jd-aig / nlp_baai Goto Github PK
View Code? Open in Web Editor NEWNLP models and codes for BAAI-JD joint project.
NLP models and codes for BAAI-JD joint project.
你好,看到你们介绍有提供数据集,但是2020年的数据集在哪里下载哈?可以提供下下载链接吗
比赛模型文件过大,几个G的大小。使用贵比赛提供的“提交评审”页面中“上传文件”按钮,选择了模型文件并上传后,无回复反应。请问该如何处理呢?
model = model_class.from_pretrained(args.model_name_or_path, from_tf=bool('.ckpt' in args.model_name_or_path), config=config)
,在进行检索baseline model调试过程中出现预训练模型加载时出现Segmentation fault (core dumped)。
transformer version:2.9.1, 2.11.0 都会出现这样的错误
pytorch version: 1.3.1
File "raw_data_preprocess.py", line 279, in <module>
train_items, dev_items = do_preprocess(args.directory, args.sess_turns)
File "raw_data_preprocess.py", line 81, in do_preprocess
waiter = word[4]
难道我们用的不是同一个训练文件吗?我用的是data_train.txt,这个文件里面商品编号
这一列有部分是空的
你好! 感谢提供baseline model
官网中对于问题的描述是这样的
输入(上下文信息,用户问题):此轮对话用户提出的问题Qn,其中Qn可能为纯文本、纯图片的问题或图文相结合的问题;此轮对话前n-1轮的对话历史信息,C={Q0, A0, Q1, A1, …, Qn-1, An-1},历史对话信息也可能为多模态形式。 输出(文本预测答案):根据输入信息,输出满足第n轮用户问题Qn所期望的答案,该答案应该是通顺、逻辑一致且含有丰富知识的文本回答。
我在阅读代码时发现
input_conversations = [conv[:-1] for conv in conversations] target_conversations = [conv[1:] for conv in conversations]
即,baseline model在train、evaluate以及test过程中,不仅需要根据{Q0, A0, Q1, A1, …, Qn-1, An-1,Qn}预测An,
还需要分别根据{Q0}预测A0,根据{Q0,A0}预测Q1,根据{Q0,A0,Q1}预测A1,...,根据{Q0,A0,...,Qn-1,An-1}预测Qn?
这与paper中的描述似乎不太一样? 并且模型在训练过程中还需要预测客户问题,不太清楚这是否合理
点击链接,提示访问权限被限制,不知道是否还开放下载权限?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.