§
    ’©ºj  ã                   ó´   — 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edede j	        d	e j
        d
ef
d„Zdededed
efd„Z	 	 ddedededed         d
ef
d„ZdS )é    N)ÚTensor)Úconv2d)ÚLiteral)Údim_zero_catÚwin_sizeÚsigmaÚdtypeÚdeviceÚreturnc                 ó  — t          j        | ||¬¦  «        | dz
  dz  z
  }|dz  }t          j        |                     d¦  «        |                     d¦  «        z    d|dz  z  z  ¦  «        }|t          j        |¦  «        z  }|S )N©r	   r
   é   é   r   ç       @)ÚtorchÚarangeÚexpÚ	unsqueezeÚsum)r   r   r	   r
   ÚcoordsÚgs         úq/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/torchmetrics/functional/image/vif.pyÚ_filterr      s�   € õ
 Œ\˜(¨%¸Ð?Ñ?Ô?À8ÈaÁ<ÐSTÑBTÑT€FØ�‰	€AÝŒ	�A—K’K ‘N”N Q§[¢[°¡^¤^Ñ3Ð4¸¸eÀQ¹h¹ÑGÑHÔH€AØ�Œ�1‰ŒÑ€AØ€Hó    ÚpredsÚtargetÚ
sigma_n_sqc           
      óz  — | j         }| j        }|                      d¦  «        } |                     d¦  «        }t          j        d||¬¦  «        }t          j        |||¬¦  «        }t          j        |                      d¦  «        ||¬¦  «        }t          j        |                      d¦  «        ||¬¦  «        }t          d¦  «        D �]ì}dd|z
  z  dz   }	t          |	|	dz  ||¬¦  «        d d d d …f         }
|dk    rHt          ||
¦  «        d d …d d …d d d…d d d…f         }t          | |
¦  «        d d …d d …d d d…d d d…f         } t          ||
¦  «        }t          | |
¦  «        }|dz  }|dz  }||z  }t          j
        t          |dz  |
¦  «        |z
  d	¬
¦  «        }t          j
        t          | dz  |
¦  «        |z
  d	¬
¦  «        }t          || z  |
¦  «        |z
  }|||z   z  }|||z  z
  }||k     }d||<   ||         ||<   d||<   ||k     }d||<   d||<   |dk     }||         ||<   d||<   t          j
        ||¬
¦  «        }|t          j        t          j        d|dz  |z  ||z   z  z   ¦  «        g d¢¬¦  «        z  }|t          j        t          j        d||z  z   ¦  «        g d¢¬¦  «        z  }�Œî||z  S )Nr   g»½×Ùß|Û=r   r   é   r   é   r   g        )Úming      ð?)r   r   é   ©Údim)r	   r
   r   r   ÚtensorÚzerosÚsizeÚranger   r   Úclampr   Úlog10)r   r   r   r	   r
   ÚepsÚ	preds_vifÚ
target_vifÚscaleÚnÚkernelÚ	mu_targetÚmu_predsÚmu_target_sqÚmu_preds_sqÚmu_target_predsÚsigma_target_sqÚsigma_preds_sqÚsigma_target_predsr   Ú
sigma_v_sqÚmasks                         r   Ú_vif_per_channelr;   "   s   € ØŒK€EØŒ\€Fà�OŠO˜AÑÔ€EØ×Ò˜aÑ Ô €Få
Œ,�u E°&Ð
9Ñ
9Ô
9€Cå”˜j°¸fÐEÑEÔE€Jå”˜EŸJšJ q™MœM°¸vÐFÑFÔF€IÝ”˜UŸZšZ¨™]œ]°%ÀÐGÑGÔG€Jå�q‘”ð $`ñ $`ˆØ�A˜‘IÑ Ñ"ˆÝ˜˜A ™E¨°vÐ>Ñ>Ô>¸tÀTÈ1È1È1¸}ÔMˆà�1Š9ˆ9Ý˜F FÑ+Ô+¨A¨A¨A¨q¨q¨q°#°#°A°#°s°s¸°s¨NÔ;ˆFÝ˜5 &Ñ)Ô)¨!¨!¨!¨Q¨Q¨Q°°°!°°S°S°q°S¨.Ô9ˆEå˜6 6Ñ*Ô*ˆ	Ý˜% Ñ(Ô(ˆØ  !‘|ˆØ ‘kˆØ# hÑ.ˆåœ+¥f¨V°Q©Y¸Ñ&?Ô&?À,Ñ&NÐTWÐXÑXÔXˆÝœ¥V¨E°1©H°fÑ%=Ô%=ÀÑ%KÐQTÐUÑUÔUˆÝ# F¨U¡N°FÑ;Ô;¸oÑMÐà /°CÑ"7Ñ8ˆØ# aÐ*<Ñ&<Ñ<ˆ
à Ò$ˆØˆˆ$‰Ø)¨$Ô/ˆ
�4ÑØ !ˆ˜Ñà Ò#ˆØˆˆ$‰Øˆ
�4Ñà�1ŠuˆØ)¨$Ô/ˆ
�4ÑØˆˆ$‰Ý”[ °Ð5Ñ5Ô5ˆ
à•U”Y�uœ{¨3°!°S±&¸OÑ1KÈzÐ\fÑOfÑ1gÑ+gÑhÔhÐnwÐnwÐnwÐxÑxÔxÑxˆ	Ø•e”i¥¤¨C°/ÀJÑ2NÑ,NÑ OÔ OÐU^ÐU^ÐU^Ð_Ñ_Ô_Ñ_ˆ
‰
à�zÑ!Ð!r   r   ÚmeanÚ	reduction©r<   Únonec                 ó’  ‡ ‡‡— ‰                       d¦  «        dk     s‰                       d¦  «        dk     r<t          d‰                       d¦  «        › d‰                       d¦  «        › d�¦  «        ‚‰                      d¦  «        dk     s‰                      d¦  «        dk     r<t          d‰                      d¦  «        › d‰                      d¦  «        › d�¦  «        ‚‰ j        ‰j        k    r t          d‰ j        › d	‰j        › d
�¦  «        ‚|dvrt          d|› �¦  «        ‚ˆ ˆˆfd„t          ‰                       d¦  «        ¦  «        D ¦   «         }t	          ‰                       d¦  «        dk    r)t          j        |d¬¦  «                             d¦  «        n|d         ¦  «        }|dk    r|                     ¦   «         S |S )a  Compute Pixel-Based Visual Information Fidelity (VIF-P).

    VIF is a full-reference metric that measures the amount of visual information
    preserved in a distorted image compared to the reference image.

    Args:
        preds: Predicted images of shape (N, C, H, W). Height and width must be at least 41.
        target: Ground truth images of shape (N, C, H, W). Must match preds in shape.
        sigma_n_sq: Variance of the visual noise. Default: 2.0.
        reduction: Method for reducing the metric across the batch.
            - "mean": Return a tensor average over the batch.
            - "none": Return a VIF score for each sample as a 1D tensor of shape (N,).

    Returns:
        torch.Tensor: VIF score(s). The shape depends on the `reduction` argument:
            - If ``reduction="mean"``, returns a scalar tensor.
            - If ``reduction="none"``, returns a tensor of shape ``(N,)``.

    Raises:
        ValueError: If input dimensions are smaller than ``41x41``.
        ValueError: If ``preds`` and ``target`` shapes don't match.
        ValueError: If ``reduction`` is not ``"mean"`` or ``"none"``.

    Example:
        >>> from torchmetrics.functional.image import visual_information_fidelity
        >>> preds = torch.randn(4, 3, 41, 41, generator=torch.Generator().manual_seed(42))
        >>> target = torch.randn(4, 3, 41, 41, generator=torch.Generator().manual_seed(43))
        >>> visual_information_fidelity(preds, target, reduction="none")
        tensor([0.0040, 0.0049, 0.0017, 0.0039])

    éÿÿÿÿé)   éþÿÿÿz8Invalid size of preds. Expected at least 41x41, but got Úxú!z9Invalid size of target. Expected at least 41x41, but got z7`preds` and `target` must have the same shape, but got z vs ú.r>   z7Argument `reduction` must be 'mean' or 'none', but got c                 ór   •— g | ]3}t          ‰d d …|d d …d d …f         ‰d d …|d d …d d …f         ‰¦  «        ‘Œ4S )N)r;   )Ú.0Úir   r   r   s     €€€r   ú
<listcomp>z/visual_information_fidelity.<locals>.<listcomp>�   se   ø€ ð ð ð ØPQÕ˜˜q˜q˜q ! Q Q Q¨¨¨˜zÔ*¨F°1°1°1°a¸¸¸¸A¸A¸A°:Ô,>À
ÑKÔKðð ð r   r   r   r#   r<   )r'   Ú
ValueErrorÚshaper(   r   r   Ústackr<   )r   r   r   r=   Úper_channel_scoresÚvif_per_samples   ```   r   Úvisual_information_fidelityrP   Y   s   øøø€ ðR ‡z‚z�"�~„~˜ÒÐ˜eŸjšj¨™nœn¨rÒ1Ð1ÝÐvÐTY×T^ÒT^Ð_aÑTbÔTbÐvÐvÐej×eoÒeoÐprÑesÔesÐvÐvÐvÑwÔwÐwà‡{‚{�2�„˜ÒÐ˜vŸ{š{¨2™œ°Ò3Ð3ÝØlÈÏÊÐTVÉÌÐlÐlÐZ`×ZeÒZeÐfhÑZiÔZiÐlÐlÐlñ
ô 
ð 	
ð „{�f”lÒ"Ð"ÝÐsÐSXÔS^ÐsÐsÐdjÔdpÐsÐsÐsÑtÔtÐtàÐ(Ð(Ð(ÝÐ^ÐS\Ð^Ð^Ñ_Ô_Ð_ðð ð ð ð ð ÝUZÐ[`×[eÒ[eÐfgÑ[hÔ[hÑUiÔUiðñ ô Ðõ "Ø:?¿*º*ÀQ¹-¼-È!Ò:KÐ:K�ŒÐ&¨AÐ.Ñ.Ô.×3Ò3°AÑ6Ô6Ð6ÐQcÐdeÔQfñô €Nð �FÒÐØ×"Ò"Ñ$Ô$Ð$ØÐr   )r   r<   )r   r   Útorch.nn.functionalr   Útyping_extensionsr   Útorchmetrics.utilities.datar   Úfloatr	   r
   r   r;   rP   © r   r   ú<module>rV      s/  ðð €€€Ø Ð Ð Ð Ð Ð Ø &Ð &Ð &Ð &Ð &Ð &Ø %Ð %Ð %Ð %Ð %Ð %à 4Ð 4Ð 4Ð 4Ð 4Ð 4ð	�eð 	 Eð 	°%´+ð 	ÀuÄ|ð 	ÐX^ð 	ð 	ð 	ð 	ð4"˜Fð 4"¨Fð 4"Àð 4"È&ð 4"ð 4"ð 4"ð 4"ðt Ø)/ð	Að AØðAàðAð ðAð �~Ô&ð	Að
 ðAð Að Að Að Að Ar   