https://github.com/guitaowufeng/LR-CNN

https://github.com/guitaowufeng/LR-CNN

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/guitaowufeng/LR-CNN
• Owner: guitaowufeng
• Fork: true (rtmaww/LR-CNN)
• Created: 2019-05-30T07:39:00.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2019-05-30T07:36:32.000Z (about 5 years ago)
• Last Synced: 2024-01-17T00:57:14.584Z (5 months ago)
• Language: Python
• Size: 0 Bytes
• Stars: 29
• Watchers: 1
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Lists

• awesome-stars-copy - guitaowufeng/LR-CNN - (Python)

README

        
# LR-CNN

Source codes for paper:

CNN-Based Chinese NER with Lexicon Rethinking  

Requirement:

======

Python 3.6

Pytorch 0.4.1  

Input format:

======

CoNLL format, with each character and its label splited by a whitespace in a line. The "BMES" tag scheme is prefered.

	别 O  

	错 O  

	过 O  

	邻 O  

	近 O  

	大 B-LOC  

	鹏 M-LOC  

	湾 E-LOC  

	的 O  

	湿 O  

	地 O  

Pretrain embedding:

======

The pretrained embeddings(word embedding, char embedding) are the same with Lattice LSTM(https://github.com/jiesutd/LatticeLSTM)  

Run the code:

======

1. Download the character embeddings and word embeddings and put them in the `data` folder.

2. To train/test the demo: `sh train.sh` / `sh test.sh`

3. To train/test your own data: modify the 'train.sh' or 'test.sh' file with your file path, and run the shell file.