§
    ’©ºj·  ã                   óŠ   — d dl Z d dl mZ d dlmZ d dlmZ ddededed	eeef         fd
„Zdeded	efd„Z	ddededed	efd„Z
dS )é    N)ÚTensor)Ú_check_same_shape)Ú_safe_xlogyç        ÚpredsÚtargetsÚpowerÚreturnc                 ó²  — t          | |¦  «         t          j        | j        | j        ¬¦  «        }d|cxk     rdk     rn nt          d|› d�¦  «        ‚|dk    rt          j        || z
  d¬¦  «        }�n|dk    rat          j        | dk    ¦  «        st          j        |dk     ¦  «        rt          d|› d	�¦  «        ‚dt          ||| z  ¦  «        | z   |z
  z  }�n¸|dk    rht          j        | dk    ¦  «        st          j        |dk    ¦  «        rt          d|› d
�¦  «        ‚dt          j	        | |z  ¦  «        || z  z   dz
  z  }�nJ|dk     r,t          j        | dk    ¦  «        rt          d|› d�¦  «        ‚n—d|cxk     rdk     rGn nDt          j        | dk    ¦  «        st          j        |dk     ¦  «        rt          d|› d�¦  «        ‚nCt          j        | dk    ¦  «        st          j        |dk    ¦  «        rt          d|› d
�¦  «        ‚t          j        t          j
        ||¦  «        d|z
  ¦  «        d|z
  d|z
  z  z  }|t          j        | d|z
  ¦  «        z  d|z
  z  }t          j        | d|z
  ¦  «        d|z
  z  }d||z
  |z   z  }t          j        |¦  «        }t          j        t          j        |¦  «        | j        ¬¦  «        }	||	fS )aî  Update and returns variables required to compute Deviance Score for the given power.

    Check for same shape of input tensors.

    Args:
        preds: Predicted tensor
        targets: Ground truth tensor
        power: see :func:`tweedie_deviance_score`

    Example:
        >>> targets = torch.tensor([1.0, 2.0, 3.0, 4.0])
        >>> preds = torch.tensor([4.0, 3.0, 2.0, 1.0])
        >>> _tweedie_deviance_score_update(preds, targets, power=2)
        (tensor(4.8333), tensor(4))

    )Údevicer   é   z(Deviance Score is not defined for power=ú.é   )Úexponentz
For power=zG, 'preds' has to be strictly positive and 'targets' cannot be negative.z:, both 'preds' and 'targets' have to be strictly positive.z&, 'preds' has to be strictly positive.zG, 'targets' has to be strictly positive and 'preds' cannot be negative.)r   ÚtorchÚzerosÚshaper   Ú
ValueErrorÚpowÚanyr   ÚlogÚmaxÚsumÚtensorÚnumel)
r   r   r	   Úzero_tensorÚdeviance_scoreÚterm_1Úterm_2Úterm_3Úsum_deviance_scoreÚnum_observationss
             úƒ/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/torchmetrics/functional/regression/tweedie_deviance.pyÚ_tweedie_deviance_score_updater$      s-  € õ" �e˜WÑ%Ô%Ð%å”+˜eœk°%´,Ð?Ñ?Ô?€Kàˆ5€}€}‚}€}�1‚}€}€}€}€}ÝÐLÀEÐLÐLÐLÑMÔMÐMà�‚z€zÝœ 7¨U¡?¸QÐ?Ñ?Ô?ˆ‰Ø	�!ŠˆåŒ9�U˜a’ZÑ Ô ð 	¥E¤I¨g¸ªkÑ$:Ô$:ð 	ÝØk˜UÐkÐkÐkñô ð ð �k¨'°7¸U±?ÑCÔCÀeÑKÈgÑUÑVˆ‰Ø	�!ŠˆåŒ9�U˜a’ZÑ Ô ð 	m¥E¤I¨g¸ªlÑ$;Ô$;ð 	mÝÐk¨%ÐkÐkÐkÑlÔlÐlà�eœi¨°©Ñ8Ô8¸GÀe¹OÑLÈqÑPÑQˆ‰à�1Š9ˆ9ÝŒy˜ !šÑ$Ô$ð ]Ý Ð![¨eÐ![Ð![Ð![Ñ\Ô\Ð\ð]à�ˆ]ˆ]Š]ˆ]˜Š]ˆ]ˆ]ˆ]ˆ]ÝŒy˜ !šÑ$Ô$ð ­¬	°'¸A²+Ñ(>Ô(>ð Ý Øo ÐoÐoÐoñô ð ðõ
 Œy˜ !šÑ$Ô$ð q­¬	°'¸Q²,Ñ(?Ô(?ð qÝ Ð!o¨eÐ!oÐ!oÐ!oÑpÔpÐpå”�5œ9 W¨kÑ:Ô:¸AÀ¹IÑFÔFÈ1ÈuÉ9ÐYZÐ]bÑYbÑJcÑdˆØ�5œ9 U¨A°©IÑ6Ô6Ñ6¸!¸e¹)ÑDˆÝ”˜5 ! e¡)Ñ,Ô,°°E±	Ñ:ˆØ˜f v™o°Ñ6Ñ7ˆåœ >Ñ2Ô2ÐÝ”|¥E¤K°Ñ$?Ô$?ÈÌÐUÑUÔUÐàÐ/Ð/Ð/ó    r!   r"   c                 ó   — | |z  S )a  Compute Deviance Score.

    Args:
        sum_deviance_score: Sum of deviance scores accumulated until now.
        num_observations: Number of observations encountered until now.

    Example:
        >>> targets = torch.tensor([1.0, 2.0, 3.0, 4.0])
        >>> preds = torch.tensor([4.0, 3.0, 2.0, 1.0])
        >>> sum_deviance_score, num_observations = _tweedie_deviance_score_update(preds, targets, power=2)
        >>> _tweedie_deviance_score_compute(sum_deviance_score, num_observations)
        tensor(1.2083)

    © )r!   r"   s     r#   Ú_tweedie_deviance_score_computer(   V   s   € ð Ð 0Ñ0Ð0r%   c                 óL   — t          | ||¬¦  «        \  }}t          ||¦  «        S )a´  Compute the `Tweedie Deviance Score`_.

    .. math::
        deviance\_score(\hat{y},y) =
        \begin{cases}
        (\hat{y} - y)^2, & \text{for }p=0\\
        2 * (y * log(\frac{y}{\hat{y}}) + \hat{y} - y),  & \text{for }p=1\\
        2 * (log(\frac{\hat{y}}{y}) + \frac{y}{\hat{y}} - 1),  & \text{for }p=2\\
        2 * (\frac{(max(y,0))^{2 - p}}{(1 - p)(2 - p)} - \frac{y(\hat{y})^{1 - p}}{1 - p} + \frac{(
            \hat{y})^{2 - p}}{2 - p}), & \text{otherwise}
        \end{cases}

    where :math:`y` is a tensor of targets values, :math:`\hat{y}` is a tensor of predictions, and
    :math:`p` is the `power`.

    Args:
        preds: Predicted tensor with shape ``(N,...)``
        targets: Ground truth tensor with shape ``(N,...)``
        power:
            - `power < 0` : Extreme stable distribution. (Requires: preds > 0.)
            - `power = 0` : Normal distribution. (Requires: targets and preds can be any real numbers.)
            - `power = 1` : Poisson distribution. (Requires: targets >= 0 and y_pred > 0.)
            - `1 < p < 2` : Compound Poisson distribution. (Requires: targets >= 0 and preds > 0.)
            - `power = 2` : Gamma distribution. (Requires: targets > 0 and preds > 0.)
            - `power = 3` : Inverse Gaussian distribution. (Requires: targets > 0 and preds > 0.)
            - `otherwise` : Positive stable distribution. (Requires: targets > 0 and preds > 0.)

    Example:
        >>> from torchmetrics.functional.regression import tweedie_deviance_score
        >>> targets = torch.tensor([1.0, 2.0, 3.0, 4.0])
        >>> preds = torch.tensor([4.0, 3.0, 2.0, 1.0])
        >>> tweedie_deviance_score(preds, targets, power=2)
        tensor(1.2083)

    )r	   )r$   r(   )r   r   r	   r!   r"   s        r#   Útweedie_deviance_scorer*   h   s5   € õH ,JÈ%ÐQXÐ`eÐ+fÑ+fÔ+fÑ(ÐÐ(Ý*Ð+=Ð?OÑPÔPÐPr%   )r   )r   r   Útorchmetrics.utilities.checksr   Útorchmetrics.utilities.computer   ÚfloatÚtupler$   r(   r*   r'   r%   r#   ú<module>r/      sÿ   ðð €€€Ø Ð Ð Ð Ð Ð à ;Ð ;Ð ;Ð ;Ð ;Ð ;Ø 6Ð 6Ð 6Ð 6Ð 6Ð 6ð=0ð =0¨&ð =0¸6ð =0È%ð =0ÐZ_Ð`fÐhnÐ`nÔZoð =0ð =0ð =0ð =0ð@1¸ð 1ÐRXð 1Ð]cð 1ð 1ð 1ð 1ð$%Qð %Q &ð %Q°6ð %QÀ%ð %QÐRXð %Qð %Qð %Qð %Qð %Qð %Qr%   