torcheval.metrics.WordErrorRate

class torcheval.metrics.WordErrorRate(*, device: Optional[device] = None)

Compute the word error rate of the predicted word sequence(s) with the reference word sequence(s). Its functional version is torcheval.metrics.functional.word_error_rate().

Examples

>>> import torch
>>> from torcheval.metrics import WordErrorRate

>>> metric = WordErrorRate()
>>> metric.update(["this is the prediction", "there is an other sample"],
["this is the reference", "there is another one"])
>>> metric.compute()
tensor(0.5)

>>> metric = WordErrorRate()
>>> metric.update(["this is the prediction", "there is an other sample"],
["this is the reference", "there is another one"])
>>> metric.update(["hello world", "welcome to the facebook"],
["hello metaverse", "welcome to meta"])
>>> metric.compute()
tensor(0.53846)

__init__(*, device: Optional[device] = 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, or a dictionary with torch.Tensor as values

Methods

__init__(*[, device])	Initialize a metric object and its internal states.
compute()	Return the word error rate score
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 edit distance and the length of the reference sequence.

Attributes

device

The last input device of Metric.to().