9df673c499
* move docs to /docs/zh * fix * v1 with gpt-4o * add deepseek part1. * add deepseek translate part2. * fix .readthedocs.yaml * fix yaml. * fix readme. * update. * optimizer docs. * refine docs.
220 lines
7.0 KiB
Plaintext
220 lines
7.0 KiB
Plaintext
# SOME DESCRIPTIVE TITLE.
|
||
# Copyright (C) 2021, PaddleNLP
|
||
# This file is distributed under the same license as the PaddleNLP package.
|
||
# FIRST AUTHOR <EMAIL@ADDRESS>, 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 <EMAIL@ADDRESS>\n"
|
||
"Language-Team: LANGUAGE <LL@li.org>\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"
|
||
|
||
#: ../model_zoo/embeddings.md:1
|
||
msgid "PaddleNLP Embedding API"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:3 ../model_zoo/embeddings.md:24
|
||
msgid "介绍"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:4 ../model_zoo/embeddings.md:28
|
||
msgid "用法"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:5 ../model_zoo/embeddings.md:30
|
||
msgid "TokenEmbedding参数"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:6 ../model_zoo/embeddings.md:69
|
||
msgid "初始化"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:7 ../model_zoo/embeddings.md:101
|
||
msgid "查询embedding结果"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:8 ../model_zoo/embeddings.md:115
|
||
msgid "可视化embedding结果"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:9 ../model_zoo/embeddings.md:140
|
||
msgid "计算词向量cosine相似度"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:10 ../model_zoo/embeddings.md:147
|
||
msgid "计算词向量内积"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:11 ../model_zoo/embeddings.md:155
|
||
msgid "训练"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:12 ../model_zoo/embeddings.md:171
|
||
msgid "切词"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:13 ../model_zoo/embeddings.md:183
|
||
msgid "预训练模型"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:14 ../model_zoo/embeddings.md:189
|
||
msgid "中文词向量"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:15 ../model_zoo/embeddings.md:343
|
||
msgid "英文词向量"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:16 ../model_zoo/embeddings.md:345
|
||
msgid "Word2Vec"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:17 ../model_zoo/embeddings.md:362
|
||
msgid "GloVe"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:18 ../model_zoo/embeddings.md:395
|
||
msgid "FastText"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:19 ../model_zoo/embeddings.md:416
|
||
msgid "使用方式"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:20 ../model_zoo/embeddings.md:427
|
||
msgid "模型信息"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:21 ../model_zoo/embeddings.md:751
|
||
msgid "致谢"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:22 ../model_zoo/embeddings.md:756
|
||
msgid "参考论文"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:26
|
||
msgid "PaddleNLP提供多个开源的预训练词向量模型，用户仅需在使用paddlenlp.embeddings.TokenEmbedding时，指定预训练模型的名称，即可加载相对应的预训练模型。以下将介绍TokenEmbeddign详细用法，并列出PaddleNLP所支持的预训练Embedding模型。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:116
|
||
msgid "使用深度学习可视化工具VisualDL的High Dimensional组件可以对embedding结果进行可视化展示，便于对其直观分析，步骤如下："
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:128
|
||
msgid "执行完毕后会在当前路径下生成一个visualize目录，并将日志存放在其中，我们在命令行启动VisualDL即可进行查看，启动命令为："
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:132
|
||
msgid "启动后打开浏览器即可看到可视化结果"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:138
|
||
msgid "使用VisualDL除可视化embedding结果外，还可以对标量、图片、音频等进行可视化，有效提升训练调参效率。关于VisualDL更多功能和详细介绍，可参考VisualDL使用文档。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:157
|
||
msgid ""
|
||
"以下为TokenEmbedding简单的组网使用方法。有关更多TokenEmbedding训练流程相关的使用方法，请参考Word "
|
||
"Embedding with PaddleNLP。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:185
|
||
msgid "以下将列举PaddleNLP支持的Embedding预训练模型。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:186
|
||
msgid "模型命名方式为：${训练模型}.${语料}.${词向量类型}.${co-occurrence type}.dim${维度}。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:187
|
||
msgid "模型有三种，分别是Word2Vec(w2v, skip-gram), GloVe(glove)和FastText(fasttext)。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:191
|
||
msgid "以下预训练词向量由Chinese-Word-Vectors提供。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:193
|
||
msgid "根据不同类型的上下文为每个语料训练多个目标词向量，第二列开始表示不同类型的上下文。以下为上下文类别："
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:195
|
||
msgid "Word表示训练时目标词预测的上下文是一个Word。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:196
|
||
msgid ""
|
||
"Word + "
|
||
"N-gram表示训练时目标词预测的上下文是一个Word或者Ngram，其中bigram表示2-grams，ngram.1-2表示1-gram或者2-grams。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:197
|
||
msgid ""
|
||
"Word + Character表示训练时目标词预测的上下文是一个Word或者Character，其中word-"
|
||
"character.char1-2表示上下文是1个或2个Character。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:198
|
||
msgid ""
|
||
"Word + Character + Ngram表示训练时目标词预测的上下文是一个Word、Character或者Ngram。bigram-"
|
||
"char表示上下文是2-grams或者1个Character。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:284
|
||
msgid "特别地，对于百度百科语料，在不同的 Co-occurrence类型下分别提供了目标词与上下文向量："
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:418
|
||
msgid ""
|
||
"以上所述的模型名称可直接以参数形式传入padddlenlp.embeddings.TokenEmbedding，加载相对应的模型。比如要加载语料为Wiki2017，通过FastText训练的预训练模型（fasttext"
|
||
".wiki-news.target.word-word.dim300.en），只需执行以下代码："
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:752
|
||
msgid "感谢 Chinese-Word-Vectors提供Word2Vec中文预训练词向量。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:753
|
||
msgid "感谢 GloVe Project提供的GloVe英文预训练词向量。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:754
|
||
msgid "感谢 FastText Project提供的英文预训练词向量。"
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:757
|
||
msgid ""
|
||
"Li, Shen, et al. \"Analogical reasoning on chinese morphological and "
|
||
"semantic relations.\" arXiv preprint arXiv:1805.06504 (2018)."
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:758
|
||
msgid ""
|
||
"Qiu, Yuanyuan, et al. \"Revisiting correlations between intrinsic and "
|
||
"extrinsic evaluations of word embeddings.\" Chinese Computational "
|
||
"Linguistics and Natural Language Processing Based on Naturally Annotated "
|
||
"Big Data. Springer, Cham, 2018. 209-221."
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:759
|
||
msgid ""
|
||
"Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. "
|
||
"GloVe: Global Vectors for Word Representation."
|
||
msgstr ""
|
||
|
||
#: ../model_zoo/embeddings.md:760
|
||
msgid ""
|
||
"T. Mikolov, E. Grave, P. Bojanowski, C. Puhrsch, A. Joulin. Advances in "
|
||
"Pre-Training Distributed Word Representations."
|
||
msgstr ""
|
||
|