Meanmetric

meanmetric

Metric MeanMetric from torchmetrics (torchmetrics.MeanMetric)

When to invoke this skill

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

Reference signature

from torchmetrics import MeanMetric

# MeanMetric(nan_strategy: Union[Literal['error', 'warn', 'ignore', 'disable'], float] = 'warn', **kwargs: Any) -> None

Library docstring

Aggregate a stream of value into their mean value.

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

- ``value`` (:class:`~float` or :class:`~torch.Tensor`): a single float or an tensor of float values with
  arbitrary shape ``(...,)``.
- ``weight`` (:class:`~float` or :class:`~torch.Tensor`): a single float or an tensor of float value with
  arbitrary shape ``(...,)``. Needs to be broadcastable with the shape of ``value`` tensor.

As output of `forward` and `compute` the metric returns the following output

- ``agg`` (:class:`~torch.Tensor`): scalar float tensor with aggregated (weighted) mean over all inputs received

Args:
    nan_strategy: options:
        - ``'error'``: if any `nan` values are encountered will give a RuntimeError
        - ``'warn'``: if any `nan` values are encountered will give a warning and continue
        - ``'ignore'``: all `nan` values are silently removed
        - ``'disable'``: disable all `nan` checks
        - a float: if a float is provided will impute any `nan` values with this value

    kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info.

Raises:
    ValueError:
        If ``nan_strategy`` is not one of ``error``, ``warn``, ``ignore``, ``disable`` or a float

Example:
    >>> from torchmetrics.aggregation import MeanMetric
    >>> metric = MeanMetric()
    >>> metric.update(1)
    >>> metric.update(torch.tensor([2, 3]))
    >>> metric.compute()
    tensor(2.)

Quick recipe

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