xiaolinandy / csds Goto Github PK
View Code? Open in Web Editor NEWThis is the official repo for paper "CSDS: A Fine-grained Chinese Dataset for Customer Service Dialogue Summarization", accepted by EMNLP 2021
This is the official repo for paper "CSDS: A Fine-grained Chinese Dataset for Customer Service Dialogue Summarization", accepted by EMNLP 2021
您好,能不能分享一下原始的JDDC对话数据集,他们那篇论文里的链接打不开了
运行 run.sh 出现如题报错,请问该怎样解决
你好,有没有将自己数据集处理成CSDS数据集格式的代码呢?
您好,截止目前,TX提供的version v0.2.0一共有四个版本。
Dimension 200 / small | Large
Dimension 100 / small | Large
请问您实际使用的是哪个版本?
作者您好:
在csds数据集中有”intent“字段,但论文中和代码中好像没有看到对其相关的使用,请问这方面有可以参考的内容吗
Instruction for BERT (BERTExt/BERTAbs)
1.Go to the models/BERT/ directory.
2.Download the CSDS dataset, create a new folder named data/ and put CSDS under the data/ folder.
3.Download the pretrained BERT model, create a new folder named bert_base_chinese/ and put it into the bert_base_chinese/ folder.
4.Run the bash file run.sh to train and test.
请问,报错应该是与step 3有较大的关系吧,怎样下载?到哪里下载?
感谢大佬的git,我这里训练完TDS+SATM*,想请问一下几个数据集的问题
1.rouge-f的分数有些可以高到0.7879/0.6517/0.7220,但有些低到只有0.3947/0.1647/0.3926,比较两个对话,分数低的turn较多,所以turn越多效果会越差? 还是有其他原因?
2.承1 turn太多有些是這樣
(26) 【客户】 你 们 已 经 有 邮 件 告 诉 我 了
(27) 【客户】 就 是 告 诉 我 在 清 关
(28) 【客户】 你 们 不 告 诉 怎 么 清 关
(29) 【客户】 让 我 做 什 么 也 不 说
(30) 【客户】 然 后 又 是 一 个 礼 拜 过 去 了
(31) 【客户】 我 也 是 有 耐 心 的
从26~31都是客户,感觉有些turn是可以合并的,如果合并是否能提升效果?
或是有什么办法改善这种对话数据?
It seems that for PGN and Fast-RL, the max target length is 84 words, while for BERTAbs, the max target length is 84 characters. If so, this issue may result in the difference of ROUGE and BLEU metrics.
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.