Multiclassjaccardindex

multiclassjaccardindex

Metric MulticlassJaccardIndex from torchmetrics (torchmetrics.classification.MulticlassJaccardIndex)

When to invoke this skill

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

Reference signature

from torchmetrics.classification import MulticlassJaccardIndex

# MulticlassJaccardIndex(num_classes: int, average: Optional[Literal['micro', 'macro', 'weighted', 'none']] = 'macro', ignore_index: Optional[int] = None, validate_args: bool = True, zero_division: float = 0, **kwargs: Any) -> None

Library docstring

Calculate the Jaccard index for multiclass tasks.

The `Jaccard index`_ (also known as the intersection over union or jaccard similarity coefficient) is an statistic
that can be used to determine the similarity and diversity of a sample set. It is defined as the size of the
intersection divided by the union of the sample sets:

.. math:: J(A,B) = \frac{|A\cap B|}{|A\cup B|}

As input to ``forward`` and ``update`` the metric accepts the following input:

- ``preds`` (:class:`~torch.Tensor`): A int tensor of shape ``(N, ...)`` or float tensor of shape ``(N, C, ..)``.
  If preds is a floating point we apply ``torch.argmax`` along the ``C`` dimension to automatically convert
  probabilities/logits into an int tensor.
- ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, ...)``.

.. tip::
   Additional dimension ``...`` will be flattened into the batch dimension.

As output to ``forward`` and ``compute`` the metric returns the following output:

- ``mcji`` (:class:`~torch.Tensor`): A tensor containing the Multi-class Jaccard Index.

Args:
    num_classes: Integer specifying the number of classes
    ignore_index:
        Specifies a target value that is ignored and does not contribute to the metric calculation
    average:
        Defines the reduction that is applied over labels. Should be one of the following:

        - ``micro``: Sum statistics over all labels
        - ``macro``: Calculate statistics for each label and average them
        - ``weighted``: calculates statistics for each label and computes weighted average using their support
        - ``"none"`` or ``None``: calculates statistic for each label and applies no reduction

    validate_args: bool indicating if input arguments and tensors should be validated for correctness.
        Set to ``False`` for faster computations.
    zero_division:
        Value to replace when there is a division by zero. Should be `0` or `1`.
    kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info.

Example (pred is integer tensor):
    >>> from torch import tensor
    >>> from torchmetrics.classification import MulticlassJaccardIndex
    >>> target = tensor([2, 1, 0, 0])
    >>> pre

Quick recipe

import torchmetrics.classification as _m
score = _m.MulticlassJaccardIndex(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).