Relative Average Spectral Error (RASE)

Module Interface

class torchmetrics.image.RelativeAverageSpectralError(window_size=8, **kwargs)

Computes Relative Average Spectral Error (RASE) (RelativeAverageSpectralError).

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

• preds (Tensor): Predictions from model of shape (N,C,H,W)

• target (Tensor): Ground truth values of shape (N,C,H,W)

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

• rase (Tensor): returns float scalar tensor with average RASE value over sample

Parameters:

• window_size (int) – Sliding window used for rmse calculation

• kwargs (Dict[str, Any]) – Additional keyword arguments, see Advanced metric settings for more info.

Returns:

Relative Average Spectral Error (RASE)

Example

>>> import torch
>>> from torchmetrics.image import RelativeAverageSpectralError
>>> g = torch.manual_seed(22)
>>> preds = torch.rand(4, 3, 16, 16)
>>> target = torch.rand(4, 3, 16, 16)
>>> rase = RelativeAverageSpectralError()
>>> rase(preds, target)
tensor(5114.6641)

Raises:

ValueError – If window_size is not a positive integer.

plot(val=None, ax=None)

Plot a single or multiple values from the metric.

Parameters:

• val (Union[Tensor, Sequence[Tensor], None]) – 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 (Optional[Axes]) – An matplotlib axis object. If provided will add plot to that axis

Return type:

Tuple[Figure, Union[Axes, ndarray]]

Returns:

Figure and Axes object

Raises:

ModuleNotFoundError – If matplotlib is not installed

>>> # Example plotting a single value
>>> import torch
>>> from torchmetrics.image import RelativeAverageSpectralError
>>> metric = RelativeAverageSpectralError()
>>> metric.update(torch.rand(4, 3, 16, 16), torch.rand(4, 3, 16, 16))
>>> fig_, ax_ = metric.plot()

>>> # Example plotting multiple values
>>> import torch
>>> _ = torch.manual_seed(42)
>>> from torchmetrics.image import RelativeAverageSpectralError
>>> metric = RelativeAverageSpectralError()
>>> values = [ ]
>>> for _ in range(10):
...     values.append(metric(torch.rand(4, 3, 16, 16), torch.rand(4, 3, 16, 16)))
>>> fig_, ax_ = metric.plot(values)

Functional Interface

torchmetrics.functional.image.relative_average_spectral_error(preds, target, window_size=8)

Compute Relative Average Spectral Error (RASE) (RelativeAverageSpectralError).

Parameters:

• preds (Tensor) – Deformed image

• target (Tensor) – Ground truth image

• window_size (int) – Sliding window used for rmse calculation

Return type:

Tensor

Returns:

Relative Average Spectral Error (RASE)

Example

>>> from torchmetrics.functional.image import relative_average_spectral_error
>>> g = torch.manual_seed(22)
>>> preds = torch.rand(4, 3, 16, 16)
>>> target = torch.rand(4, 3, 16, 16)
>>> relative_average_spectral_error(preds, target)
tensor(5114.6641)

Raises:

ValueError – If window_size is not a positive integer.