Statscores

statscores

Metric StatScores from torchmetrics (torchmetrics.StatScores)

When to invoke this skill

The user has predictions + ground truth and asks to evaluate with StatScores, or mentions torchmetrics.StatScores directly, or wants the standard torchmetrics implementation.

Reference signature

from torchmetrics import StatScores

# StatScores(task: Literal['binary', 'multiclass', 'multilabel'], threshold: float = 0.5, num_classes: Optional[int] = None, num_labels: Optional[int] = None, average: Optional[Literal['micro', 'macro', 'weighted', 'none']] = 'micro', multidim_average: Optional[Literal['global', 'samplewise']] = 'global', top_k: Optional[int] = 1, ignore_index: Optional[int] = None, validate_args: bool = True, **kwargs: Any) -> torchmetrics.metric.Metric

Library docstring

Compute the number of true positives, false positives, true negatives, false negatives and the support.

This function is a simple wrapper to get the task specific versions of this metric, which is done by setting the
``task`` argument to either ``'binary'``, ``'multiclass'`` or ``'multilabel'``. See the documentation of
:class:`~torchmetrics.classification.BinaryStatScores`, :class:`~torchmetrics.classification.MulticlassStatScores`
and :class:`~torchmetrics.classification.MultilabelStatScores` for the specific details of each argument influence
and examples.

Legacy Example:
    >>> from torch import tensor
    >>> preds  = tensor([1, 0, 2, 1])
    >>> target = tensor([1, 1, 2, 0])
    >>> stat_scores = StatScores(task="multiclass", num_classes=3, average='micro')
    >>> stat_scores(preds, target)
    tensor([2, 2, 6, 2, 4])
    >>> stat_scores = StatScores(task="multiclass", num_classes=3, average=None)
    >>> stat_scores(preds, target)
    tensor([[0, 1, 2, 1, 1],
            [1, 1, 1, 1, 2],
            [1, 0, 3, 0, 1]])

Quick recipe

import torchmetrics as _m
score = _m.StatScores(y_true, y_pred)

Don'ts

• Don't reimplement when the library version handles edge cases (NaN, ties, empty inputs) better than a hand-rolled formula.
• Always check the library version's argument order — sklearn is (y_true, y_pred) while torchmetrics is (preds, target).