torcheval.metrics.functional.auc

torcheval.metrics.functional.auc(x: Tensor, y: Tensor, reorder: bool = False) → Tensor

Computes Area Under the Curve (AUC) using the trapezoidal rule. :param x: x-coordinates :param y: y-coordinates :param reorder: sorts the x input tensor in order, default value is False

Returns:

Tensor containing AUC score (float)

Raises:

ValueError – If both x and y don’t have the same shape. If both x and y have atleast 1 element.

Example

>>> from torcheval.metrics.functional.aggregation.auc import auc
>>> x = torch.tensor([0,.1,.2,.3])
>>> y = torch.tensor([1,1,1,1])
>>> auc(x, y)
tensor([0.3000])
>>> y = torch.tensor([[0, 4, 0, 4, 3],
                [1, 1, 2, 1, 1],
                [4, 3, 1, 4, 4],
                [1, 0, 0, 3, 0]])
>>> x = torch.tensor([[0.2535, 0.1138, 0.1324, 0.1887, 0.3117],
                [0.1434, 0.4404, 0.1100, 0.1178, 0.1883],
                [0.2344, 0.1743, 0.3110, 0.0393, 0.2410],
                [0.1381, 0.1564, 0.0320, 0.2220, 0.4515]])
>>> auc(x, y, reorder=True) # Reorders X and calculates AUC.
tensor([0.3667, 0.3343, 0.8843, 0.5048])