links

docs/source/links.rst

@@ -64,10 +64,10 @@
 .. _Improved Techniques for Training GANs: https://arxiv.org/abs/1606.03498
 .. _KID Score: https://github.com/toshas/torch-fidelity/blob/v0.3.0/torch_fidelity/metric_kid.py
 .. _Demystifying MMD GANs: https://arxiv.org/abs/1801.01401
-.. _Computes Peak Signal-to-Noise Ratio: https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio
+.. _Compute Peak Signal-to-Noise Ratio: https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio
 .. _Turn a Metric into a Bootstrapped: https://en.wikipedia.org/wiki/Bootstrapping_(statistics)
 .. _Metric Test for Reset: https://github.com/Lightning-AI/pytorch-lightning/pull/7055
-.. _Computes Mean Absolute Error: https://en.wikipedia.org/wiki/Mean_absolute_error
+.. _Compute Mean Absolute Error: https://en.wikipedia.org/wiki/Mean_absolute_error
 .. _Mean Absolute Percentage Error: https://en.wikipedia.org/wiki/Mean_absolute_percentage_error
 .. _mean squared error: https://en.wikipedia.org/wiki/Mean_squared_error
 .. _Aggregate the statistics from multiple devices: https://stackoverflow.com/questions/68395368/estimate-running-correlation-on-multiple-nodes

src/torchmetrics/classification/accuracy.py

@@ -32,7 +32,7 @@
 
 
 class BinaryAccuracy(BinaryStatScores):
-    r"""Computes `Accuracy`_ for binary tasks:
+    r"""Compute `Accuracy`_ for binary tasks.
 
     .. math::
         \text{Accuracy} = \frac{1}{N}\sum_i^N 1(y_i = \hat{y}_i)
@@ -41,16 +41,16 @@ class BinaryAccuracy(BinaryStatScores):
 
     As input to ``forward`` and ``update`` the metric accepts the following input:
 
-    - ``preds`` (:class:`~torch.Tensor`): An int or float tensor of shape ``(N, ...)``. If preds is a floating
-      point tensor with values outside [0,1] range we consider the input to be logits and will auto apply sigmoid
-      per element. Addtionally, we convert to int tensor with thresholding using the value in ``threshold``.
-    - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)``
+        - ``preds`` (:class:`~torch.Tensor`): An int or float tensor of shape ``(N, ...)``. If preds is a floating
+          point tensor with values outside [0,1] range we consider the input to be logits and will auto apply sigmoid
+          per element. Addtionally, we convert to int tensor with thresholding using the value in ``threshold``.
+        - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)``
 
     As output to ``forward`` and ``compute`` the metric returns the following output:
 
-    - ``ba`` (:class:`~torch.Tensor`): If ``multidim_average`` is set to ``global``, the metric returns a scalar value.
-      If ``multidim_average`` is set to ``samplewise``, the metric returns ``(N,)`` vector consisting of a scalar
-      value per sample.
+        - ``ba`` (:class:`~torch.Tensor`): If ``multidim_average`` is set to ``global``, metric returns a scalar value.
+          If ``multidim_average`` is set to ``samplewise``, the metric returns ``(N,)`` vector consisting of a scalar
+          value per sample.
 
     Args:
         threshold: Threshold for transforming probability to binary {0,1} predictions
@@ -161,25 +161,25 @@ class MulticlassAccuracy(MulticlassStatScores):
 
     As input to ``forward`` and ``update`` the metric accepts the following input:
 
-    - ``preds`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)`` or float tensor of shape ``(N, C, ..)``.
-      If preds is a floating point we apply ``torch.argmax`` along the ``C`` dimension to automatically convert
-      probabilities/logits into an int tensor.
-    - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)``
+        - ``preds`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)`` or float tensor
+          of shape ``(N, C, ..)``. If preds is a floating point we apply ``torch.argmax`` along the ``C`` dimension
+          to automatically convert probabilities/logits into an int tensor.
+        - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)``
 
     As output to ``forward`` and ``compute`` the metric returns the following output:
 
-    - ``mca`` (:class:`~torch.Tensor`): A tensor with the accuracy score whose returned shape depends on the
-      ``average`` and ``multidim_average`` arguments:
+        - ``mca`` (:class:`~torch.Tensor`): A tensor with the accuracy score whose returned shape depends on the
+          ``average`` and ``multidim_average`` arguments:
 
-        - If ``multidim_average`` is set to ``global``:
+            - If ``multidim_average`` is set to ``global``:
 
-          - If ``average='micro'/'macro'/'weighted'``, the output will be a scalar tensor
-          - If ``average=None/'none'``, the shape will be ``(C,)``
+              - If ``average='micro'/'macro'/'weighted'``, the output will be a scalar tensor
+              - If ``average=None/'none'``, the shape will be ``(C,)``
 
-        - If ``multidim_average`` is set to ``samplewise``:
+            - If ``multidim_average`` is set to ``samplewise``:
 
-          - If ``average='micro'/'macro'/'weighted'``, the shape will be ``(N,)``
-          - If ``average=None/'none'``, the shape will be ``(N, C)``
+              - If ``average='micro'/'macro'/'weighted'``, the shape will be ``(N,)``
+              - If ``average=None/'none'``, the shape will be ``(N, C)``
 
     Args:
         num_classes: Integer specifing the number of classes

src/torchmetrics/functional/image/psnr.py

@@ -27,7 +27,7 @@ def _psnr_compute(
     base: float = 10.0,
     reduction: Literal["elementwise_mean", "sum", "none", None] = "elementwise_mean",
 ) -> Tensor:
-    """Computes peak signal-to-noise ratio.
+    """Compute peak signal-to-noise ratio.
 
     Args:
         sum_squared_error: Sum of square of errors over all observations
@@ -60,13 +60,13 @@ def _psnr_update(
     target: Tensor,
     dim: Optional[Union[int, Tuple[int, ...]]] = None,
 ) -> Tuple[Tensor, Tensor]:
-    """Updates and returns variables required to compute peak signal-to-noise ratio.
+    """Update and return variables required to compute peak signal-to-noise ratio.
 
     Args:
         preds: Predicted tensor
         target: Ground truth tensor
-        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.
+        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.
     """
 
     if dim is None:

src/torchmetrics/functional/regression/mae.py

@@ -37,7 +37,7 @@ def _mean_absolute_error_update(preds: Tensor, target: Tensor) -> Tuple[Tensor,
 
 
 def _mean_absolute_error_compute(sum_abs_error: Tensor, n_obs: int) -> Tensor:
-    """Computes Mean Absolute Error.
+    """Compute Mean Absolute Error.
 
     Args:
         sum_abs_error: Sum of absolute value of errors over all observations
@@ -55,7 +55,7 @@ def _mean_absolute_error_compute(sum_abs_error: Tensor, n_obs: int) -> Tensor:
 
 
 def mean_absolute_error(preds: Tensor, target: Tensor) -> Tensor:
-    """Computes mean absolute error.
+    """Compute mean absolute error.
 
     Args:
         preds: estimated labels

src/torchmetrics/image/psnr.py

@@ -23,7 +23,7 @@
 
 
 class PeakSignalNoiseRatio(Metric):
-    r"""Computes `Computes Peak Signal-to-Noise Ratio`_ (PSNR):
+    r"""`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)
 

src/torchmetrics/regression/mae.py

@@ -20,7 +20,7 @@
 
 
 class MeanAbsoluteError(Metric):
-    r"""`Computes Mean Absolute Error`_ (MAE):
+    r"""`Compute Mean Absolute Error`_ (MAE):
 
     .. math:: \text{MAE} = \frac{1}{N}\sum_i^N | y_i - \hat{y_i} |
 
@@ -70,5 +70,5 @@ def update(self, preds: Tensor, target: Tensor) -> None:  # type: ignore
         self.total += n_obs
 
     def compute(self) -> Tensor:
-        """Computes mean absolute error over state."""
+        """Compute mean absolute error over state."""
         return _mean_absolute_error_compute(self.sum_abs_error, self.total)