torcheval.metrics.HitRate

class torcheval.metrics.HitRate(*, k: int | None = None, device: device | None = None)

Compute the hit rate of the correct class among the top predicted classes. Its functional version is torcheval.metrics.functional.hit_rate().

Parameters:

k (int, optional) – Number of top class probabilities to be considered. If k is None, all classes are considered and a hit rate of 1.0 is returned.

Examples:

>>> import torch
>>> from torcheval.metrics import HitRate

>>> metric = HitRate()
>>> metric.update(torch.tensor([[0.3, 0.1, 0.6], [0.5, 0.2, 0.3]]), torch.tensor([2, 1]))
>>> metric.update(torch.tensor([[0.2, 0.1, 0.7], [0.3, 0.3, 0.4]]), torch.tensor([1, 0]))
>>> metric.compute()
tensor([1., 1., 1., 1.])

>>> metric = HitRate(k=2)
>>> metric.update(torch.tensor([[0.3, 0.1, 0.6], [0.5, 0.2, 0.3]]), torch.tensor([2, 1]))
>>> metric.update(torch.tensor([[0.2, 0.1, 0.7], [0.3, 0.3, 0.4]]), torch.tensor([1, 0]))
>>> metric.compute()
tensor([1., 0., 0., 1.])

__init__(*, k: int | None = None, device: device | None = None) → None

Initialize a metric object and its internal states.

Use self._add_state() to initialize state variables of your metric class. The state variables should be either torch.Tensor, a list of torch.Tensor, a dictionary with torch.Tensor as values, or a deque of torch.Tensor.

Methods

__init__(*[, k, device])	Initialize a metric object and its internal states.
compute()	Return the concatenated hite rate scores.
load_state_dict(state_dict[, strict])	Loads metric state variables from state_dict.
merge_state(metrics)	Merge the metric state with its counterparts from other metric instances.
reset()	Reset the metric state variables to their default value.
state_dict()	Save metric state variables in state_dict.
to(device, *args, **kwargs)	Move tensors in metric state variables to device.
update(input, target)	Update the metric state with the ground truth labels and predictions.

Attributes

device

The last input device of Metric.to().