o
    £ôj  ã                   @   sœ   d dl mZ d dlmZmZmZmZ d dlmZm	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 d	d
giZesDdgZG dd	„ d	eƒZdS )é    )ÚSequence)ÚAnyÚCallableÚOptionalÚUnion)ÚTensorÚtensor)ÚLiteral)Úpermutation_invariant_training)ÚMetric)Ú_MATPLOTLIB_AVAILABLE)Ú_AX_TYPEÚ_PLOT_OUT_TYPEÚPermutationInvariantTrainingÚpitú!PermutationInvariantTraining.plotc                       sÜ   e Zd ZU dZdZeed< dZeed< eed< eed< dZ	e
e ed	< dZe
e ed
< 		d deded ded deddf
‡ fdd„Zdededdfdd„Zdefdd„Zd!deeee df de
e defdd„Z‡  ZS )"r   aQ  Calculate `Permutation invariant training`_ (PIT).

    This metric can evaluate models for speaker independent multi-talker speech separation in a permutation
    invariant way.

    As input to ``forward`` and ``update`` the metric accepts the following input

    - ``preds`` (:class:`~torch.Tensor`): float tensor with shape ``(batch_size,num_speakers,...)``
    - ``target`` (:class:`~torch.Tensor`): float tensor with shape ``(batch_size,num_speakers,...)``

    As output of `forward` and `compute` the metric returns the following output

    - ``pesq`` (:class:`~torch.Tensor`): float scalar tensor with average PESQ value over samples

    Args:
        metric_func:
            a metric function accept a batch of target and estimate.

            if `mode`==`'speaker-wise'`, then ``metric_func(preds[:, i, ...], target[:, j, ...])`` is called
            and expected to return a batch of metric tensors ``(batch,)``;

            if `mode`==`'permutation-wise'`, then ``metric_func(preds[:, p, ...], target[:, :, ...])`` is called,
            where `p` is one possible permutation, e.g. [0,1] or [1,0] for 2-speaker case, and expected to return
            a batch of metric tensors ``(batch,)``;
        mode:
            can be `'speaker-wise'` or `'permutation-wise'`.
        eval_func:
            the function to find the best permutation, can be 'min' or 'max', i.e. the smaller the better
            or the larger the better.
        kwargs: Additional keyword arguments for either the ``metric_func`` or distributed communication,
            see :ref:`Metric kwargs` for more info.

    Example:
        >>> from torch import randn
        >>> from torchmetrics.audio import PermutationInvariantTraining
        >>> from torchmetrics.functional.audio import scale_invariant_signal_noise_ratio
        >>> preds = randn(3, 2, 5) # [batch, spk, time]
        >>> target = randn(3, 2, 5) # [batch, spk, time]
        >>> pit = PermutationInvariantTraining(scale_invariant_signal_noise_ratio,
        ...     mode="speaker-wise", eval_func="max")
        >>> pit(preds, target)
        tensor(-2.1065)

    FÚfull_state_updateTÚis_differentiableÚsum_pit_metricÚtotalNÚplot_lower_boundÚplot_upper_boundúspeaker-wiseÚmaxÚmetric_funcÚmode)r   zpermutation-wiseÚ	eval_func)r   ÚminÚkwargsÚreturnc                    sz   |  dd¡|  dd ¡|  dd ¡dœ}tƒ jdi |¤Ž || _|| _|| _|| _| jdtdƒdd	� | jd
tdƒdd	� d S )NÚdist_sync_on_stepFÚprocess_groupÚdist_sync_fn)r    r!   r"   r   g        Úsum)ÚdefaultÚdist_reduce_fxr   r   © )	ÚpopÚsuperÚ__init__r   r   r   r   Ú	add_stater   )Úselfr   r   r   r   Zbase_kwargs©Ú	__class__r&   úi/var/www/html/academy/tools/voices/clone/cosy-venv/lib/python3.10/site-packages/torchmetrics/audio/pit.pyr)   T   s   


ýz%PermutationInvariantTraining.__init__ÚpredsÚtargetc                 C   sL   t ||| j| j| jfi | j¤Žd }|  j| ¡ 7  _|  j| ¡ 7  _dS )z*Update state with predictions and targets.r   N)	r
   r   r   r   r   r   r#   r   Únumel)r+   r/   r0   Z
pit_metricr&   r&   r.   Úupdatei   s   ÿÿþz#PermutationInvariantTraining.updatec                 C   s   | j | j S )zCompute metric.)r   r   )r+   r&   r&   r.   Úcomputer   s   z$PermutationInvariantTraining.computeÚvalÚaxc                 C   s   |   ||¡S )aŒ  Plot a single or multiple values from the metric.

        Args:
            val: Either a single result from calling `metric.forward` or `metric.compute` or a list of these results.
                If no value is provided, will automatically call `metric.compute` and plot that result.
            ax: An matplotlib axis object. If provided will add plot to that axis

        Returns:
            Figure and Axes object

        Raises:
            ModuleNotFoundError:
                If `matplotlib` is not installed

        .. plot::
            :scale: 75

            >>> # Example plotting a single value
            >>> import torch
            >>> from torchmetrics.audio import PermutationInvariantTraining
            >>> from torchmetrics.functional.audio import scale_invariant_signal_noise_ratio
            >>> preds = torch.randn(3, 2, 5) # [batch, spk, time]
            >>> target = torch.randn(3, 2, 5) # [batch, spk, time]
            >>> metric = PermutationInvariantTraining(scale_invariant_signal_noise_ratio,
            ...     mode="speaker-wise", eval_func="max")
            >>> metric.update(preds, target)
            >>> fig_, ax_ = metric.plot()

        .. plot::
            :scale: 75

            >>> # Example plotting multiple values
            >>> import torch
            >>> from torchmetrics.audio import PermutationInvariantTraining
            >>> from torchmetrics.functional.audio import scale_invariant_signal_noise_ratio
            >>> preds = torch.randn(3, 2, 5) # [batch, spk, time]
            >>> target = torch.randn(3, 2, 5) # [batch, spk, time]
            >>> metric = PermutationInvariantTraining(scale_invariant_signal_noise_ratio,
            ...     mode="speaker-wise", eval_func="max")
            >>> values = [ ]
            >>> for _ in range(10):
            ...     values.append(metric(preds, target))
            >>> fig_, ax_ = metric.plot(values)

        )Ú_plot)r+   r4   r5   r&   r&   r.   Úplotv   s   .r   )r   r   )NN)Ú__name__Ú
__module__Ú__qualname__Ú__doc__r   ÚboolÚ__annotations__r   r   r   r   Úfloatr   r   r	   r   r)   r2   r3   r   r   r   r   r7   Ú__classcell__r&   r&   r,   r.   r      s0   
 -üþýüûú	2N)Úcollections.abcr   Útypingr   r   r   r   Útorchr   r   Útyping_extensionsr	   Ú!torchmetrics.functional.audio.pitr
   Útorchmetrics.metricr   Útorchmetrics.utilities.importsr   Útorchmetrics.utilities.plotr   r   Ú__doctest_requires__Ú__doctest_skip__r   r&   r&   r&   r.   Ú<module>   s   
