a  d@sddlmZmZmZmZddlmZddlmZddl m Z ddl m Z ddl mZddlmZmZddlmZmZes~d gZd ZGd d d eZd S))AnyOptionalSequenceUnion)Tensor)Literal)_bleu_score_update)_SacreBLEUTokenizer) BLEUScore)_MATPLOTLIB_AVAILABLE_REGEX_AVAILABLE)_AX_TYPE_PLOT_OUT_TYPESacreBLEUScore.plotnone13azhintlcharc seZdZUdZdZeed<dZeed<dZeed<dZ e ed<d Z e ed <de ee deeee ed dfdd Zeeeeed dddZdeeeeefeeedddZZS)SacreBLEUScoreaCalculate `BLEU score`_ of machine translated text with one or more references. This implementation follows the behaviour of `SacreBLEU`_. The SacreBLEU implementation differs from the NLTK BLEU implementation in tokenization techniques. As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~Sequence`): An iterable of machine translated corpus - ``target`` (:class:`~Sequence`): An iterable of iterables of reference corpus As output of ``forward`` and ``compute`` the metric returns the following output: - ``sacre_bleu`` (:class:`~torch.Tensor`): A tensor with the SacreBLEU Score Args: n_gram: Gram value ranged from 1 to 4 smooth: Whether to apply smoothing, see `SacreBLEU`_ tokenize: Tokenization technique to be used. Supported tokenization: ``['none', '13a', 'zh', 'intl', 'char']`` lowercase: If ``True``, BLEU score over lowercased text is calculated. kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. weights: Weights used for unigrams, bigrams, etc. to calculate BLEU score. If not provided, uniform weights are used. Raises: ValueError: If ``tokenize`` not one of 'none', '13a', 'zh', 'intl' or 'char' ValueError: If ``tokenize`` is set to 'intl' and `regex` is not installed ValueError: If a length of a list of weights is not ``None`` and not equal to ``n_gram``. Example: >>> from torchmetrics.text import SacreBLEUScore >>> preds = ['the cat is on the mat'] >>> target = [['there is a cat on the mat', 'a cat is on the mat']] >>> sacre_bleu = SacreBLEUScore() >>> sacre_bleu(preds, target) tensor(0.7598) Additional References: - Automatic Evaluation of Machine Translation Quality Using Longest Common Subsequence and Skip-Bigram Statistics by Chin-Yew Lin and Franz Josef Och `Machine Translation Evolution`_ Fis_differentiableThigher_is_betterfull_state_updategplot_lower_boundg?plot_upper_boundrNr)n_gramsmoothtokenize lowercaseweightskwargsreturnc s\tjf|||d||tvr8tdtd|d|dkrLtsLtdt|||_dS)N)rrr!z*Argument `tokenize` expected to be one of z but got .rzv`'intl'` tokenization requires that `regex` is installed. Use `pip install regex` or `pip install torchmetrics[text]`.)super__init__AVAILABLE_TOKENIZERS ValueErrorr ModuleNotFoundErrorr tokenizer)selfrrrr r!r" __class__e/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/torchmetrics/text/sacre_bleu.pyr&^s  zSacreBLEUScore.__init__)predstargetr#c Cs.t|||j|j|j|j|j|j\|_|_dS)z*Update state with predictions and targets.N)r numerator denominatorZ preds_lenZ target_lenrr*)r+r0r1r.r.r/updaterszSacreBLEUScore.update)valaxr#cCs |||S)aPlot a single or multiple values from the metric. Args: val: Either a single result from calling `metric.forward` or `metric.compute` or a list of these results. If no value is provided, will automatically call `metric.compute` and plot that result. ax: An matplotlib axis object. If provided will add plot to that axis Returns: Figure and Axes object Raises: ModuleNotFoundError: If `matplotlib` is not installed .. plot:: :scale: 75 >>> # Example plotting a single value >>> from torchmetrics.text import SacreBLEUScore >>> metric = SacreBLEUScore() >>> preds = ['the cat is on the mat'] >>> target = [['there is a cat on the mat', 'a cat is on the mat']] >>> metric.update(preds, target) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> from torchmetrics.text import SacreBLEUScore >>> metric = SacreBLEUScore() >>> preds = ['the cat is on the mat'] >>> target = [['there is a cat on the mat', 'a cat is on the mat']] >>> values = [ ] >>> for _ in range(10): ... values.append(metric(preds, target)) >>> fig_, ax_ = metric.plot(values) )Z_plot)r+r5r6r.r.r/plots*r)rFrFN)NN)__name__ __module__ __qualname____doc__rbool__annotations__rrrfloatrintrrrrr&strr4rrr rr7 __classcell__r.r.r,r/r&s4 1      rN)typingrrrrZtorchrZtyping_extensionsrZ!torchmetrics.functional.text.bleurZ'torchmetrics.functional.text.sacre_bleur Ztorchmetrics.text.bleur Ztorchmetrics.utilities.importsr r Ztorchmetrics.utilities.plotr rZ__doctest_skip__r'rr.r.r.r/s