Skip to content

An Analysis Tool to Models for Chinese Spell Checking Released on ACL2023.

Notifications You must be signed in to change notification settings

piglaker/ConfusionCluster

Folders and files

NameName
Last commit message
Last commit date

Latest commit

May 11, 2023
3f6200c · May 11, 2023

History

31 Commits
May 10, 2023
May 3, 2022
May 10, 2023
May 11, 2023
Apr 21, 2022
Apr 21, 2022
Jun 25, 2022
Sep 1, 2022
May 7, 2023
May 3, 2022
Sep 1, 2022
May 4, 2022
Sep 1, 2022
May 7, 2023
May 4, 2022
May 7, 2023
Aug 25, 2022

Repository files navigation

ConfusionCluster

分析及CCCR Probe 指标对于ACL 2023 (Findings): Investigating Glyph Phonetic Information for Chinese Spell Checking: What Works and What's Next

训练纠错模型的代码及训练数据见另一github仓库:SpecialEdition 论文中的Probe实验一见github仓库:

Environment setting

python >= 3.7

conda create -n ctcSE python=3.7
conda activate ctcSE

pip install -r requirements.txt

Quick Start

将main.py中路径设置为你要测试的纠错模型的路径。

bash run.sh

⚠️注意最好该模型是huggingface对象输出能够返回logits用于分析,否则请参考本项目中对ReaLiSe的处理,提前运行一次inference将模型logits保存到文件再读入。

About

An Analysis Tool to Models for Chinese Spell Checking Released on ACL2023.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published