File size: 5,535 Bytes
9c6594c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 |
# Copyright The PyTorch Lightning team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from collections.abc import Sequence
from typing import Any, Optional, Union
import torch
from torch import Tensor, tensor
from torchmetrics.functional.image.psnrb import _psnrb_compute, _psnrb_update
from torchmetrics.metric import Metric
from torchmetrics.utilities.imports import _MATPLOTLIB_AVAILABLE
from torchmetrics.utilities.plot import _AX_TYPE, _PLOT_OUT_TYPE
if not _MATPLOTLIB_AVAILABLE:
__doctest_skip__ = ["PeakSignalNoiseRatioWithBlockedEffect.plot"]
class PeakSignalNoiseRatioWithBlockedEffect(Metric):
r"""Computes `Peak Signal to Noise Ratio With Blocked Effect`_ (PSNRB).
.. math::
\text{PSNRB}(I, J) = 10 * \log_{10} \left(\frac{\max(I)^2}{\text{MSE}(I, J)-\text{B}(I, J)}\right)
Where :math:`\text{MSE}` denotes the `mean-squared-error`_ function. This metric is a modified version of PSNR that
better supports evaluation of images with blocked artifacts, that oftens occur in compressed images.
.. attention::
Metric only supports grayscale images. If you have RGB images, please convert them to grayscale first.
As input to ``forward`` and ``update`` the metric accepts the following input
- ``preds`` (:class:`~torch.Tensor`): Predictions from model of shape ``(N,1,H,W)``
- ``target`` (:class:`~torch.Tensor`): Ground truth values of shape ``(N,1,H,W)``
As output of `forward` and `compute` the metric returns the following output
- ``psnrb`` (:class:`~torch.Tensor`): float scalar tensor with aggregated PSNRB value
Args:
block_size: integer indication the block size
kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info.
Example:
>>> from torch import rand
>>> metric = PeakSignalNoiseRatioWithBlockedEffect()
>>> preds = rand(2, 1, 10, 10)
>>> target = rand(2, 1, 10, 10)
>>> metric(preds, target)
tensor(7.2893)
"""
is_differentiable: bool = True
higher_is_better: bool = True
full_state_update: bool = False
sum_squared_error: Tensor
total: Tensor
bef: Tensor
data_range: Tensor
def __init__(
self,
block_size: int = 8,
**kwargs: Any,
) -> None:
super().__init__(**kwargs)
if not isinstance(block_size, int) and block_size < 1:
raise ValueError("Argument ``block_size`` should be a positive integer")
self.block_size = block_size
self.add_state("sum_squared_error", default=tensor(0.0), dist_reduce_fx="sum")
self.add_state("total", default=tensor(0), dist_reduce_fx="sum")
self.add_state("bef", default=tensor(0.0), dist_reduce_fx="sum")
self.add_state("data_range", default=tensor(0), dist_reduce_fx="max")
def update(self, preds: Tensor, target: Tensor) -> None:
"""Update state with predictions and targets."""
sum_squared_error, bef, num_obs = _psnrb_update(preds, target, block_size=self.block_size)
self.sum_squared_error += sum_squared_error
self.bef += bef
self.total += num_obs
self.data_range = torch.maximum(self.data_range, torch.max(target) - torch.min(target))
def compute(self) -> Tensor:
"""Compute peak signal-to-noise ratio over state."""
return _psnrb_compute(self.sum_squared_error, self.bef, self.total, self.data_range)
def plot(
self, val: Optional[Union[Tensor, Sequence[Tensor]]] = None, ax: Optional[_AX_TYPE] = None
) -> _PLOT_OUT_TYPE:
"""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.image import PeakSignalNoiseRatioWithBlockedEffect
>>> metric = PeakSignalNoiseRatioWithBlockedEffect()
>>> metric.update(torch.rand(2, 1, 10, 10), torch.rand(2, 1, 10, 10))
>>> fig_, ax_ = metric.plot()
.. plot::
:scale: 75
>>> # Example plotting multiple values
>>> import torch
>>> from torchmetrics.image import PeakSignalNoiseRatioWithBlockedEffect
>>> metric = PeakSignalNoiseRatioWithBlockedEffect()
>>> values = [ ]
>>> for _ in range(10):
... values.append(metric(torch.rand(2, 1, 10, 10), torch.rand(2, 1, 10, 10)))
>>> fig_, ax_ = metric.plot(values)
"""
return self._plot(val, ax)
|