# SOME DESCRIPTIVE TITLE. # Copyright (C) 2021, PaddleNLP # This file is distributed under the same license as the PaddleNLP package. # FIRST AUTHOR , 2022. # #, fuzzy msgid "" msgstr "" "Project-Id-Version: PaddleNLP \n" "Report-Msgid-Bugs-To: \n" "POT-Creation-Date: 2022-03-18 21:31+0800\n" "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n" "Last-Translator: FULL NAME \n" "Language-Team: LANGUAGE \n" "MIME-Version: 1.0\n" "Content-Type: text/plain; charset=utf-8\n" "Content-Transfer-Encoding: 8bit\n" "Generated-By: Babel 2.9.0\n" #: ../source/paddlenlp.metrics.bleu.rst:2 msgid "bleu" msgstr "" #: of paddlenlp.metrics.bleu.BLEU:1 msgid "基类：:class:`paddle.metric.metrics.Metric`" msgstr "" #: of paddlenlp.metrics.bleu.BLEU:1 msgid "" "BLEU (bilingual evaluation understudy) is an algorithm for evaluating the" " quality of text which has been machine-translated from one natural " "language to another. This metric uses a modified form of precision to " "compare a candidate translation against multiple reference translations." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:6 msgid "" "BLEU could be used as `paddle.metric.Metric` class, or an ordinary class." " When BLEU is used as `paddle.metric.Metric` class. A function is needed " "that transforms the network output to reference string list, and " "transforms the label to candidate string. By default, a default function " "`default_trans_func` is provided, which gets target sequence id by " "calculating the maximum probability of each step. In this case, user must" " provide `vocab`. It should be noted that the BLEU here is different from" " the BLEU calculated in prediction, and it is only for observation during" " training and evaluation." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:16 msgid "" "BP & =\n" "\\begin{cases}\n" "1, & \\text{if }c>r \\\\\n" "e_{1-r/c}, & \\text{if }c\\leq r\n" "\\end{cases}\n" "\n" "BLEU & = BP\\exp(\\sum_{n=1}^N w_{n} \\log{p_{n}})" msgstr "" #: of paddlenlp.metrics.bleu.BLEU:26 msgid "" "where `c` is the length of candidate sentence, and `r` is the length of " "reference sentence." msgstr "" #: of paddlenlp.metrics.bleu.BLEU paddlenlp.metrics.bleu.BLEU.add_inst #: paddlenlp.metrics.bleu.BLEUForDuReader #: paddlenlp.metrics.bleu.BLEUForDuReader.add_inst msgid "参数" msgstr "" #: of paddlenlp.metrics.bleu.BLEU:28 msgid "`trans_func` transforms the network output to string to calculate." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:31 msgid "" "Vocab for target language. If `trans_func` is None and BLEU is used as " "`paddle.metric.Metric` instance, `default_trans_func` will be performed " "and `vocab` must be provided." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:36 paddlenlp.metrics.bleu.BLEUForDuReader:5 msgid "Number of gram for BLEU metric. Defaults to 4." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:38 msgid "The weights of precision of each gram. Defaults to None." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:41 msgid "Name of `paddle.metric.Metric` instance. Defaults to \"bleu\"." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:46 msgid "实际案例" msgstr "" #: of paddlenlp.metrics.bleu.BLEU:47 msgid "Using as a general evaluation object." msgstr "" #: of paddlenlp.metrics.bleu.BLEU:58 msgid "Using as an instance of `paddle.metric.Metric`." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.update:1 msgid "Update states for metric" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.update:3 msgid "" "Inputs of :code:`update` is the outputs of :code:`Metric.compute`, if " ":code:`compute` is not defined, the inputs of :code:`update` will be " "flatten arguments of **output** of mode and **label** from data: " ":code:`update(output1, output2, ..., label1, label2,...)`" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.update:8 msgid "see :code:`Metric.compute`" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.add_inst:1 #: paddlenlp.metrics.bleu.BLEUForDuReader.add_inst:1 msgid "Update the states based on a pair of candidate and references." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.add_inst:3 #: paddlenlp.metrics.bleu.BLEUForDuReader.add_inst:3 msgid "Tokenized candidate sentence." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.add_inst:5 #: paddlenlp.metrics.bleu.BLEUForDuReader.add_inst:5 msgid "List of tokenized ground truth sentences." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.reset:1 msgid "Reset states and result" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.accumulate:1 msgid "Calculates and returns the final bleu metric." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.accumulate msgid "返回" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.accumulate:3 msgid "Returns the accumulated metric `bleu` and its data type is float64." msgstr "" #: of paddlenlp.metrics.bleu.BLEU.accumulate msgid "返回类型" msgstr "" #: of paddlenlp.metrics.bleu.BLEU.name:1 msgid "Returns metric name" msgstr "" #: of paddlenlp.metrics.bleu.BLEUForDuReader:1 msgid "基类：:class:`paddlenlp.metrics.bleu.BLEU`" msgstr "" #: of paddlenlp.metrics.bleu.BLEUForDuReader:1 msgid "BLEU metric with bonus for DuReader contest." msgstr "" #: of paddlenlp.metrics.bleu.BLEUForDuReader:3 msgid "" "Please refer to `DuReader " "Homepage`_ for " "more details." msgstr "" #: of paddlenlp.metrics.bleu.BLEUForDuReader:7 msgid "" "Weight of YesNo dataset when adding bonus for DuReader contest. Defaults " "to 1.0." msgstr "" #: of paddlenlp.metrics.bleu.BLEUForDuReader:9 msgid "" "Weight of Entity dataset when adding bonus for DuReader contest. Defaults" " to 1.0." msgstr ""