:jddlmZddlmZmZmZmZddlmZm Z ddl m Z ddl m Z mZddlmZddlmZddlmZdd lmZdd lmZmZesd gZGd d eZy))Sequence)AnyListOptionalUnion)Tensortensor)Literal) _uqi_compute _uqi_update)Metric)rank_zero_warn) dim_zero_cat)_MATPLOTLIB_AVAILABLE)_AX_TYPE_PLOT_OUT_TYPEzUniversalImageQualityIndex.plotc "eZdZUdZdZeed<dZeed<dZeed<dZ e ed<d Z e ed <e e ed <e e ed <e ed <e ed< ddeedee deddeddf fd Zd e d e ddfdZde fdZ ddeee ee fdeedefdZxZS)UniversalImageQualityIndexaCompute Universal Image Quality Index (UniversalImageQualityIndex_). 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 - ``uiqi`` (:class:`~torch.Tensor`): if ``reduction!='none'`` returns float scalar tensor with average UIQI value over sample else returns tensor of shape ``(N,)`` with UIQI values per sample Args: kernel_size: size of the gaussian kernel sigma: Standard deviation of the gaussian kernel 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 kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Return: Tensor with UniversalImageQualityIndex score Example: >>> import torch >>> from torchmetrics.image import UniversalImageQualityIndex >>> preds = torch.rand([16, 1, 16, 16]) >>> target = preds * 0.75 >>> uqi = UniversalImageQualityIndex() >>> uqi(preds, target) tensor(0.9216) Tis_differentiablehigher_is_betterFfull_state_updateplot_lower_boundg?plot_upper_boundpredstargetsum_uqinumel kernel_sizesigma reductionelementwise_meansumnoneNkwargsreturnNc ^t|di||dvrtd|d||dk(r4td|j dgd|j d gdn:|j d t d d |j dt dd ||_||_||_y)Nr"zThe `reduction` zI is not valid. Valid options are `elementwise_mean`, `sum`, `none`, None.r%zMetric `UniversalImageQualityIndex` will save all targets and predictions in the buffer when using`reduction=None` or `reduction='none'. For large datasets, this may lead to a large memory footprint.rcat)defaultdist_reduce_fxrrrr$)r+rr) super__init__ ValueErrorr add_stater rr r!)selfrr r!r& __class__s k/home/mcse/projects/srt_converter/srt-converter-venv/lib/python3.12/site-packages/torchmetrics/image/uqi.pyr.z#UniversalImageQualityIndex.__init__Ps "6" E E"9+-vw    V 3 x  NN7BuN E NN8RN F NN9fSk%N H NN7F1IeN D& "ct||\}}|j|jdk(r7|jj||jj|yt |||j |jd}|xj|z c_|j}|xj|d|dz|d|j dz dzz|d|j dz dzzz c_ y) z*Update state with predictions and targets.Nr%r$)r!r) r r!rappendrr rr rshaper)r1rr uqi_scorepss r3updatez!UniversalImageQualityIndex.updatejs#E62 v >> !T^^v%= JJ  e $ KK  v &$UFD4D4Ddjj\abI LLI %LB JJ"Q%"Q%-2a543C3CA3F+F+JKrRSuW[WgWghiWjOjmnOno oJr4cN|jdk(s |jWt|j}t|j}t |||j |j |jS|jdk(r|j|jz S|jS)z&Compute explained variance over state.r%r#) r!rrrr rr rr)r1rrs r3computez"UniversalImageQualityIndex.computevs >>V #t~~'= ,E!$++.Fvt/?/?T^^\ \,0NN>P,Pt||djj(bVZVbVbbr4valaxc&|j||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 UniversalImageQualityIndex >>> preds = torch.rand([16, 1, 16, 16]) >>> target = preds * 0.75 >>> metric = UniversalImageQualityIndex() >>> metric.update(preds, target) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> import torch >>> from torchmetrics.image import UniversalImageQualityIndex >>> preds = torch.rand([16, 1, 16, 16]) >>> target = preds * 0.75 >>> metric = UniversalImageQualityIndex() >>> values = [ ] >>> for _ in range(10): ... values.append(metric(preds, target)) >>> fig_, ax_ = metric.plot(values) )_plot)r1r@rAs r3plotzUniversalImageQualityIndex.plot~sXzz#r""r4)) rE)?rFr#)NN)__name__ __module__ __qualname____doc__rbool__annotations__rrrfloatrrrrintr rr.r=r?rrrrrD __classcell__)r2s@r3rrs#J#t"!d!#t#!e!!e! < L O M&.!+FX #c]##BC #  #  #4 pF pF pt pcc_c,#E&(6*:":;<,#IQRZI[,# ,#r4rN)collections.abcrtypingrrrrtorchrr typing_extensionsr !torchmetrics.functional.image.uqir r torchmetrics.metricr torchmetrics.utilitiesrtorchmetrics.utilities.datartorchmetrics.utilities.importsrtorchmetrics.utilities.plotrr__doctest_skip__rr,r4r3r[sB%-- %G&14@@9:K#K#r4