o .wi$@sddlmZddlmZmZmZmZddlmZddl m Z ddl m Z ddl mZmZmZmZddlmZmZddlmZmZesEd gZd d giZGd d d e ZdS))Sequence)AnyCallableOptionalUnion)Tensor)Literal)Metric)ALLOWED_ACCUMULATE_VALUESALLOWED_ROUGE_KEYS_rouge_score_compute_rouge_score_update)_MATPLOTLIB_AVAILABLE_NLTK_AVAILABLE)_AX_TYPE_PLOT_OUT_TYPEROUGEScore.plot) ROUGEScorenltkcsFeZdZUdZdZeed<dZeed<dZeed<dZ e ed<d Z e ed <  d%dede e egefde e egeefdeddeeeedffdedd ffdd Zdeeeefdeeeeeeefdd fddZdeeeffddZdefdd Z d&d!e eeeefd"e edefd#d$ZZS)'ra `Calculate Rouge Score`_, used for automatic summarization. This implementation should imitate the behaviour of the ``rouge-score`` package `Python ROUGE Implementation` As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~Sequence`): An iterable of predicted sentences or a single predicted sentence - ``target`` (:class:`~Sequence`): An iterable of target sentences or an iterable of interables of target sentences or a single target sentence As output of ``forward`` and ``compute`` the metric returns the following output: - ``rouge`` (:class:`~Dict`): A dictionary of tensor rouge scores for each input str rouge key Args: use_stemmer: Use Porter stemmer to strip word suffixes to improve matching. normalizer: A user's own normalizer function. If this is ``None``, replacing any non-alpha-numeric characters with spaces is default. This function must take a ``str`` and return a ``str``. tokenizer: A user's own tokenizer function. If this is ``None``, splitting by spaces is default This function must take a ``str`` and return ``Sequence[str]`` accumulate: Useful in case of multi-reference rouge score. - ``avg`` takes the avg of all references with respect to predictions - ``best`` takes the best fmeasure score obtained between prediction and multiple corresponding references. rouge_keys: A list of rouge types to calculate. Keys that are allowed are ``rougeL``, ``rougeLsum``, and ``rouge1`` through ``rouge9``. kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Example: >>> from torchmetrics.text.rouge import ROUGEScore >>> preds = "My name is John" >>> target = "Is your name John" >>> rouge = ROUGEScore() >>> from pprint import pprint >>> pprint(rouge(preds, target)) {'rouge1_fmeasure': tensor(0.7500), 'rouge1_precision': tensor(0.7500), 'rouge1_recall': tensor(0.7500), 'rouge2_fmeasure': tensor(0.), 'rouge2_precision': tensor(0.), 'rouge2_recall': tensor(0.), 'rougeL_fmeasure': tensor(0.5000), 'rougeL_precision': tensor(0.5000), 'rougeL_recall': tensor(0.5000), 'rougeLsum_fmeasure': tensor(0.5000), 'rougeLsum_precision': tensor(0.5000), 'rougeLsum_recall': tensor(0.5000)} Raises: ValueError: If the python packages ``nltk`` is not installed. ValueError: If any of the ``rouge_keys`` does not belong to the allowed set of keys. Fis_differentiableThigher_is_betterfull_state_updategplot_lower_boundg?plot_upper_boundNbestrouge1rouge2rougeL rougeLsum use_stemmer normalizer tokenizer accumulate)avgr rouge_keys.kwargsreturnc  stjd i||sd|vrtstdddl}t|ts!|f}|D]}|tvr3td|dtq#|t vrBtd|dt ||_ dd|D|_ |rU|j j nd|_||_||_||_|j D]} d D]} |j| d | gdd qhqddS) NrzUStemmer and/or `rougeLsum` requires that `nltk` is installed. Use `pip install nltk`.rzGot unknown rouge key z. Expected to be one of zGot unknown accumulate value cSsg|]}t|qS)r ).0keyr(r(T/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/torchmetrics/text/rouge.py sz'ROUGEScore.__init__..fmeasure precisionrecall_)dist_reduce_fxr()super__init__rModuleNotFoundErrorr isinstancetupler ValueErrorr r%rouge_keys_valuesstemporter PorterStemmerstemmerr!r"r# add_state) selfr r!r"r#r%r&rr* rouge_keyscore __class__r(r+r4js8   zROUGEScore.__init__predstargetc Cst|trtdd|Drt|tr|gndd|D}t|tr%|g}t|tr.|gg}t|||j|j|j|j|j d}| D]$\}}|D]}| D]\}}t |d|d| | |jqNqHqBdS) z*Update state with predictions and targets.css|]}t|tVqdS)N)r6strr)tgtr(r(r+ sz$ROUGEScore.update..cSsg|]}|gqSr(r(rGr(r(r+r,sz%ROUGEScore.update..)r=r!r"r#rouger1N)r6listallrFr r9r=r!r"r#itemsgetattrappendtodevice) r?rDrEoutputr@metricsmetrictpvaluer(r(r+updates,   &zROUGEScore.updatecCsHi}|jD]}dD]}t|d|d||d|d|<q qt|S)zCCalculate (Aggregate and provide confidence intervals) ROUGE score.r-rJr1)r9rNr )r? update_outputr@rUr(r(r+computes  (zROUGEScore.computecCsH|jjg}|jD]}t||}t|trt|}||qtt|S)z>Return a unique hash for the specific instance of this metric.) rC__name__ _defaultsrNr6rKr7rOhash)r? hash_valsr*rVr(r(r+__hash__s      zROUGEScore.__hash__valaxcCs |||S)aFPlot 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.rouge import ROUGEScore >>> metric = ROUGEScore() >>> preds = "My name is John" >>> target = "Is your name John" >>> metric.update(preds, target) >>> fig_, ax_ = metric.plot() .. plot:: :scale: 75 >>> # Example plotting multiple values >>> from torchmetrics.text.rouge import ROUGEScore >>> metric = ROUGEScore() >>> preds = "My name is John" >>> target = "Is your name John" >>> values = [ ] >>> for _ in range(10): ... values.append(metric(preds, target)) >>> fig_, ax_ = metric.plot(values) )_plot)r?r_r`r(r(r+plots *r)FNNrr)NN)rZ __module__ __qualname____doc__rbool__annotations__rrrfloatrrrrFrrrr7rr4rWdictrrYintr^rrrb __classcell__r(r(rBr+r%sX  >    (  rN)collections.abcrtypingrrrrtorchrtyping_extensionsr torchmetricsr "torchmetrics.functional.text.rouger r r r torchmetrics.utilities.importsrrtorchmetrics.utilities.plotrr__doctest_skip____doctest_requires__rr(r(r(r+s