Skip to content

This repository is a class project for the course Nature Language Process, Fall 2018, CS@NJU

Notifications You must be signed in to change notification settings

zzzzwj/BiLSTM-CRF-for-Chinese-NER

Repository files navigation

Usage

If you wanna to train a model, run the code below:
python main.py --train_data data/train_vec.txt --dev_data data/dev_vec.txt --test_data data/test_vec.txt --max_epoch 30 --lr 0.0001 --dropout 0.2 --batch_size 128 --mode train

And if you wanna predict your own data, change the --mode to 'test'. The sample data is in the directory ./data.

Running environment

Python 3.6
tensorflow 1.8

Ackonwledgement

This repository is a simple version of https://github.com/Determined22/zh-NER-TF. If you want to know more about BiLSTM-CRF on Chinese NER task, you should visit that repository.

About

This repository is a class project for the course Nature Language Process, Fall 2018, CS@NJU

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published