torcheval.metrics.BLEUScore

class torcheval.metrics.BLEUScore(*, n_gram: int, weights: Optional[Tensor] = None, device: Optional[device] = None)

Compute BLEU score (https://en.wikipedia.org/wiki/BLEU) given translations and references. Its functional version is torcheval.metrics.functional.text.bleu.

Parameters:

• n_gram – Maximum n-gram to use when computing BLEU score. Can be 1, 2, 3, or 4.
• weights – Optional weight distribution of n-grams. Requires len(weights) = n_gram. If unspecified, will use uniform weights.

Examples

>>> import torch
>>> from torcheval.metrics import BLEUScore
>>> metric = BLEUScore(n_gram=4)
>>> candidates = ["the squirrel is eating the nut", "the cat is on the mat"]
>>> references = [["a squirrel is eating a nut", "the squirrel is eating a tasty nut"], ["there is a cat on the mat", "a cat is on the mat"]]
>>> metric.update(candidates, references)
>>> metric.compute()
tensor(0.65341892)
>>> candidates = ["i like ice cream and apple pie"]
>>> references = [["i like apple pie with ice cream on top", "i like ice cream with my apple pie", "i enjoy my apple pie with ice cream"]]
>>> metric.update(candidates, references)
>>> metric.compute()
tensor(0.56377503)

__init__(*, n_gram: int, weights: Optional[Tensor] = None, 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__(*, n_gram[, weights, device])	Initialize a metric object and its internal states.
compute()	Returns the running BLEUScore.
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 new inputs.

Attributes

device

The last input device of Metric.to().