o
    yi                     @   sH   d dl mZmZmZmZmZ d dlmZ d dlm	Z	 G dd de	Z
dS )    )AnyCallableDictListOptional)Tensor)Metricc                       s   e Zd ZdZddedeee  ddf fddZde	de
eef fd	d
ZdededefddZdededdfddZde
ee	f fddZdddZdedefddZdedefddZ  ZS )ClasswiseWrapperaZ
  Wrapper class for altering the output of classification metrics that returns multiple values to include
    label information.

    Args:
        metric: base metric that should be wrapped. It is assumed that the metric outputs a single
            tensor that is split along the first dimension.
        labels: list of strings indicating the different classes.

    Example:
        >>> import torch
        >>> _ = torch.manual_seed(42)
        >>> from torchmetrics import ClasswiseWrapper
        >>> from torchmetrics.classification import MulticlassAccuracy
        >>> metric = ClasswiseWrapper(MulticlassAccuracy(num_classes=3, average=None))
        >>> preds = torch.randn(10, 3).softmax(dim=-1)
        >>> target = torch.randint(3, (10,))
        >>> metric(preds, target)  # doctest: +NORMALIZE_WHITESPACE
        {'multiclassaccuracy_0': tensor(0.5000),
        'multiclassaccuracy_1': tensor(0.7500),
        'multiclassaccuracy_2': tensor(0.)}

    Example (labels as list of strings):
        >>> import torch
        >>> from torchmetrics import ClasswiseWrapper
        >>> from torchmetrics.classification import MulticlassAccuracy
        >>> metric = ClasswiseWrapper(
        ...    MulticlassAccuracy(num_classes=3, average=None),
        ...    labels=["horse", "fish", "dog"]
        ... )
        >>> preds = torch.randn(10, 3).softmax(dim=-1)
        >>> target = torch.randint(3, (10,))
        >>> metric(preds, target)  # doctest: +NORMALIZE_WHITESPACE
        {'multiclassaccuracy_horse': tensor(0.3333),
        'multiclassaccuracy_fish': tensor(0.6667),
        'multiclassaccuracy_dog': tensor(0.)}

    Example (in metric collection):
        >>> import torch
        >>> from torchmetrics import ClasswiseWrapper, MetricCollection
        >>> from torchmetrics.classification import MulticlassAccuracy, MulticlassRecall
        >>> labels = ["horse", "fish", "dog"]
        >>> metric = MetricCollection(
        ...     {'multiclassaccuracy': ClasswiseWrapper(MulticlassAccuracy(num_classes=3, average=None), labels),
        ...     'multiclassrecall': ClasswiseWrapper(MulticlassRecall(num_classes=3, average=None), labels)}
        ... )
        >>> preds = torch.randn(10, 3).softmax(dim=-1)
        >>> target = torch.randint(3, (10,))
        >>> metric(preds, target)  # doctest: +NORMALIZE_WHITESPACE
        {'multiclassaccuracy_horse': tensor(0.),
         'multiclassaccuracy_fish': tensor(0.3333),
         'multiclassaccuracy_dog': tensor(0.4000),
         'multiclassrecall_horse': tensor(0.),
         'multiclassrecall_fish': tensor(0.3333),
         'multiclassrecall_dog': tensor(0.4000)}
    Nmetriclabelsreturnc                    sj   t    t|tstd| |d ur*t|tr#tdd |D s*td| || _|| _d| _	d S )NzNExpected argument `metric` to be an instance of `torchmetrics.Metric` but got c                 s   s    | ]}t |tV  qd S N)
isinstancestr).0lab r   S/home/ubuntu/.local/lib/python3.10/site-packages/torchmetrics/wrappers/classwise.py	<genexpr>R   s    z,ClasswiseWrapper.__init__.<locals>.<genexpr>zLExpected argument `labels` to either be `None` or a list of strings but got    )
super__init__r   r   
ValueErrorlistallr
   r   _update_count)selfr
   r   	__class__r   r   r   N   s   

$
zClasswiseWrapper.__init__xc                    sH   | j jj  | jd u r fddt|D S  fddt| j|D S )Nc                        i | ]\}}  d | |qS _r   )r   ivalnamer   r   
<dictcomp>[        z-ClasswiseWrapper._convert.<locals>.<dictcomp>c                    r    r!   r   )r   r   r$   r%   r   r   r'   \   r(   )r
   r   __name__lowerr   	enumeratezip)r   r   r   r%   r   _convertX   s   
zClasswiseWrapper._convertargskwargsc                 O   s   |  | j|i |S r   )r-   r
   r   r.   r/   r   r   r   forward^      zClasswiseWrapper.forwardc                 O   s   | j j|i | d S r   )r
   updater0   r   r   r   r3   a   r2   zClasswiseWrapper.updatec                 C   s   |  | j S r   )r-   r
   computer   r   r   r   r4   d   s   zClasswiseWrapper.computec                 C   s   | j   d S r   )r
   resetr5   r   r   r   r6   g   s   zClasswiseWrapper.resetr3   c                 C      |S zOverwrite to do nothing.r   )r   r3   r   r   r   _wrap_updatej      zClasswiseWrapper._wrap_updater4   c                 C   r7   r8   r   )r   r4   r   r   r   _wrap_computen   r:   zClasswiseWrapper._wrap_computer   )r   N)r)   
__module____qualname____doc__r   r   r   r   r   r   r   r   r-   r1   r3   r4   r6   r   r9   r;   __classcell__r   r   r   r   r	      s    $8

r	   N)typingr   r   r   r   r   torchr   torchmetricsr   r	   r   r   r   r   <module>   s   