§
    ’©ºjè  ã            
       óŒ   — d dl mZ d dlZd dlmZ d dlmZ dedededefd	„Zdeded
ededef
d„Z	ddeded
ee         defd„Z
dS )é    )ÚOptionalN)ÚTensor)Ú"_check_retrieval_functional_inputsÚtargetÚpredsÚdiscount_cumsumÚreturnc                 óä  — t          j        | dd¬¦  «        \  }}}t          j        |t           j        ¬¦  «        }|                     d||                      |j        ¬¦  «        ¦  «         ||z  }|                     d¬¦  «        dz
  }t          j        |t           j        ¬¦  «        }||d                  |d<   ||                              ¦   «         |dd…<   ||z   	                    ¦   «         S )aI  Translated version of sklearns `_tie_average_dcg` function.

    Args:
        target: ground truth about each document relevance.
        preds: estimated probabilities of each document to be relevant.
        discount_cumsum: cumulative sum of the discount.

    Returns:
        The cumulative gain of the tied elements.

    T)Úreturn_inverseÚreturn_counts)Údtyper   ©Údimé   N)
ÚtorchÚuniqueÚ
zeros_likeÚfloat32Úscatter_add_Útor   ÚcumsumÚdiffÚsum)	r   r   r   Ú_ÚinvÚcountsÚrankedÚgroupsÚdiscount_sumss	            úv/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/torchmetrics/functional/retrieval/ndcg.pyÚ_tie_average_dcgr!      sâ   € õ ”\ 5 &¸ÈTÐRÑRÔR�N€A€sˆFÝÔ˜f­E¬MÐ:Ñ:Ô:€FØ
×Ò˜˜3 §	¢	°´ 	Ñ =Ô =Ñ>Ô>Ð>Ø�f‰_€FØ�]Š]˜qˆ]Ñ!Ô! AÑ%€FÝÔ$ Vµ5´=ÐAÑAÔA€MØ& v¨a¤yÔ1€M�!ÑØ'¨Ô/×4Ò4Ñ6Ô6€M�!�"�"ÑØ�]Ñ"×'Ò'Ñ)Ô)Ð)ó    Útop_kÚignore_tiesc                 óN  — dt          j        t          j        | j        d         | j        ¬¦  «        dz   ¦  «        z  }d||d…<   |r6|                     d¬¦  «        }| |         }||z                       ¦   «         }n'|                     d¬	¦  «        }t          | ||¦  «        }|S )
ay  Translated version of sklearns `_dcg_sample_scores` function.

    Args:
        target: ground truth about each document relevance.
        preds: estimated probabilities of each document to be relevant.
        top_k: consider only the top k elements
        ignore_ties: If True, ties are ignored. If False, ties are averaged.

    Returns:
        The cumulative gain

    g      ð?éÿÿÿÿ)Údeviceg       @g        NT)Ú
descendingr   )	r   Úlog2ÚarangeÚshaper'   Úargsortr   r   r!   )	r   r   r#   r$   ÚdiscountÚrankingr   Úcumulative_gainr   s	            r    Ú_dcg_sample_scoresr0   -   s«   € ð •e”j¥¤¨f¬l¸2Ô.>ÀvÄ}Ð!UÑ!UÔ!UÐX[Ñ![Ñ\Ô\Ñ]€HØ€HˆUˆVˆVÑàð KØ—-’-¨4�-Ñ0Ô0ˆØ˜”ˆØ# fÑ,×1Ò1Ñ3Ô3ˆˆà"Ÿ/š/¨b˜/Ñ1Ô1ˆÝ*¨6°5¸/ÑJÔJˆØÐr"   c                 ó\  — t          | |d¬¦  «        \  } }|€| j        d         n|}t          |t          ¦  «        r|dk    st	          d¦  «        ‚t          || |d¬¦  «        }t          |||d¬¦  «        }|dk    }d||<   || xx         ||          z  cc<   |                     ¦   «         S )	a  Compute `Normalized Discounted Cumulative Gain`_ (for information retrieval).

    ``preds`` and ``target`` should be of the same shape and live on the same device.
    ``target`` must be either `bool` or `integers` and ``preds`` must be ``float``,
    otherwise an error is raised.

    Args:
        preds: estimated probabilities of each document to be relevant.
        target: ground truth about each document relevance.
        top_k: consider only the top k elements (default: ``None``, which considers them all)

    Return:
        A single-value tensor with the nDCG of the predictions ``preds`` w.r.t. the labels ``target``.

    Raises:
        ValueError:
            If ``top_k`` parameter is not `None` or an integer larger than 0

    Example:
        >>> from torchmetrics.functional.retrieval import retrieval_normalized_dcg
        >>> preds = torch.tensor([.1, .2, .3, 4, 70])
        >>> target = torch.tensor([10, 0, 0, 1, 5])
        >>> retrieval_normalized_dcg(preds, target)
        tensor(0.6957)

    T)Úallow_non_binary_targetNr&   r   z,`top_k` has to be a positive integer or NoneF)r$   )r   r+   Ú
isinstanceÚintÚ
ValueErrorr0   Úmean)r   r   r#   ÚgainÚnormalized_gainÚall_irrelevants         r    Úretrieval_normalized_dcgr:   G   sÏ   € õ6 7°u¸fÐ^bÐcÑcÔc�M€Eˆ6à$˜}ˆEŒK˜ŒOˆO°%€Eå�u�cÑ"Ô"ð I u¨q¢y yÝÐGÑHÔHÐHå˜f e¨UÀÐFÑFÔF€DÝ(¨°¸ÈDÐQÑQÔQ€Oð %¨Ò)€NØ€DˆÑØˆ.ˆÐÐÔ˜_¨n¨_Ô=Ñ=ÐÐÑà�9Š9‰;Œ;Ðr"   )N)Útypingr   r   r   Útorchmetrics.utilities.checksr   r!   r4   Úboolr0   r:   © r"   r    ú<module>r?      só   ðð Ð Ð Ð Ð Ð à €€€Ø Ð Ð Ð Ð Ð à LÐ LÐ LÐ LÐ LÐ Lð*˜Vð *¨Fð *ÀVð *ÐPVð *ð *ð *ð *ð.˜vð ¨fð ¸Sð Ètð ÐX^ð ð ð ð ð4*ð * Fð *°Fð *À8ÈCÄ=ð *Ð\bð *ð *ð *ð *ð *ð *r"   