GithubHelp home page GithubHelp logo

dcn's People

Contributors

destwang avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

dcn's Issues

如何测试模型

您好,在运行sh train.sh了训练模型完成之后,如何在SIGHAN上测试模型得到您展示的结果呢?

How to download the datasets?

The dataset link need username and code, but i only see the code.
Can you provide the username to download the train set?
Thank you.

report a potential bug

In the file transformers/modeling_dcn.py, line 550.

self.word_pinyin_idx = torch.tensor(self.word_pinyin_idx, device="cuda")

I think device="cuda" would cause lots of bugs. Maybe you can fix it, or maybe I'm wrong. Thanks.

关于运行环境

您好,我这边本地测试,用的2080TI,12G显存,batch size只能设置为6才不会OOM,这正常吗?您这边用的什么机器可以跑bz=32?

Truth中存在一些错误

您好,感谢您公开了代码,我按照您的流程跑了一下,只是batch_size设置不同,到那时结果大概低一个百分点;另外分析sighan15的测试集结果,发现Truth数据集中有一些错误,比如: (pid=A2-1311-2) 这女人告所他的大学在那里。 这里预测出来告诉,但是truth中实际上是没有这个错误的,所以数据集是没有经过清洗是吗?

训练代码报错

Traceback (most recent call last):
File "train_DCN.py", line 511, in
main()
File "train_DCN.py", line 228, in main
model = DCNForMaskedLM.from_pretrained(
File "/home/zhaojianhui/DCN-main/transformers/modeling_utils.py", line 671, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
File "/home/zhaojianhui/DCN-main/transformers/modeling_dcn.py", line 1124, in init
self.cls = BertOnlyMLMHead(config)
File "/home/zhaojianhui/DCN-main/transformers/modeling_dcn.py", line 586, in init
self.predictions = BertLMPredictionHead(config)
File "/home/zhaojianhui/DCN-main/transformers/modeling_dcn.py", line 543, in init
self.pinyin_embeddings = nn.Embedding(config.pinyin_vocab_size,
AttributeError: 'BertConfig' object has no attribute 'pinyin_vocab_size'

预训练DCN的数据集问题

论文中的最好表现是需要在wiki data上预训练,这部分预训练具体是咋做的,数据是哪些,可以答疑解惑一下么。。

sighan15数据集无法实现论文效果,略低

********* Evaluation Test sighan15 Sentence-level ************
1100 1100
detect_sent_precision=0.756432, detect_sent_recall=0.801818, detect_Fscore=0.778464
correct_sent_precision=0.728988, correct_sent_recall=0.772727, correct_Fscore=0.750221

结果接近没有预训练的Roberta-DCN,batch_size 为30.

数据集问题

image
您提供的 train.txt 是上图的四个数据集吗? 这四者加起来有 277,803 条数据,但是 train.txt 有 28 万多,请问多出来的哪里的呢?期待回复

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.