  «                             t          j
        ¬¦  «        }t          t)          ||¦  «        ¦  «        D �]Ï\  } \  }!}"t          j        |!¦  «        }!t          j        |"¦  «        }"t          |!¦  «        }#|!t          j
        |¦  «        z  }$|!|"|!z   t,          z   z                       ¦   «         }%dg|z  }&dg|z  }'|#r|$d         nd	|| |||f<   t1          |#dz
  d	d¦  «        D ]%}(|%|(         |%|(dz
           k    r|%|(         |%|(dz
  <   Œ&t          j        |$‰ j        j        d¬¦  «        })|)                     ¦   «         }*t          |*¦  «        D ]d\  }+},t5          |,¦  «        }-d	|-cxk    rt          |%¦  «        k     r6n Œ1d	|-cxk    rt          |¦  «        k     rn ŒN|%|-         |&|+<   ||-         |'|+<   Œet          j        |&t          j
        ¬¦  «        || dd…|||f<   t          j        |'t          j
        ¬¦  «        |	| dd…|||f<   �ŒÑ�ŒÃ�Œß�Œþ‰ j        |||||gt6          j                             ¦   «                              d¦  «        |||	dœ‰ _        d d„}.t?          t@          ¦  «         !                    t@          j"        ¦  «        }/‰ j        j         !                    d¦  «        }0|dd…dd…dd…|/|0f         }1 |.|1¦  «        \  }2}3t?          t@          ¦  «         !                    t@          j#        ¦  «        }4|dd…dd…dd…|4|0f         }5 |.|5¦  «        \  }6}7t?          t@          ¦  «         !                    t@          j$        ¦  «        }8|dd…dd…dd…|8|0f         }9 |.|9¦  «        \  }:};t?          t@          ¦  «         !                    t@          j%        ¦  «        }<|dd…dd…dd…|<|0f         }= |.|=¦  «        \  }>}?‰ j        |||||gt6          j                             ¦   «                              d¦  «        |||	|2|3|6|7|:|;|>|?dœ‰ _        dS )!z^
        Accumulate per image evaluation results and store the result in self.results
        r  c                ó,   — h | ]}t          |¦  «        ’ŒS r1   ©Útuple)rÇ   Úas     r3   ú	<setcomp>z,COCOEvaluator._accumulate.<locals>.<setcomp>Ü  s+   € ð 3
ð 3
ð 3
Ø�E�!‰HŒHð3
ð 3
ð 3
r2   c                ó"   •— g | ]\  }}|‰v ¯	|‘ŒS r1   r1   )rÇ   ÚnÚkÚset_categoriess      €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>ã  s.   ø€ ð !
ð !
ð !
Ù�!�Q¸AÀÐ<OÐ<OˆAÐ<OÐ<OÐ<Or2   c                ó   •— g | ]}|‰v ¯|‘Œ	S r1   r1   )rÇ   ÚmÚset_max_detectionss     €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>ç  s+   ø€ ð #
ð #
ð #
Ø¨qÐ4FÐ/FÐ/FˆAÐ/FÐ/FÐ/Fr2   c                ó<   •— g | ]\  }}t          |¦  «        ‰v ¯|‘ŒS r1   rž  )rÇ   rë   r    Úset_area_rangess      €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>ë  s8   ø€ ð $
ð $
ð $
á��TÝ�T‰{Œ{˜oÐ-Ð-ð à-Ð-Ð-r2   c                ó"   •— g | ]\  }}|‰v ¯	|‘ŒS r1   r1   )rÇ   r£  rå   Úset_image_idss      €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>ñ  s.   ø€ ð 
ð 
ð 
Ù�!�Q¸AÀÐ<NÐ<NˆAÐ<NÐ<NÐ<Nr2   c                ó6   •— g | ]}‰j         ‰‰z   |z            ‘ŒS r1   )rA  )rÇ   rå   Úarea_offsetÚ
cat_offsetrN   s     €€€r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s7   ø€ ð )ð )ð )ØIJ˜œ z°KÑ'?À!Ñ'CÔDð)ð )ð )r2   c                ó   — g | ]}|®|‘ŒS rÿ   r1   ©rÇ   Úes     r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s#   € ð Gð Gð GØ¸¸˜¸¸¸r2   r   c                ó0   •— g | ]}|d          d‰…         ‘ŒS )r@   r   r1   ©rÇ   r²  r:   s     €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s&   ø€ ÐIÐIÐI°a˜˜:œ q¨ yÔ1ÐIÐIÐIr2   rV  rW  c                ó8   •— g | ]}|d          dd…d‰…f         ‘ŒS )r>   Nr   r1   r´  s     €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s.   ø€ ÐMÐMÐM¸!˜˜;œ¨¨¨¨1¨W¨9¨Ô5ÐMÐMÐMr2   r~   r	  Nc                ó8   •— g | ]}|d          dd…d‰…f         ‘ŒS )rB   Nr   r1   r´  s     €r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s.   ø€ ÐLÐLÐL¸˜˜:œ q q q¨!¨G¨) |Ô4ÐLÐLÐLr2   c                ó   — g | ]
}|d          ‘ŒS )rA   r1   r±  s     r3   rÉ   z-COCOEvaluator._accumulate.<locals>.<listcomp>  s   € Ð0VÐ0VÐ0VÀ1°°:´Ð0VÐ0VÐ0Vr2   r2  rJ   Úleft)Úsidez%Y-%m-%d %H:%M:%S)rE  ÚcountsÚdateÚ	precisionÚrecallÚscoresÚprecision_slicerP  rH   ú7tuple[npt.NDArray[np.float64], npt.NDArray[np.float64]]c                óÌ   ‡‡— | dk    Št          j        ‰| t          j        d¦  «        ¦  «        Šdˆˆfd„} |d¦  «        } |d	¦  «                             d	d
¦  «        }||fS )z<Compute average precision while handling -1 sentinel values.rJ   r2  r
  úint | tuple[int, ...]rH   rY   c           	     óö   •— ‰                      | t          j        ¬¦  «        }‰                      | ¬¦  «        }t          j        ||t          j        |j        dt          j        ¬¦  «        |dk    ¬¦  «        }|S )N)r
  r  r	  ç      ð¿r  r   )Úoutr  )r  r  r  ÚdivideÚfullr  )r
  Úsumsrº  ÚmeansÚ
valid_maskÚvalid_precisions       €€r3   Úmean_with_maskzTCOCOEvaluator._accumulate.<locals>.compute_average_precision.<locals>.mean_with_masku  st   ø€ ð '×*Ò*°½B¼JÐ*ÑGÔG�Ø#Ÿš¨T˜Ñ2Ô2�Ý13´ØØÝœ ¤
¨D½¼
ÐCÑCÔCØ  1š*ð	2ñ 2ô 2�ð �r2   )r~   rx   r~   r   )r
  rÂ  rH   rY   )r  r  r  Ú	transpose)r¿  rÌ  rK   rZ   rÊ  rË  s       @@r3   Úcompute_average_precisionz<COCOEvaluator._accumulate.<locals>.compute_average_precisionn  s‰   øø€ ð )¨BÒ.ˆJÝ œh z°?ÅBÄJÈsÁOÄOÑTÔTˆOðð ð ð ð ð ð ð (˜¨Ñ/Ô/ˆJØ)˜>¨!Ñ,Ô,×6Ò6°q¸!Ñ<Ô<ˆLØ˜|Ð+Ð+r2   r3  )rE  rº  r»  r¼  r½  r¾  ÚmAP_scores_all_sizesÚap_per_class_all_sizesÚmAP_scores_smallÚap_per_class_smallÚmAP_scores_mediumÚap_per_class_mediumÚmAP_scores_largeÚap_per_class_large)r¿  rP  rH   rÀ  )&rL   rE  r6  r7  rÃ   r9   r8  rÁ   r  Úonesr  râ   rã   Úconcatenaterj  Úcount_nonzerorˆ  Úlogical_notÚcumsumr  Úzipri  ÚEPSÚtolistr‘   Úsearchsortedr   ÚdatetimeÚnowÚstrftimerB  r±   r$  rh   r)  r*  r+  r,  )GrN   Únum_iou_thresholdsÚnum_recall_thresholdsÚnum_categoriesÚnum_area_rangesÚnum_max_detectionsÚnum_imgsr¼  r½  r¾  Úselected_category_idsÚselected_max_detectionsÚselected_area_ranges_idsÚ
image_indsÚcat_idxÚcat_eval_idxÚarea_idxÚarea_eval_idxÚmax_det_idxÚeval_img_data_rawÚeval_img_dataÚ	dt_scoresrn  Údt_scores_sortedr�  Ú
dt_ignoredÚ
gt_ignoredÚnum_non_ignored_gtÚtrue_positivesÚfalse_positivesÚtp_sumÚfp_sumÚiou_thresh_idxÚtpÚfpÚnum_tpsÚrcÚprÚprecision_at_recallÚscore_at_recallrå   Úrecall_indsÚrecall_inds_listÚriÚpos_idx_valueÚpos_idx_intrÎ  Úarea_range_idxÚmax_100_dets_idxÚaverage_precision_all_sizesrÏ  rÐ  Úsmall_area_range_idxÚaverage_precision_smallrÑ  rÒ  Úmedium_area_range_idxÚaverage_precision_mediumrÓ  rÔ  Úlarge_area_range_idxÚaverage_precision_largerÕ  rÖ  r®  r¯  r:   rª  r¥  r¬  r¨  sG   `                                                               @@@@@@@r3   Ú_accumulatezCOCOEvaluator._accumulate³  sã	  øøøøøøøø€ õ
 ! ¤Ô!5Ñ6Ô6ÐÝ # D¤KÔ$8Ñ 9Ô 9ÐÝ˜Tœ[Ô0Ñ1Ô1ˆÝ˜dœkÔ4Ñ5Ô5ˆÝ  ¤Ô!5Ñ6Ô6ÐÝ�t”{Ô*Ñ+Ô+ˆõ ”Wà"Ø%ØØØ"ðõ ”*ð	
ñ 	
ô 	
ð 	
ˆ	õ ”'Ø °ÐBTÐUÝ”*ð
ñ 
ô 
ð 
ˆõ ”'à"Ø%ØØØ"ðõ ”*ð	
ñ 	
ô 	
ð 	
ˆõ ˜Tœ[Ô0Ñ1Ô1ˆð3
ð 3
Ø"œkÔ4ð3
ñ 3
ô 3
ˆõ ! ¤Ô!5Ñ6Ô6ÐÝ˜DœKÔ/Ñ0Ô0ˆð!
ð !
ð !
ð !
Ý# D¤KÔ$7Ñ8Ô8ð!
ñ !
ô !
Ðð#
ð #
ð #
ð #
Ø”{Ô+ð#
ñ #
ô #
Ðð$
ð $
ð $
ð $
å& t¤{Ô'=Ñ>Ô>ð$
ñ $
ô $
Ð ð
ð 
ð 
ð 
Ý# D¤KÔ$7Ñ8Ô8ð
ñ 
ô 
ˆ
õ &/Ð/DÑ%EÔ%Eð c	ñ c	Ñ!ˆG�\Ø%¨Ñ7¸(ÑBˆJõ ,5Ð5MÑ+NÔ+Nð _ñ _Ñ'�˜-Ø+¨hÑ6�õ -6Ð6MÑ,NÔ,Nð [ñ [Ñ(�K ð)ð )ð )ð )ð )ð )ØNXð)ñ )ô )Ð%ðGð GØ#4ðGñ Gô G�Mõ
 ˜=Ñ)Ô)¨QÒ.Ð.Ø õ !#¤ØIÐIÐIÐI¸=ÐIÑIÔIñ!ô !�Iõ œ: y j°xÐ@Ñ@Ô@�DØ'0°¤Ð$õ "$¤ØMÐMÐMÐM¸}ÐMÑMÔMÐTUð"ñ "ô "à�a�a˜�gô"�Jõ "$¤ØLÐLÐLÐL¸mÐLÑLÔLÐSTð"ñ "ô "à�a�a˜�gô"�Jõ
 "$¤Ð0VÐ0VÈÐ0VÑ0VÔ0VÑ!WÔ!W�JÝ)+Ô)9¸*Èº/Ñ)JÔ)JÐ&ð *¨QÒ.Ð.Ù õ &(¤^Ø"¥B¤N°:Ñ$>Ô$>ñ&ô &�Nõ ')¤nÝœ zÑ2Ô2µB´NÀ:Ñ4NÔ4Nñ'ô '�Oõ  œY ~¸AÐ>Ñ>Ô>×EÒEÍBÌJÐEÑWÔW�FÝœY ¸QÐ?Ñ?Ô?×FÒFÍRÌZÐFÑXÔX�Fõ 5>½cÀ&È&Ñ>QÔ>QÑ4RÔ4Rð ,ñ ,Ñ0˜©¨¨RÝœX b™\œ\˜ÝœX b™\œ\˜Ý"% b¡'¤'˜à¥"¤*Ð-?Ñ"@Ô"@Ñ@˜à  B¨¡G­c¡MÑ2×:Ò:Ñ<Ô<˜à/2¨eÐ6KÑ.KÐ+à+.¨%Ð2GÑ*G˜ð
 '.Ð4˜B˜rœF˜F°1ð ˜~¨w¸À+ÐMÑNõ
 "' w°¡{°A°rÑ!:Ô!:ð 2ð 2˜AØ! !œu r¨!¨a©%¤yÒ0Ð0Ø,.¨q¬E  1 q¡5¡	øå<>¼OØ ¤Ô 4¸6ð=ñ =ô =˜ð 7B×6HÒ6HÑ6JÔ6JÐ(Ý1:Ð;KÑ1LÔ1Lð Tð TÑ-˜B å/2°=Ñ/AÔ/A˜KØ  KÐ9Ð9Ò9Ð9µ#°b±'´'Ò9Ð9Ð9Ð9Ð9¸aÀ;ð ?ð ?ò ?ð ?ÕQTØ 0ñRô Rò ?ð ?ð ?ð ?ð ?ð ;=¸[¼/Ð 3°BÑ 7Ø6FÀ{Ô6S °Ñ 3øõ œHÐ%8ÅÄ
ÐKÑKÔKð " .°!°!°!°W¸hÈÐ"SÑTõ
 œH _½B¼JÐGÑGÔGð ˜~¨q¨q¨q°'¸8À[ÐPÑQÑQñU,ñ_[ñ	_ðD ”kà"Ø%ØØØ"ðõ Ô%×)Ò)Ñ+Ô+×4Ò4Ð5HÑIÔIØ"ØØð
ð 
ˆŒð 	,ð 	,ð 	,ð 	,õ2 �jÑ)Ô)×/Ò/µ
´Ñ?Ô?ˆØœ;Ô/×5Ò5°cÑ:Ô:Ðà&/ØˆAˆAˆqˆqˆq�!�!�!�^Ð%5Ð5ô'
Ð#ð
 8QÐ7PØ'ñ8
ô 8
Ñ4ÐÐ4õ
  $¥JÑ/Ô/×5Ò5µjÔ6FÑGÔGÐØ"+ØˆAˆAˆqˆqˆq�!�!�!Ð)Ð+;Ð;ô#
Ðð 0IÐ/HØ#ñ0
ô 0
Ñ,ÐÐ,õ
 !%¥ZÑ 0Ô 0× 6Ò 6µzÔ7HÑ IÔ IÐØ#,ØˆAˆAˆqˆqˆq�!�!�!Ð*Ð,<Ð<ô$
Ð ð 2KÐ1JØ$ñ2
ô 2
Ñ.ÐÐ.õ
  $¥JÑ/Ô/×5Ò5µjÔ6FÑGÔGÐØ"+ØˆAˆAˆqˆqˆq�!�!�!Ð)Ð+;Ð;ô#
Ðð 0IÐ/HØ#ñ0
ô 0
Ñ,ÐÐ,ð
 ”kà"Ø%ØØØ"ðõ Ô%×)Ò)Ñ+Ô+×4Ò4Ð5HÑIÔIØ"ØØØ$8Ø&<Ø 0Ø"4Ø!2Ø#6Ø 0Ø"4ð)
ð 
ˆŒˆˆr2   c                ó¸   ‡ ‡— ddt           j        dfdˆ fd„Šdˆˆ fd„}t          ‰ j        ¦  «        dk    r# |¦   «                              ¦   «         ‰ _        dS dS )zM
        Compute and display summary metrics for evaluation results.
        TNr3  Úuse_aprF   Úiou_thrúfloat | Noner9   r$  r8  r   rH   r   c           
     óÌ  •— d}| rdnd}| rdnd}|€)‰j         j        d         d›d‰j         j        d	         d›�n|d›}t          t          ¦  «        }|                     |¦  «        }	‰j         j                             |¦  «        }
| rqt          j        ‰j        d
         t          j	        ¬¦  «        }|�0t          j
        |‰j         j        k    ¦  «        d         }||         }|d d …d d …d d …|	|
f         }nmt          j        ‰j        d         t          j	        ¬¦  «        }|�0t          j
        |‰j         j        k    ¦  «        d         }||         }|d d …d d …|	|
f         }t          ||d	k             ¦  «        dk    rd}n+t          t          j        ||d	k             ¦  «        ¦  «        }t                               |                     ||||||¦  «        ¦  «         |S )NzC {:<18} {} @[ IoU={:<9} | area={:>6s} | maxDets={:>3d} ] = {:0.10f}zAverage PrecisionzAverage Recallz(AP)z(AR)r   z0.2fú:rJ   r¼  r  r½  rÄ  )rE  r6  r±   r$  rh   r8  r  r  rB  r  r  rL   r   rM   ÚloggerÚinfoÚformat)r  r  r9   r8  ÚiStrÚtitleStrÚtypeStrÚiou_strÚall_object_sizesr
  Úmax_detections_idxÚsÚtÚmean_srN   s                 €r3   Ú
_summarizez8COCOEvaluator._pycocotools_summarize.<locals>._summarizeÊ  s
  ø€ ð YˆDØ.4ÐJÐ*Ð*Ð:JˆHØ &Ð2�f�f¨FˆGð �?ð ”;Ô'¨Ô*ÐQÐQÐQ°$´+Ô2FÀrÔ2JÐQÐQÐQÐQàÐ&Ð&ð õ
  $¥JÑ/Ô/ÐØ-×3Ò3°JÑ?Ô?ˆNØ!%¤Ô!5×!;Ò!;¸HÑ!EÔ!EÐØð @õ .0¬ZØ”L Ô-µR´Zð.ñ .ô .�ð Ð&Ýœ ¨D¬KÔ,@Ò!@ÑAÔAÀ!ÔD�AØ˜!œ�AØ�a�a�a˜˜˜˜A˜A˜A˜~Ð/AÐAÔB��õ ”J˜tœ|¨HÔ5½R¼ZÐHÑHÔH�ØÐ&Ýœ ¨D¬KÔ,@Ò!@ÑAÔAÀ!ÔD�AØ˜!œ�AØ�a�a�a˜˜˜˜NÐ,>Ð>Ô?�Ý�1�Q˜’V”9‰~Œ~ Ò"Ð"Ø��å�rœw q¨¨Rª¤yÑ1Ô1Ñ2Ô2�Ý�KŠKØ—’˜H g¨w¸
ÀHÈfÑUÔUñô ð ð ˆMr2   rP  c                 ó®  •— t          j        dt           j        ¬¦  «        }  ‰d¬¦  «        | d<    ‰dd‰j        j        d         ¬¦  «        | d	<    ‰dd
‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<    ‰d‰j        j        d         ¬¦  «        | d<    ‰d‰j        j        d	         ¬¦  «        | d<    ‰d‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<    ‰dt
          j        ‰j        j        d         ¬¦  «        | d<   | S )N)é   r  T)r  r   r0  rx   )r  r  r8  r~   g      è?)r  r9   r8  rw   ry   rW   F)r  r8  r|   é   é   é	   r{   é   )	r  rƒ  r  rE  r8  r$  r*  r+  r,  )rF  r&  rN   s    €€r3   Ú_summarize_predictionszDCOCOEvaluator._pycocotools_summarize.<locals>._summarize_predictions÷  s   ø€ Ý-/¬X°eÅ2Ä:Ð-NÑ-NÔ-NˆEØ!�z¨Ð.Ñ.Ô.ˆE�!‰HØ!�zØ S°4´;Ô3GÈÔ3Jðñ ô ˆE�!‰Hð "�zØ T°D´KÔ4HÈÔ4Kðñ ô ˆE�!‰Hð "�zØÝ%Ô+ØœÔ-¨aÔ0ðñ ô ˆE�!‰Hð
 "�zØÝ%Ô,ØœÔ-¨aÔ0ðñ ô ˆE�!‰Hð
 "�zØÝ%Ô+ØœÔ-¨aÔ0ðñ ô ˆE�!‰Hð
 "�z¨¸¼Ô9MÈaÔ9PÐQÑQÔQˆE�!‰HØ!�z¨¸¼Ô9MÈaÔ9PÐQÑQÔQˆE�!‰HØ!�z¨¸¼Ô9MÈaÔ9PÐQÑQÔQˆE�!‰HØ!�zØÝ%Ô+ØœÔ-¨aÔ0ðñ ô ˆE�!‰Hð
 #˜
ØÝ%Ô,ØœÔ-¨aÔ0ðñ ô ˆE�"‰Ið
 #˜
ØÝ%Ô+ØœÔ-¨aÔ0ðñ ô ˆE�"‰Ið
 ˆLr2   r   )
r  rF   r  r  r9   r$  r8  r   rH   r   )rH   rP  )r$  r)  rL   rB  rÞ  rF  )rN   r-  r&  s   ` @r3   Ú_pycocotools_summarizez$COCOEvaluator._pycocotools_summarizeÅ  s›   øø€ ð  Ø$(Ý%/¤^Øð	+	ð +	ð +	ð +	ð +	ð +	ð +	ðZ*	ð *	ð *	ð *	ð *	ð *	ð *	õX ˆtŒ|ÑÔ Ò!Ð!Ø/Ð/Ñ1Ô1×8Ò8Ñ:Ô:ˆDŒJˆJˆJð "Ð!r2   c                ó  ‡ ‡— t          t          j        ‰ j        j        ¦  «        ¦  «        ‰ j        _        t          t          j        ‰ j        j        ¦  «        ¦  «        ‰ j        _        t          ‰ j        j        ¦  «        ‰ j        _        ‰                      ¦   «          ˆ fd„‰ j        j        D ¦   «         ‰ _	        ‰ j        j        d         Šˆˆ fd„‰ j        j        D ¦   «         ‰ _
        ‰                      ¦   «          dS )z„
        Start the per image evaluation on all images and keeep results in
        self.eval_imgs (a list of dictionaries).
        c                ó`   •— i | ]*}‰j         j        D ]}||f‰                     ||¦  «        “ŒŒ+S r1   )rE  rÃ   rs  )rÇ   rÈ   ræ   rN   s      €r3   ú
<dictcomp>z*COCOEvaluator.evaluate.<locals>.<dictcomp>3  s\   ø€ ð 
ð 
ð 
àØœ+Ô-ð
ð 
ð ð �VÐ˜d×/Ò/°¸Ñ?Ô?ð
ð 
ð 
ð 
r2   rJ   c           
     ó|   •— g | ]8}‰j         j        D ])}‰j         j        D ]}‰                     |||‰¦  «        ‘ŒŒ*Œ9S r1   )rE  r9   rÁ   r›  )rÇ   ræ   r9   rÈ   r:   rN   s       €€r3   rÉ   z*COCOEvaluator.evaluate.<locals>.<listcomp>=  su   ø€ ð 
ð 
ð 
àØ"œkÔ4ð
ð 
ð Øœ+Ô-ð	
ð 
ð ð × Ò  ¨°¸WÑEÔEð
ð 
ð 
ð 
ð 
r2   N)r±   r  ÚuniquerE  rÁ   rÃ   rH  r8  rO  rG  rA  r  )rN   r:   s   `@r3   ÚevaluatezCOCOEvaluator.evaluate&  sÿ   øø€ õ #¥2¤9¨T¬[Ô-@Ñ#AÔ#AÑBÔBˆŒÔÝ"¥2¤9¨T¬[Ô-@Ñ#AÔ#AÑBÔBˆŒÔÝ% d¤kÔ&:Ñ;Ô;ˆŒÔà×-Ò-Ñ/Ô/Ð/ð
ð 
ð 
ð 
àœ+Ô-ð
ñ 
ô 
ˆŒ	ð ”+Ô& rÔ*ˆð
ð 
ð 
ð 
ð 
àœ+Ô-ð
ñ 
ô 
ˆŒð 	×ÒÑÔÐÐÐr2   N)r>  r©   r?  r©   rE   r   rH   rt   r¥   )rÈ   r   ræ   r   rH   rP  )
rÈ   r   ræ   r   r9   r8   r:   r   rH   rt  )r-   r.   r/   r¦   r   ÚBOXESrµ   rO  rs  r›  r  r.  r4  r1   r2   r3   r=  r=  ¨  sÎ   € € € € € ðð ð '3Ô&8ð	,Kð ,Kð ,Kð ,Kð ,Kð\!ð !ð !ð !ðF6&ð 6&ð 6&ð 6&ðp{
ð {
ð {
ð {
ðzP
ð P
ð P
ð P
ðd_;ð _;ð _;ð _;ðBð ð ð ð ð r2   r=  c                  óf   — e Zd ZdZej        dddfd&d„Zd'd„Zd(d„Zd)d„Z	d*d„Z
d+d!„Zd,d#„Zd-d%„ZdS ).ÚMeanAveragePrecisiona  
    Mean Average Precision (mAP) is a metric used to evaluate object detection models.
    It is the average of the precision-recall curves at different IoU thresholds.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> from supervision.metrics import MeanAveragePrecision
        >>> predictions = sv.Detections(
        ...     xyxy=np.array([[0, 0, 10, 10]]),
        ...     class_id=np.array([0]),
        ...     confidence=np.array([0.9])
        ... )
        >>> targets = sv.Detections(
        ...     xyxy=np.array([[0, 0, 10, 10]]),
        ...     class_id=np.array([0])
        ... )
        >>> map_metric = MeanAveragePrecision()
        >>> map_result = map_metric.update(predictions, targets).compute()
        >>> round(float(map_result.map50), 2)
        1.0
        >>> print(map_result)
        Average Precision (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 1.000
        Average Precision (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 1.000
        Average Precision (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 1.000
        Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 1.000
        Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = -1.000
        Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = -1.000

        ```

    ![example_plot](
        https://media.roboflow.com/supervision-docs/metrics/mAP_plot_example.png
    ){ align=center width="800" }
    FNrE   r   Úclass_agnosticrF   Úclass_mappingúdict[int, int] | NoneÚimage_indicesrÂ   rH   rt   c                óZ   — || _         || _        g | _        g | _        || _        || _        dS )aW  
        Initialize the Mean Average Precision metric.

        Args:
            metric_target: The type of detection data to use.
            class_agnostic: Whether to treat all data as a single class.
            class_mapping: A dictionary to map class IDs to new IDs.
            image_indices: The indices of the images to use.
        N)Ú_metric_targetÚ_class_agnosticÚ_predictions_listÚ_targets_listÚ_class_mappingÚ_image_indices)rN   rE   r8  r9  r;  s        r3   rµ   zMeanAveragePrecision.__init__n  s:   € ð  ,ˆÔØ-ˆÔà35ˆÔØ/1ˆÔØ+ˆÔØ+ˆÔÐÐr2   c                ó"   — g | _         g | _        dS )zR
        Reset the metric to its initial state, clearing all stored data.
        N)r?  r@  rS   s    r3   ÚresetzMeanAveragePrecision.reset†  s   € ð "$ˆÔØˆÔÐÐr2   rí   úDetections | list[Detections]rª   c                ó  — t          |t          ¦  «        s|g}t          |t          ¦  «        s|g}t          |¦  «        t          |¦  «        k    r0t          dt          |¦  «        › dt          |¦  «        › d�¦  «        ‚| j        rNt          |¦  «        }t          |¦  «        }|D ]}|j        �d|j        dd…<   Œ|D ]}|j        �d|j        dd…<   Œ| j                             |¦  «         | j	                             |¦  «         | S )a  
        Add new predictions and targets to the metric, but do not compute the result.

        Args:
            predictions: The predicted detections.
            targets: The ground-truth detections.

        Returns:
            The updated metric instance.
        úThe number of predictions (ú) and targets (z%) during the update must be the same.NrJ   )
rñ   r±   rL   rò   r>  r   Úclass_idr?  Úextendr@  )rN   rí   rª   Ú
predictionÚtargets        r3   ÚupdatezMeanAveragePrecision.update�  sD  € õ ˜+¥tÑ,Ô,ð 	(Ø&˜-ˆKÝ˜'¥4Ñ(Ô(ð 	 Ø�iˆGåˆ{ÑÔ�s 7™|œ|Ò+Ð+ÝðQ­c°+Ñ.>Ô.>ð Qð QÝ  ™\œ\ðQð Qð Qñô ð ð
 Ôð 		,Ý" ;Ñ/Ô/ˆKÝ˜wÑ'Ô'ˆGà)ð 0ð 0�
ØÔ&Ð2Ø-/�JÔ'¨¨¨Ñ*øØ!ð ,ð ,�Ø”?Ð.Ø)+�F”O A A AÑ&øàÔ×%Ò% kÑ2Ô2Ð2ØÔ×!Ò! 'Ñ*Ô*Ð*àˆr2   Ú
detectionsr   únpt.NDArray[Any] | Nonec                óB  — | j         t          j        k    st          |¦  «        dk    rdS | j         t          j        k    rG|j        €t          d¦  «        ‚t          j        |j        ¦  «         	                    t          ¦  «        S | j         t          j        k    rn|j                             t          ¦  «        }|€t          dt          › d�¦  «        ‚t          j        |t          j        ¬¦  «                             ddd	¦  «        S t          d
| j         › �¦  «        ‚)z‚Return per-detection masks or oriented boxes for the metric target,
        or `None` for the box target and for empty detections.r   Nz^MeanAveragePrecision with `MetricTarget.MASKS` requires masks on both predictions and targets.zKMeanAveragePrecision with `MetricTarget.ORIENTED_BOUNDING_BOXES` requires `z,` in `data` on both predictions and targets.r  rJ   ry   rx   zInvalid metric target: )r=  r   r5  rL   rk  Úmaskrò   r  r  r  rF   rl  ÚdatarJ  r   r  r  )rN   rN  Úobbs      r3   Ú_detections_contentz(MeanAveragePrecision._detections_content·  s  € ð Ô¥,Ô"4Ò4Ð4½¸J¹¼È1Ò8LÐ8LØ�4ØÔ¥,Ô"4Ò4Ð4ØŒÐ&Ý ð>ñô ð õ ”:˜jœoÑ.Ô.×5Ò5µdÑ;Ô;Ð;ØÔ¥,Ô"FÒFÐFØ”/×%Ò%Õ&>Ñ?Ô?ˆCØˆ{Ý ð0å1ð0ð 0ð 0ñô ð õ ”:˜c­¬Ð4Ñ4Ô4×<Ò<¸RÀÀAÑFÔFÐFÝÐH°4Ô3FÐHÐHÑIÔIÐIr2   Úxywhr   r+   rë   r   r   c                ó®  — |€t          |d         |d         z  ¦  «        S | j        t          j        k    r't          t	          j        ||         ¦  «        ¦  «        S ||dd…df         ||dd…df         }}t          dt          t	          j        |t	          j        |d¦  «        z  t	          j        |d¦  «        |z  z
  ¦  «        ¦  «        z  ¦  «        S )z�Compute the default annotation area for the metric target: bbox area
        for boxes, pixel count for masks, polygon area for oriented boxes.Nrx   rw   r   r~   r0  rJ   )	r   r=  r   rk  r  rÙ  Úabsr  Úroll)rN   rU  r+   rë   rö   r÷   s         r3   Ú_content_areaz"MeanAveragePrecision._content_areaÏ  sÁ   € ð
 ˆ?Ý˜˜aœ 4¨¤7Ñ*Ñ+Ô+Ð+ØÔ¥,Ô"4Ò4Ð4Ý�Ô)¨'°#¬,Ñ7Ô7Ñ8Ô8Ð8Ø�s˜A˜A˜A˜q�yÔ! 7¨3°°°°1¨9Ô#5ˆ1ˆå�S�3�rœv a­"¬'°!°R©.¬.Ñ&8½2¼7À1Àb¹>¼>ÈAÑ;MÑ&MÑNÔNÑOÔOÑOÑPÔPÐPr2   úlist[Detections]údict[str, list[_TypeCocoDict]]c           
     ó  ‡ — d„ t          t          |¦  «        ¦  «        D ¦   «         }‰ j        �ˆ fd„|D ¦   «         }g }t          |¦  «        D �]\  }}‰ j        �‰ j        |         }|j        €Œ"‰                      |¦  «        }t          |j        ¦  «        D �]Í\  }}|d         |d         |d         |d         z
  |d         |d         z
  g}	d}
|j        �>|j        |         }‰ j        �‰ j        t          |¦  «                 }
nt          |¦  «        }
d}|j	        �St          |j	        v rEt          j        |j	        t                   t          j        ¬¦  «        }t          ||         ¦  «        }|€‰                      |	||¦  «        }d}|j	        �Id	|j	        v r@t          j        |j	        d	         t          j        ¬¦  «        }t          ||         ¦  «        }d}|j	        �Id
|j	        v r@t          j        |j	        d
         t          j        ¬¦  «        }t          ||         ¦  «        }||||	|
t          |¦  «        dz   |dœ}|�||         |d<    |j        |¦  «         �ŒÏ�Œd„ |D ¦   «         }d„ |D ¦   «         }|||dœS )zJTransform targets into a dictionary that can be used by the COCO evaluatorc                ó   — g | ]}d |i‘ŒS rÐ   r1   )rÇ   rÈ   s     r3   rÉ   z9MeanAveragePrecision._prepare_targets.<locals>.<listcomp>à  s   € Ð&XÐ&XÐ&X¸&¨¨f ~Ð&XÐ&XÐ&Xr2   Nc                ó:   •— g | ]}d ‰j         |d                   i‘ŒS rÐ   )rB  )rÇ   r¿   rN   s     €r3   rÉ   z9MeanAveragePrecision._prepare_targets.<locals>.<listcomp>â  s*   ø€ ÐOÐOÐOÀ�t˜TÔ0°°T´Ô;Ð<ÐOÐOÐOr2   r   r~   rx   rw   r  r!   r"   )r    r!   r   r   r   r   r"   r+   c                ó   — h | ]
}|d          ’ŒS rË   r1   )rÇ   Ú
annotations     r3   r¡  z8MeanAveragePrecision._prepare_targets.<locals>.<setcomp>%  s   € ÐOÐOÐO°Z�z -Ô0ÐOÐOÐOr2   c                ó   — g | ]}d |i‘ŒS rÐ   r1   )rÇ   ræ   s     r3   rÉ   z9MeanAveragePrecision._prepare_targets.<locals>.<listcomp>&  s   € Ð*TÐ*TÐ*T¸f¨D°&¨>Ð*TÐ*TÐ*Tr2   )r»   r   r¼   )r‘   rL   rB  rã   ÚxyxyrT  rI  rA  r   rR  r   r  r  r  r   rY  r„  r½   )rN   rª   r»   r   r   Úimage_targetsr+   Ú
target_idxrb  rU  r   Úcls_idr    Ú	area_datar!   Úiscrowd_datar"   Úignore_dataÚdict_annotationÚall_cat_idsr¼   s   `                    r3   Ú_prepare_targetsz%MeanAveragePrecision._prepare_targetsÜ  sø  ø€ ð 'YÐ&XÅEÍ#ÈgÉ,Ì,ÑDWÔDWÐ&XÑ&XÔ&XˆØÔÐ*ØOÐOÐOÐOÈÐOÑOÔOˆFà+-ˆÝ'0°Ñ'9Ô'9ð >	4ñ >	4Ñ#ˆH�mØÔ"Ð.ØÔ.¨xÔ8�ð Ô!Ð)Øà×.Ò.¨}Ñ=Ô=ˆGÝ$-¨mÔ.@Ñ$AÔ$Að 54ñ 54Ñ �
˜DØ˜Qœ  a¤¨$¨q¬'°D¸´GÑ*;¸TÀ!¼WÀtÈAÄwÑ=NÐO�ð  �à Ô)Ð5Ø*Ô3°JÔ?�FØÔ*Ð6Ø&*Ô&9½#¸f¹+¼+Ô&F˜˜å&)¨&¡k¤k˜ð �à!Ô&Ð2Ý'¨=Ô+=Ð=Ð=å9;¼Ø%Ô*­?Ô;Å2Ä:ð:ñ :ô :�Iõ ! ¨:Ô!6Ñ7Ô7�Dà�<Ø×-Ò-¨d°G¸ZÑHÔH�Dà�Ø Ô%Ð1°iÀ=ÔCUÐ6UÐ6UÝ:<¼*Ø%Ô*¨9Ô5½R¼Xð;ñ ;ô ;�Lõ " ,¨zÔ":Ñ;Ô;�Gà�Ø Ô%Ð1°hÀ-ÔBTÐ6TÐ6TÝ9;¼Ø%Ô*¨8Ô4½B¼Hð:ñ :ô :�Kõ ! ¨ZÔ!8Ñ9Ô9�Fð !Ø&Ø (Ø Ø#.Ý˜kÑ*Ô*¨QÑ.Ø$ð2ð 2�ð Ð&Ø18¸Ô1D�O IÑ.Ø"�Ô" ?Ñ3Ô3Ð3Ñ3ñk54ðn PÐOÀ;ÐOÑOÔOˆØ*TÐ*TÈÐ*TÑ*TÔ*Tˆ
ð Ø&Ø$ð
ð 
ð 	
r2   rè   c           
     óf  — g }t          |¦  «        D �]�\  }}| j        �| j        |         }|j        €Œ"|                      |¦  «        }t          |j        ¦  «        D �]P\  }}|d         |d         |d         |d         z
  |d         |d         z
  g}d}	d}
|j        �>|j        |         }| j        �| j        t          |¦  «                 }	nt          |¦  «        }	|j        �t          |j        |         ¦  «        }
d}|j	        �St          |j	        v rEt          j        |j	        t                   t          j        ¬¦  «        }t          ||         ¦  «        }|€|                      |||¦  «        }|||
|	|t          |¦  «        dz   dœ}|�||         |d	<   |                     |¦  «         �ŒR�ŒŸ|S )
z`Transform predictions into a list of predictions that can be used by the COCO
        evaluator.Nr   r~   rx   rw   r2  r  )r   r   r$   r   r    r   r+   )rã   rB  rb  rT  rI  rA  r   Ú
confidencer   rR  r   r  r  r  rY  rL   r½   )rN   rí   r?  r   Úimage_predictionsr+   Úpred_idxrb  rU  r   r$   re  r    rf  Údict_predictions                  r3   Ú_prepare_predictionsz)MeanAveragePrecision._prepare_predictions.  só  € ð
 13ÐÝ+4°[Ñ+AÔ+Að 1	9ñ 1	9Ñ'ˆHÐ'ØÔ"Ð.ØÔ.¨xÔ8�à Ô%Ð-Øà×.Ò.Ð/@ÑAÔAˆGÝ"+Ð,=Ô,BÑ"CÔ"Cð )9ñ )9‘�˜$Ø˜Qœ  a¤¨$¨q¬'°D¸´GÑ*;¸TÀ!¼WÀtÈAÄwÑ=NÐO�à�Ø�à$Ô-Ð9Ø.Ô7¸ÔA�FØÔ*Ð6Ø&*Ô&9½#¸f¹+¼+Ô&F˜˜å&)¨&¡k¤k˜à$Ô/Ð;Ý!Ð"3Ô">¸xÔ"HÑIÔI�Eð �à%Ô*Ð6Ý'Ð+<Ô+AÐAÐAå9;¼Ø)Ô.­Ô?ÅrÄzð:ñ :ô :�Iõ ! ¨8Ô!4Ñ5Ô5�Dà�<Ø×-Ò-¨d°G¸XÑFÔF�Dð !)Ø Ø"Ø#.Ø ÝÐ.Ñ/Ô/°!Ñ3ð2ð 2�ð Ð&Ø18¸Ô1B�O IÑ.Ø ×'Ò'¨Ñ8Ô8Ð8Ñ8ñS)9ðT  Ðr2   rD   c                ó¤  — t          | j        ¦  «        }t          | j        ¦  «        }||k    rt          d|› d|› d�¦  «        ‚|                      | j        ¦  «        }|                      | j        ¦  «        }t          |¬¦  «        }|                     |¦  «        }t          ||| j	        ¬¦  «        }| 
                    ¦   «          t          | j	        | j        t          j        |j        d         t          j        ¬¦  «        t          j        |j        d         t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        ¬	¦  «        }t          | j	        | j        t          j        |j        d
         t          j        ¬¦  «        t          j        |j        d         t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        ¬	¦  «        }	t          | j	        | j        t          j        |j        d         t          j        ¬¦  «        t          j        |j        d         t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        ¬	¦  «        }
t          | j	        | j        t          j        |j        d         t          j        ¬¦  «        t          j        |j        d         t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        t          j        |j        j        t          j        ¬¦  «        ||	|
¬¦	  «	        }|S )a6  
        Calculate Mean Average Precision based on predicted and ground-truth
        detections at different thresholds using the COCO evaluation metrics.
        Source: https://github.com/rafaelpadilla/review_object_detection_metrics

        Returns:
            The Mean Average Precision result.
        rG  rH  z)) during the evaluation must be the same.r·   )rE   rÑ  r  rÒ  )rE   rG   rK   rZ   r[   r\   rÓ  rÔ  rÕ  rÖ  rÏ  rÐ  )	rE   rG   rK   rZ   r[   r\   r]   r^   r_   )rL   r?  r@  rò   rk  rq  r©   rþ   r=  r=  r4  rD   r>  r  r  rB  r  rE  r6  rÃ   Úint32)rN   Útotal_images_predictionsÚtotal_images_targetsÚdict_targetsÚlst_predictionsÚcoco_gtÚcoco_detÚcocoEvalÚ	mAP_smallÚ
mAP_mediumÚ	mAP_largeÚ
mAP_results               r3   ÚcomputezMeanAveragePrecision.computeh  su  € õ $' tÔ'=Ñ#>Ô#>Ð Ý" 4Ô#5Ñ6Ô6Ðà#Ð';Ò;Ð;ÝðÐ.Fð ð Ø1ðð ð ñô ð ð
 ×,Ò,¨TÔ-?Ñ@Ô@ˆØ×3Ò3°DÔ4JÑKÔKˆå#¨LÐ9Ñ9Ô9ˆà×+Ò+¨OÑ<Ô<ˆå  ¨(À$ÔBUÐVÑVÔVˆð 	×ÒÑÔÐõ /ØÔ-Ø"Ô2Ý”zØÔ Ð!3Ô4½B¼Jðñ ô õ œØÔ Ð!5Ô6½b¼jðñ ô õ œ: h¤oÔ&>ÅbÄjÐQÑQÔQÝœJ x¤Ô'>ÅbÄhÐOÑOÔOð
ñ 
ô 
ˆ	õ 0ØÔ-Ø"Ô2Ý”zØÔ Ð!4Ô5½R¼Zðñ ô õ œØÔ Ð!6Ô7½r¼zðñ ô õ œ: h¤oÔ&>ÅbÄjÐQÑQÔQÝœJ x¤Ô'>ÅbÄhÐOÑOÔOð
ñ 
ô 
ˆ
õ /ØÔ-Ø"Ô2Ý”zØÔ Ð!3Ô4½B¼Jðñ ô õ œØÔ Ð!5Ô6½b¼jðñ ô õ œ: h¤oÔ&>ÅbÄjÐQÑQÔQÝœJ x¤Ô'>ÅbÄhÐOÑOÔOð
ñ 
ô 
ˆ	õ 0ØÔ-Ø"Ô2Ý”zØÔ Ð!7Ô8ÅÄ
ðñ ô õ œØÔ Ð!9Ô:Å"Ä*ðñ ô õ œ: h¤oÔ&>ÅbÄjÐQÑQÔQÝœJ x¤Ô'>ÅbÄhÐOÑOÔOØ#Ø%Ø#ð
ñ 
ô 
ˆ
ð Ðr2   )
rE   r   r8  rF   r9  r:  r;  rÂ   rH   rt   r¥   )rí   rE  rª   rE  rH   r7  )rN  r   rH   rO  )rU  r   r+   rO  rë   r   rH   r   )rª   rZ  rH   r[  )rí   rZ  rH   rè   )rH   rD   )r-   r.   r/   r¦   r   r5  rµ   rD  rM  rT  rY  rk  rq  r  r1   r2   r3   r7  r7  H  sê   € € € € € ð#ð #ðN '3Ô&8Ø$Ø/3Ø*.ð,ð ,ð ,ð ,ð ,ð0 ð  ð  ð  ð(ð (ð (ð (ðTJð Jð Jð Jð0Qð Qð Qð QðP
ð P
ð P
ð P
ðd8 ð 8 ð 8 ð 8 ðtVð Vð Vð Vð Vð Vr2   r7  )r  r  r  r  r  r  rH   rY   )@Ú
__future__r   rô   rà  rÑ   Úcollectionsr   r   Údataclassesr   Úenumr   Útypingr   r	   r
   r   Únumpyr  Únumpy.typingÚnptÚsupervision.configr   r   Úsupervision.detection.corer   Ú'supervision.detection.utils.iou_and_nmsr   r   r   Úsupervision.draw.colorr   Úsupervision.metrics.corer   r   Úsupervision.metrics.utils.utilsr   Úsupervision.utils.loggerr   r-   r  ri   rm   r   Údictr&   r±   r5   r0   r7   rD   r©   r:  r;  r9  Úfinfor  r  rÝ  r"  r$  r.  r=  r7  r1   r2   r3   ú<module>r‘     s†  ðØ "Ð "Ð "Ð "Ð "Ð "Ð "à €€€Ø €€€Ø Ð Ð Ð Ø #Ð #Ð #Ð #Ð #Ð #Ø Ð Ð Ð Ð Ð Ø !Ð !Ð !Ð !Ð !Ð !Ø Ð Ð Ð Ð Ð Ø ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;Ð ;à Ð Ð Ð Ø Ð Ð Ð Ð Ð à HÐ HÐ HÐ HÐ HÐ HÐ HÐ HØ 1Ð 1Ð 1Ð 1Ð 1Ð 1ðð ð ð ð ð ð ð ð ð ð
 8Ð 7Ð 7Ð 7Ð 7Ð 7Ø 9Ð 9Ð 9Ð 9Ð 9Ð 9Ð 9Ð 9Ø CÐ CÐ CÐ CÐ CÐ CØ 0Ð 0Ð 0Ð 0Ð 0Ð 0à	ˆ�XÑ	Ô	€àð ØÐÐÐðð ð ð ð �I Uð ñ ô ð ð, # 3¨¨]Ô(;Ð#;Ô<Ð Ð <Ð <Ð <Ñ <ð$ð $ð $ð $ð $ ñ $ô $ð $ð ðYð Yð Yð Yð Yñ Yô Yñ „ðYðxV#ð V#ð V#ð V#ð V#ñ V#ô V#ð V#ðt Ð ØÐ ØÐ ð €b„hˆrŒzÑÔÔ€ð-ð -ð -ð -ð`ð ð ð ð �ñ ô ð ð#
ð #
ð #
ð #
ð #
ñ #
ô #
ð #
ðL]
ð ]
ð ]
ð ]
ð ]
ñ ]
ô ]
ð ]
ð@vð vð vð vð v˜6Ð"<Ô=ñ vô vð vð vð vr2   