i1ddlmZddlmZmZmZddlmZmZddl m Z m Z m Z ddl mZddlmZddlmZmZesgdZGd d eZGd d eZGd deZy))Sequence)AnyOptionalUnion)Tensortensor)*complex_scale_invariant_signal_noise_ratio"scale_invariant_signal_noise_ratiosignal_noise_ratio)Metric)_MATPLOTLIB_AVAILABLE)_AX_TYPE_PLOT_OUT_TYPE)zSignalNoiseRatio.plotz#ScaleInvariantSignalNoiseRatio.plotz*ComplexScaleInvariantSignalNoiseRatio.plotceZdZUdZdZeed<dZeed<dZeed<e ed<e ed<d Z e e ed <d Z e e ed < dd ed edd ffd Zde de dd fdZde fdZ dde ee ee fde edefdZxZS)SignalNoiseRatioa3Calculate `Signal-to-noise ratio`_ (SNR_) meric for evaluating quality of audio. .. math:: \text{SNR} = \frac{P_{signal}}{P_{noise}} where :math:`P` denotes the power of each signal. The SNR metric compares the level of the desired signal to the level of background noise. Therefore, a high value of SNR means that the audio is clear. As input to `forward` and `update` the metric accepts the following input - ``preds`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` - ``target`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` As output of `forward` and `compute` the metric returns the following output - ``snr`` (:class:`~torch.Tensor`): float scalar tensor with average SNR value over samples Args: zero_mean: if to zero mean target and preds or not kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Raises: TypeError: if target and preds have a different shape Example: >>> from torch import tensor >>> from torchmetrics.audio import SignalNoiseRatio >>> target = tensor([3.0, -0.5, 2.0, 7.0]) >>> preds = tensor([2.5, 0.0, 2.0, 8.0]) >>> snr = SignalNoiseRatio() >>> snr(preds, target) tensor(16.1805) Ffull_state_updateTis_differentiablehigher_is_bettersum_snrtotalNplot_lower_boundplot_upper_bound zero_meankwargsreturnc t|di|||_|jdt dd|jdt ddy)Nrsumdefaultdist_reduce_fxrr)super__init__r add_staterselfrr __class__s k/Volumes/fast/ai/experiments/voice-extract-mac/.venv/lib/python3.12/site-packages/torchmetrics/audio/snr.pyr$zSignalNoiseRatio.__init__QsH "6"" y&+eL wq %Hpredstargetct|||j}|xj|jz c_|xj|j z c_y*Update state with predictions and targets.)r+r,rN)r rrrrnumel)r'r+r, snr_batchs r)updatezSignalNoiseRatio.update\s=&U6T^^\  '  ioo'' r*c4|j|jz SzCompute metric.)rrr's r)computezSignalNoiseRatio.computecs||djj((r*valaxc&|j||S)aPlot a single or multiple values from the metric. Args: val: Either a single result from calling `metric.forward` or `metric.compute` or a list of these results. If no value is provided, will automatically call `metric.compute` and plot that result. ax: An matplotlib axis object. If provided will add plot to that axis Returns: Figure and Axes object Raises: ModuleNotFoundError: If `matplotlib` is not installed .. plot:: :scale: 75 >>> # Example plotting a single value >>> import torch >>> from torchmetrics.audio import SignalNoiseRatio >>> metric = SignalNoiseRatio() >>> metric.update(torch.rand(4), torch.rand(4)) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> import torch >>> from torchmetrics.audio import SignalNoiseRatio >>> metric = SignalNoiseRatio() >>> values = [ ] >>> for _ in range(10): ... values.append(metric(torch.rand(4), torch.rand(4))) >>> fig_, ax_ = metric.plot(values) _plotr'r7r8s r)plotzSignalNoiseRatio.plotgsPzz#r""r*FNN)__name__ __module__ __qualname____doc__rbool__annotations__rrrrrfloatrrr$r2r6rrrrr= __classcell__r(s@r)rr$s"H$t#"t"!d! O M(,huo,(,huo,  I I I  I(F(F(t()) _c(#E&(6*:":;<(#IQRZI[(# (#r*rceZdZUdZdZeed<eed<dZdZe e ed<dZ e e ed<de d dffd Z d ed ed dfd Zd efdZddeeeedfde ed efdZxZS)ScaleInvariantSignalNoiseRatioa7Calculate `Scale-invariant signal-to-noise ratio`_ (SI-SNR) metric for evaluating quality of audio. As input to `forward` and `update` the metric accepts the following input - ``preds`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` - ``target`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` As output of `forward` and `compute` the metric returns the following output - ``si_snr`` (:class:`~torch.Tensor`): float scalar tensor with average SI-SNR value over samples Args: kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Raises: TypeError: if target and preds have a different shape Example: >>> import torch >>> from torch import tensor >>> from torchmetrics.audio import ScaleInvariantSignalNoiseRatio >>> target = tensor([3.0, -0.5, 2.0, 7.0]) >>> preds = tensor([2.5, 0.0, 2.0, 8.0]) >>> si_snr = ScaleInvariantSignalNoiseRatio() >>> si_snr(preds, target) tensor(15.0918) T sum_si_snrrNrrrrc t|di||jdtdd|jdtddy)NrKrrrrrr")r#r$r%r)r'rr(s r)r$z'ScaleInvariantSignalNoiseRatio.__init__sA "6" |VC[O wq %Hr*r+r,ct||}|xj|jz c_|xj|j z c_y)r/)r+r,N)r rKrrr0)r'r+r, si_snr_batchs r)r2z%ScaleInvariantSignalNoiseRatio.updates<9fU  <++-- l((** r*c4|j|jz Sr4)rKrr5s r)r6z&ScaleInvariantSignalNoiseRatio.computes++r*r7r8c&|j||S)a6Plot 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 ScaleInvariantSignalNoiseRatio >>> metric = ScaleInvariantSignalNoiseRatio() >>> metric.update(torch.rand(4), torch.rand(4)) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> import torch >>> from torchmetrics.audio import ScaleInvariantSignalNoiseRatio >>> metric = ScaleInvariantSignalNoiseRatio() >>> values = [ ] >>> for _ in range(10): ... values.append(metric(torch.rand(4), torch.rand(4))) >>> fig_, ax_ = metric.plot(values) r:r<s r)r=z#ScaleInvariantSignalNoiseRatio.plotLzz#r""r*r?)r@rArBrCrrrErrrrFrrr$r2r6rrrrr=rGrHs@r)rJrJs< M(,huo,(,huo,II I+F+F+t+,,&#fhv&6<=&#(S[J\&#hv&#r*rJceZdZUdZdZeed<eed<dZdZe e ed<dZ e e ed< dde d e d dffd Zd ed ed dfdZd efdZddeeeedfde ed efdZxZS)%ComplexScaleInvariantSignalNoiseRatioaCalculate `Complex scale-invariant signal-to-noise ratio`_ (C-SI-SNR) metric for evaluating quality of audio. As input to `forward` and `update` the metric accepts the following input - ``preds`` (:class:`~torch.Tensor`): real float tensor with shape ``(...,frequency,time,2)`` or complex float tensor with shape ``(..., frequency,time)`` - ``target`` (:class:`~torch.Tensor`): real float tensor with shape ``(...,frequency,time,2)`` or complex float tensor with shape ``(..., frequency,time)`` As output of `forward` and `compute` the metric returns the following output - ``c_si_snr`` (:class:`~torch.Tensor`): float scalar tensor with average C-SI-SNR value over samples Args: zero_mean: if to zero mean target and preds or not kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Raises: ValueError: If ``zero_mean`` is not an bool TypeError: If ``preds`` is not the shape (..., frequency, time, 2) (after being converted to real if it is complex). If ``preds`` and ``target`` does not have the same shape. Example: >>> from torch import randn >>> from torchmetrics.audio import ComplexScaleInvariantSignalNoiseRatio >>> preds = randn((1,257,100,2)) >>> target = randn((1,257,100,2)) >>> c_si_snr = ComplexScaleInvariantSignalNoiseRatio() >>> c_si_snr(preds, target) tensor(-38.8832) T ci_snr_sumnumNrrrrrc t|di|t|tst d|||_|j dtdd|j dtddy) Nz5Expected argument `zero_mean` to be an bool, but got rTrrrrUrr")r#r$ isinstancerD ValueErrorrr%rr&s r)r$z.ComplexScaleInvariantSignalNoiseRatio.__init__!sg "6")T*TU^T_`a a" |VC[O ufQiFr*r+r,ct|||j}|xj|jz c_|xj|j z c_yr.)r rrTrrUr0)r'r+r,vs r)r2z,ComplexScaleInvariantSignalNoiseRatio.update.s? 6U6]a]k]k l 1557" AGGIr*c4|j|jz Sr4)rTrUr5s r)r6z-ComplexScaleInvariantSignalNoiseRatio.compute5s))r*r7r8c&|j||S)azPlot a single or multiple values from the metric. Args: val: Either a single result from calling `metric.forward` or `metric.compute` or a list of these results. If no value is provided, will automatically call `metric.compute` and plot that result. ax: An matplotlib axis object. If provided will add plot to that axis Returns: Figure and Axes object Raises: ModuleNotFoundError: If `matplotlib` is not installed .. plot:: :scale: 75 >>> # Example plotting a single value >>> import torch >>> from torchmetrics.audio import ComplexScaleInvariantSignalNoiseRatio >>> metric = ComplexScaleInvariantSignalNoiseRatio() >>> metric.update(torch.rand(1,257,100,2), torch.rand(1,257,100,2)) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> import torch >>> from torchmetrics.audio import ComplexScaleInvariantSignalNoiseRatio >>> metric = ComplexScaleInvariantSignalNoiseRatio() >>> values = [ ] >>> for _ in range(10): ... values.append(metric(torch.rand(1,257,100,2), torch.rand(1,257,100,2))) >>> fig_, ax_ = metric.plot(values) r:r<s r)r=z*ComplexScaleInvariantSignalNoiseRatio.plot9rQr*r>r?)r@rArBrCrrrErrrrFrrDrr$r2r6rrrrr=rGrHs@r)rSrSs"H K(,huo,(,huo,  G G G  GFFt**&#fhv&6<=&#(S[J\&#hv&#r*rSN)collections.abcrtypingrrrtorchrr!torchmetrics.functional.audio.snrr r r torchmetrics.metricr torchmetrics.utilities.importsr torchmetrics.utilities.plotrr__doctest_skip__rrJrSr"r*r)res_%''  '@@k#vk#\`#V`#Fj#Fj#r*