o .wi%@sddlmZddlmZddlmZmZmZddlZddlm Z m Z ddl m Z ddl mZmZddlmZdd lmZdd lmZdd lmZmZesOd gZGd ddeZdS))Sequence)partial)AnyOptionalUnionN)Tensortensor)Literal) _psnr_compute _psnr_update)Metric)rank_zero_warn)_MATPLOTLIB_AVAILABLE)_AX_TYPE_PLOT_OUT_TYPEPeakSignalNoiseRatio.plotcseZdZUdZdZeed<dZeed<dZeed<dZ e ed<e ed <e ed < d"de e e ee e ffde dedde e eeedffdedd f fdd Zde de dd fddZde fddZ d#de e e ee fde edefd d!ZZS)$PeakSignalNoiseRatioa `Compute Peak Signal-to-Noise Ratio`_ (PSNR). .. math:: \text{PSNR}(I, J) = 10 * \log_{10} \left(\frac{\max(I)^2}{\text{MSE}(I, J)}\right) Where :math:`\text{MSE}` denotes the `mean-squared-error`_ function. As input to ``forward`` and ``update`` the metric accepts the following input - ``preds`` (:class:`~torch.Tensor`): Predictions from model of shape ``(N,C,H,W)`` - ``target`` (:class:`~torch.Tensor`): Ground truth values of shape ``(N,C,H,W)`` As output of `forward` and `compute` the metric returns the following output - ``psnr`` (:class:`~torch.Tensor`): if ``reduction!='none'`` returns float scalar tensor with average PSNR value over sample else returns tensor of shape ``(N,)`` with PSNR values per sample Args: data_range: the range of the data. If None, it is determined from the data (max - min). If a tuple is provided then the range is calculated as the difference and input is clamped between the values. The ``data_range`` must be given when ``dim`` is not None. base: a base of a logarithm to use. reduction: a method to reduce metric score over labels. - ``'elementwise_mean'``: takes the mean (default) - ``'sum'``: takes the sum - ``'none'`` or ``None``: no reduction will be applied dim: Dimensions to reduce PSNR scores over, provided as either an integer or a list of integers. Default is None meaning scores will be reduced across all dimensions and all batches. kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Raises: ValueError: If ``dim`` is not ``None`` and ``data_range`` is not given. Example: >>> from torchmetrics.image import PeakSignalNoiseRatio >>> psnr = PeakSignalNoiseRatio() >>> preds = torch.tensor([[0.0, 1.0], [2.0, 3.0]]) >>> target = torch.tensor([[3.0, 2.0], [1.0, 0.0]]) >>> psnr(preds, target) tensor(2.5527) Tis_differentiablehigher_is_betterFfull_state_updateplot_lower_bound min_target max_targetN$@elementwise_mean data_rangebase reduction)rsumnoneNdim.kwargsreturnc sbtjdi||dur|dkrtd|d|dur2|jdtddd|jdtd ddn|jdgd d|jdgd dd|_|durk|durQtd d|_|jd tdtj d|jd tdtj dn/t |t r|jdt|d|d ddt tj|d |dd|_n |jdtt|dd||_||_t |trt ||_dS||_dS)NrzThe `reduction=z.` will not have any effect when `dim` is None.sum_squared_errorrr)defaultdist_reduce_fxtotalrcatz6The `data_range` must be given when `dim` is not None.rrrmean)minmax)super__init__r add_stater clamping_fn ValueErrorrtorchr+r, isinstancetuplerclampfloatrrrr!)selfrrrr!r" __class__r-T/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/torchmetrics/image/psnr.pyr/Xs,  "zPeakSignalNoiseRatio.__init__predstargetcCs*|jdur||}||}t|||jd\}}|jdurg|jdur7t||j|_t| |j |_ t |j t sGtdt|j t |jt sWtdt|j|j |7_ |j|7_dSt |j tswtdt|j t |jtstdt|j|j ||j|dS)z*Update state with predictions and targets.N)r!z:Expected `self.sum_squared_error` to be a Tensor, but got z.Expected `self.total` to be a Tensor, but got z8Expected `self.sum_squared_error` to be a list, but got z,Expected `self.total` to be a list, but got )r1r r!rr3minimumr+rmaximumr,rr4r$r TypeErrortyper'listappend)r8r<r=r$num_obsr-r-r;updates0          zPeakSignalNoiseRatio.updatecCs|jdur|jn|j|j}t|jtjr|j}nt|jtr+tdd|jD}nt dt|j tjr:|j }nt|j trLtdd|j D}nt dt ||||j |j dS)z.Compute peak signal-to-noise ratio over state.NcSg|]}|qSr-flatten.0valuer-r-r; z0PeakSignalNoiseRatio.compute..z>Expected sum_squared_error to be a Tensor or a list of TensorscSrFr-rGrIr-r-r;rLrMz2Expected total to be a Tensor or a list of Tensors)rr)rrrr4r$r3rrBr(r@r'r rr)r8rr$r'r-r-r;computes  zPeakSignalNoiseRatio.computevalaxcCs |||S)aPlot 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.image import PeakSignalNoiseRatio >>> metric = PeakSignalNoiseRatio() >>> preds = torch.tensor([[0.0, 1.0], [2.0, 3.0]]) >>> target = torch.tensor([[3.0, 2.0], [1.0, 0.0]]) >>> metric.update(preds, target) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> import torch >>> from torchmetrics.image import PeakSignalNoiseRatio >>> metric = PeakSignalNoiseRatio() >>> preds = torch.tensor([[0.0, 1.0], [2.0, 3.0]]) >>> target = torch.tensor([[3.0, 2.0], [1.0, 0.0]]) >>> values = [ ] >>> for _ in range(10): ... values.append(metric(preds, target)) >>> fig_, ax_ = metric.plot(values) )_plot)r8rOrPr-r-r;plots ,r)NrrN)NN)__name__ __module__ __qualname____doc__rbool__annotations__rrrr7rrrr5r intrr/rErNrrrrR __classcell__r-r-r9r;r sF  /   ' r)collections.abcr functoolsrtypingrrrr3rrtyping_extensionsr "torchmetrics.functional.image.psnrr r torchmetrics.metricr torchmetrics.utilitiesr torchmetrics.utilities.importsrtorchmetrics.utilities.plotrr__doctest_skip__rr-r-r-r;s