torcheval.metrics.functional.binary_binned_auprc

torcheval.metrics.functional.binary_binned_auprc(input: Tensor, target: Tensor, *, num_tasks: int = 1, threshold: Union[int, List[float], Tensor] = 100) → Tuple[Tensor, Tensor]

Binned Version of AUPRC, which is the area under the AUPRC Curve, for binary classification. Its class version is torcheval.metrics.BinaryBinnedAUPRC.

Computation is done by computing the area under the precision/recall curve; precision and recall are computed for the buckets defined by threshold.

Parameters:

• input (Tensor) – Tensor of label predictions It should be predicted label, probabilities or logits with shape of (num_tasks, n_sample) or (n_sample, ).
• target (Tensor) – Tensor of ground truth labels with shape of (num_tasks, n_sample) or (n_sample, ).
• num_tasks (int) – Number of tasks that need binary_binned_auprc calculation. Default value is 1. binary_binned_auprc for each task will be calculated independently.
• threshold (Tensor, int, List[float]) – Either an integer representing the number of bins, a list of thresholds, or a tensor of thresholds. The same thresholds will be used for all tasks. If threshold is a tensor, it must be 1D. If list or tensor is given, the first element must be 0 and the last must be 1.

Examples

>>> import torch
>>> from torcheval.metrics.functional import binary_binned_auprc
>>> input = torch.tensor([0.2, 0.3, 0.4, 0.5])
>>> target = torch.tensor([0, 0, 1, 1])
>>> binary_binned_auprc(input, target, threshold=5)
(tensor(1.0),
tensor([0.0000, 0.2500, 0.5000, 0.7500, 1.0000]))

>>> input = torch.tensor([0.2, 0.3, 0.4, 0.5])
>>> target = torch.tensor([0, 0, 1, 1])
>>> threshold = torch.tensor([0.0000, 0.2500, 0.7500, 1.0000])
>>> binary_binned_auprc(input, target, threshold=threshold)
(tensor(0.6667),
tensor([0.0000, 0.2500, 0.7500, 1.0000]))

>>> input = torch.tensor([[0.2, 0.3, 0.4, 0.5], [0, 1, 2, 3]])
>>> target = torch.tensor([[0, 0, 1, 1], [0, 1, 1, 1]])
>>> threshold = torch.tensor([0.0000, 0.2500, 0.7500, 1.0000])
>>> binary_binned_auprc(input, target, num_tasks=2, threshold=threshold)
(tensor([0.6667, 1.0000],
tensor([0.0000, 0.2500, 0.7500, 1.0000]))