Modifiedpanopticquality

modifiedpanopticquality

Metric ModifiedPanopticQuality from torchmetrics (torchmetrics.ModifiedPanopticQuality)

When to invoke this skill

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

Reference signature

from torchmetrics import ModifiedPanopticQuality

# _ModifiedPanopticQuality(things: collections.abc.Collection[int], stuffs: collections.abc.Collection[int], allow_unknown_preds_category: bool = False, **kwargs: Any) -> None

Library docstring

Wrapper for deprecated import.

>>> from torch import tensor
>>> preds = tensor([[[0, 0], [0, 1], [6, 0], [7, 0], [0, 2], [1, 0]]])
>>> target = tensor([[[0, 1], [0, 0], [6, 0], [7, 0], [6, 0], [255, 0]]])
>>> pq_modified = _ModifiedPanopticQuality(things = {0, 1}, stuffs = {6, 7})
>>> pq_modified(preds, target)
tensor(0.7667, dtype=torch.float64)

Quick recipe

import torchmetrics as _m
score = _m.ModifiedPanopticQuality(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).