![](https://img-blog.csdnimg.cn/20201014180756754.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
自然语言处理与机器翻译
文章平均质量分 86
DreamNotOver
这个作者很懒,什么都没留下…
展开
-
USING GIZA++ TO OBTAIN WORD ALIGNMENT BETWEEN BILINGUAL SENTENCES
USING GIZA++ TO OBTAIN WORD ALIGNMENT BETWEEN BILINGUAL SENTENCESPROBLEMWord alignment is mapping of words between two sentences that have the same meaning in two different languages. Let's say we...转载 2018-08-18 00:32:08 · 238 阅读 · 0 评论 -
GIZA++ Installation and Running Tutorial
https://okapiframework.org/wiki/index.php?title=GIZA%2B%2B_Installation_and_Running_TutorialGIZA++ Installation and Running TutorialSteps List and Overview Articles on Pipelines and Steps in the ...转载 2018-09-02 22:19:45 · 550 阅读 · 0 评论 -
Topic Modeling with Gensim (Python)
转自https://www.machinelearningplus.com/nlp/topic-modeling-gensim-python/1. IntroductionOne of the primary applications of natural language processing is to automatically extract what topics people ...转载 2018-09-17 22:45:55 · 6134 阅读 · 1 评论 -
Complete Guide to Topic Modeling
转自https://nlpforhackers.io/topic-modeling/Complete Guide to Topic ModelingWhat is Topic Modeling?Topic modelling, in the context of Natural Language Processing, is described as a method of uncov...转载 2018-09-17 22:45:09 · 603 阅读 · 0 评论 -
aI_Challenger 机器翻译
AI_Challenger 机器翻译官方提供的脚本有不少错误,python脚本从2迁移到了3。 训练脚本中去掉了batch_size项,改用 --worker_gpu_memory_fraction 可以免去内存溢出风险。经尝试,在base、big模式下训练的模型效果很差,只有hparams_set=transformer_base_single_gpu 时,训练的模型效果好。最近在客户提供的...原创 2018-12-30 22:00:12 · 1284 阅读 · 3 评论 -
导出tensor2tensor模型示例
pip install tensorflow-serving-apiyum install tree -y自定义问题路径 --t2t_usr_dir=./ai_data(fastai) [root@datanode2 t2t]# tree ai_dataai_data├── __init__.py└── problem.py0 directories, 2 files...原创 2019-01-02 18:21:53 · 1232 阅读 · 1 评论