to_binary_class

sdm.evaluation.to_binary_class(pred: TableTensor, target: TableTensor | CategoricalTensor, positive_class: bool | int | float | str) → tuple[Tensor, Tensor]

Convert binary predictions and targets to positive-class form.

Parameters:

• pred (TableTensor) – Prediction table whose numerical columns contain class scores. Column names define the class order.

• target (TableTensor | CategoricalTensor) – Single-column categorical target.

• positive_class (bool | int | float | str) – Class value treated as the positive class.

Return type:

tuple[Tensor, Tensor]

>>> import torch
>>> import sdm
>>> from sdm.evaluation import to_binary_class
>>> pred = sdm.TableTensor.from_tensor(
...     torch.tensor([[0.8, 0.2], [0.1, 0.9], [0.4, 0.6]]),
...     columns=["false", "true"],
... )
>>> target = sdm.CategoricalTensor(
...     code=torch.tensor([[0], [1], [1]]),
...     categories=(sdm.StringTensor.from_list(["false", "true"]),),
... )
>>> positive_scores, binary_target = to_binary_class(
...     pred,
...     target,
...     positive_class="true",
... )
>>> positive_scores
tensor([0.2000, 0.9000, 0.6000])
>>> binary_target
tensor([False,  True,  True])
>>> pred = sdm.TableTensor.from_tensor(
...     torch.tensor([[0.2, 0.8], [0.9, 0.1], [0.6, 0.4]]),
...     columns=["true", "false"],
... )
>>> positive_scores, binary_target = to_binary_class(
...     pred,
...     target,
...     positive_class="true",
... )
>>> positive_scores
tensor([0.2000, 0.9000, 0.6000])
>>> binary_target
tensor([False,  True,  True])