English | 简体中文

A Deep Learning Based Knowledge Extraction Toolkit
for Knowledge Graph Construction

DeepKE is a knowledge extraction toolkit for knowledge graph construction supporting cnSchema，low-resource, document-level and multimodal scenarios for entity, relation and attribute extraction. We provide documents, Google Colab tutorials, online demo, and slides for beginners.

Reading materials:

Data-Efficient Knowledge Graph Construction, 高效知识图谱构建 (Tutorial on CCKS 2022) [slides]

What's New
Prediction Demo
Model Framework
Quick Start
- Requirements
- Introduction of Three Functions
  - 1. Named Entity Recognition
  - 2. Relation Extraction
  - 3. Attribute Extraction
Notebook Tutorial
Tips
To do
Citation
Developers

What's New

Aug, 2022

We have added data augmentation (Chinese, English) support for low-resource relation extraction.

June, 2022

We have added multimodal support for entity and relation extraction.

May, 2022

We have released DeepKE-cnschema with off-the-shelf knowledge extraction models.

Jan, 2022

We have released a paper DeepKE: A Deep Learning Based Knowledge Extraction Toolkit for Knowledge Base Population

Dec, 2021

We have added dockerfile to create the enviroment automatically.

Nov, 2021

The demo of DeepKE, supporting real-time extration without deploying and training, has been released.
The documentation of DeepKE, containing the details of DeepKE such as source codes and datasets, has been released.

Oct, 2021

pip install deepke
The codes of deepke-v2.0 have been released.

Aug, 2019

The codes of deepke-v1.0 have been released.

Aug, 2018

The project DeepKE startup and codes of deepke-v0.1 have been released.

Prediction Demo

There is a demonstration of prediction.

Model Framework

DeepKE contains a unified framework for named entity recognition, relation extraction and attribute extraction, the three knowledge extraction functions.
Each task can be implemented in different scenarios. For example, we can achieve relation extraction in standard, low-resource (few-shot), document-level and multimodal settings.
Each application scenario comprises of three components: Data including Tokenizer, Preprocessor and Loader, Model including Module, Encoder and Forwarder, Core including Training, Evaluation and Prediction.

Quick Start

DeepKE supports pip install deepke.
Take the fully supervised relation extraction for example.

Step1 Download the basic code

git clone https://github.com/zjunlp/DeepKE.git

Step2 Create a virtual environment using Anaconda and enter it.

We also provide dockerfile source code, which is located in the docker folder, to help users create their own mirrors.

conda create -n deepke python=3.8

conda activate deepke

Install DeepKE with source code

python setup.py install

python setup.py develop

Install DeepKE with pip
```
pip install deepke
```

Step3 Enter the task directory

cd DeepKE/example/re/standard

Step4 Download the dataset

wget 120.27.214.45/Data/re/standard/data.tar.gz

tar -xzvf data.tar.gz

Step5 Training (Parameters for training can be changed in the conf folder)

We support visual parameter tuning by using wandb.

python run.py

Step6 Prediction (Parameters for prediction can be changed in the conf folder)

Modify the path of the trained model in predict.yaml.

python predict.py

Requirements

python == 3.8

torch == 1.5
hydra-core == 1.0.6
tensorboard == 2.4.1
matplotlib == 3.4.1
transformers == 3.4.0
jieba == 0.42.1
scikit-learn == 0.24.1
pytorch-transformers == 1.2.0
seqeval == 1.2.2
tqdm == 4.60.0
opt-einsum==3.3.0
wandb==0.12.7
ujson

Introduction of Three Functions

1. Named Entity Recognition

Named entity recognition seeks to locate and classify named entities mentioned in unstructured text into pre-defined categories such as person names, organizations, locations, organizations, etc.

The data is stored in .txt files. Some instances as following:

Sentence	Person	Location	Organization
本报北京9月4日讯记者杨涌报道：部分省区人民日报宣传发行工作座谈会9月3日在4日在京举行。	杨涌	北京	人民日报
《红楼梦》是中央电视台和中国电视剧制作中心根据中国古典文学名著《红楼梦》摄制于1987年的一部古装连续剧，由王扶林导演，周汝昌、王蒙、周岭等多位红学家参与制作。	王扶林，周汝昌，王蒙，周岭	中国	中央电视台，中国电视剧制作中心
秦始皇兵马俑位于陕西省西安市，1961年被国务院公布为第一批全国重点文物保护单位，是世界八大奇迹之一。	秦始皇	陕西省，西安市	国务院

Read the detailed process in specific README
- STANDARD (Fully Supervised)
  
  Step1 Enter DeepKE/example/ner/standard. Download the dataset.
```
wget 120.27.214.45/Data/ner/standard/data.tar.gz

tar -xzvf data.tar.gz
```
  Step2 Training
  
  The dataset and parameters can be customized in the data folder and conf folder respectively.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```
- FEW-SHOT
  
  Step1 Enter DeepKE/example/ner/few-shot. Download the dataset.
```
wget 120.27.214.45/Data/ner/few_shot/data.tar.gz

tar -xzvf data.tar.gz
```
  Step2 Training in the low-resouce setting
  
  The directory where the model is loaded and saved and the configuration parameters can be cusomized in the conf folder.
```
python run.py +train=few_shot
```
  Users can modify load_path in conf/train/few_shot.yaml to use existing loaded model.
  
  Step3 Add - predict to conf/config.yaml, modify loda_path as the model path and write_path as the path where the predicted results are saved in conf/predict.yaml, and then run python predict.py
```
python predict.py
```
- MULTIMODAL
  
  Step1 Enter DeepKE/example/ner/multimodal. Download the dataset.
```
wget 120.27.214.45/Data/ner/multimodal/data.tar.gz

tar -xzvf data.tar.gz
```
  We use RCNN detected objects and visual grounding objects from original images as visual local information, where RCNN via faster_rcnn and visual grounding via onestage_grounding.
  
  Step2 Training in the multimodal setting
  - The dataset and parameters can be customized in the data folder and conf folder respectively.
  - Start with the model trained last time: modify load_path in conf/train.yamlas the path where the model trained last time was saved. And the path saving logs generated in training can be customized by log_dir.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```

2. Relation Extraction

Relationship extraction is the task of extracting semantic relations between entities from a unstructured text.

The data is stored in .csv files. Some instances as following:

Sentence	Relation	Head	Head_offset	Tail	Tail_offset
《岳父也是爹》是王军执导的电视剧，由马恩然、范明主演。	导演	岳父也是爹	1	王军	8
《九玄珠》是在纵横中文网连载的一部小说，作者是龙马。	连载网站	九玄珠	1	纵横中文网	7
提起杭州的美景，西湖总是第一个映入脑海的词语。	所在城市	西湖	8	杭州	2

Read the detailed process in specific README
- STANDARD (Fully Supervised)
  
  Step1 Enter the DeepKE/example/re/standard folder. Download the dataset.
```
wget 120.27.214.45/Data/re/standard/data.tar.gz

tar -xzvf data.tar.gz
```
  Step2 Training
  
  The dataset and parameters can be customized in the data folder and conf folder respectively.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```
- FEW-SHOT
  
  Step1 Enter DeepKE/example/re/few-shot. Download the dataset.
```
wget 120.27.214.45/Data/re/few_shot/data.tar.gz

tar -xzvf data.tar.gz
```
  Step 2 Training
  - The dataset and parameters can be customized in the data folder and conf folder respectively.
  - Start with the model trained last time: modify train_from_saved_model in conf/train.yamlas the path where the model trained last time was saved. And the path saving logs generated in training can be customized by log_dir.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```
- DOCUMENT
  
  Step1 Enter DeepKE/example/re/document. Download the dataset.
```
wget 120.27.214.45/Data/re/document/data.tar.gz

tar -xzvf data.tar.gz
```
  Step2 Training
  - The dataset and parameters can be customized in the data folder and conf folder respectively.
  - Start with the model trained last time: modify train_from_saved_model in conf/train.yamlas the path where the model trained last time was saved. And the path saving logs generated in training can be customized by log_dir.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```
- MULTIMODAL
  
  Step1 Enter DeepKE/example/re/multimodal. Download the dataset.
```
wget 120.27.214.45/Data/re/multimodal/data.tar.gz

tar -xzvf data.tar.gz
```
  We use RCNN detected objects and visual grounding objects from original images as visual local information, where RCNN via faster_rcnn and visual grounding via onestage_grounding.
  
  Step2 Training
  - The dataset and parameters can be customized in the data folder and conf folder respectively.
  - Start with the model trained last time: modify load_path in conf/train.yamlas the path where the model trained last time was saved. And the path saving logs generated in training can be customized by log_dir.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```

3. Attribute Extraction

Attribute extraction is to extract attributes for entities in a unstructed text.

The data is stored in .csv files. Some instances as following:

Sentence	Att	Ent	Ent_offset	Val	Val_offset
张冬梅，女，汉族，1968年2月生，河南淇县人	民族	张冬梅	0	汉族	6
诸葛亮，字孔明，三国时期杰出的军事家、文学家、发明家。	朝代	诸葛亮	0	三国时期	8
2014年10月1日许鞍华执导的电影《黄金时代》上映	上映时间	黄金时代	19	2014年10月1日	0

Read the detailed process in specific README
- STANDARD (Fully Supervised)
  
  Step1 Enter the DeepKE/example/ae/standard folder. Download the dataset.
```
wget 120.27.214.45/Data/ae/standard/data.tar.gz

tar -xzvf data.tar.gz
```
  Step2 Training
  
  The dataset and parameters can be customized in the data folder and conf folder respectively.
```
python run.py
```
  Step3 Prediction
```
python predict.py
```

Notebook Tutorial

This toolkit provides many Jupyter Notebook and Google Colab tutorials. Users can study DeepKE with them.

Standard Setting

NER Notebook

NER Colab

RE Notebook

RE Colab

AE Notebook

AE Colab
Low-resource

NER Notebook

NER Colab

RE Notebook

RE Colab
Document-level

RE Notebook

RE Colab
Multimodal

NER Notebook

NER Colab

RE Notebook

RE Colab

Tips

Using nearest mirror, like THU in China, will speed up the installation of Anaconda.
Using nearest mirror, like aliyun in China, will speed up pip install XXX.
When encountering ModuleNotFoundError: No module named 'past'，run pip install future .
It's slow to install the pretrained language models online. Recommend download pretrained models before use and save them in the pretrained folder. Read README.md in every task directory to check the specific requirement for saving pretrained models.
The old version of DeepKE is in the deepke-v1.0 branch. Users can change the branch to use the old version. The old version has been totally transfered to the standard relation extraction (example/re/standard).
It's recommended to install DeepKE with source codes. Because user may meet some problems in Windows system with 'pip',and the source code modification will not work,seeissue
More related low-resource knowledge extraction works can be found in Knowledge Extraction in Low-Resource Scenarios: Survey and Perspective.
Make sure the exact versions of requirements in requirements.txt.

To do

In next version, we plan to add event extraction to the toolkit.

Meanwhile, we will offer long-term maintenance to fix bugs, solve issues and meet new requests. So if you have any problems, please put issues to us.

Citation

Please cite our paper if you use DeepKE in your work

@article{zhang2022deepke,
  title={DeepKE: A Deep Learning Based Knowledge Extraction Toolkit for Knowledge Base Population},
  author={Zhang, Ningyu and Xu, Xin and Tao, Liankuan and Yu, Haiyang and Ye, Hongbin and Qiao, Shuofei and Xie, Xin and Chen, Xiang and Li, Zhoubo and Li, Lei and others},
  journal={arXiv preprint arXiv:2201.03335},
  year={2022}
}

Developers

Zhejiang University: Ningyu Zhang, Liankuan Tao, Xin Xu, Haiyang Yu, Hongbin Ye, Shuofei Qiao, Peng Wang, Xin Xie, Xiang Chen, Zhoubo Li, Lei Li, Xiaozhuan Liang, Yunzhi Yao, Shumin Deng, Wen Zhang, Guozhou Zheng, Huajun Chen

Alibaba Group: Feiyu Xiong, Qiang Chen

DAMO Academy: Zhenru Zhang, Chuanqi Tan, Fei Huang

DeepKE
DeepKE copied to clipboard

Metadata

Table of Contents

What's New

Aug, 2022

June, 2022

May, 2022

Jan, 2022

Dec, 2021

Nov, 2021

Oct, 2021

Aug, 2019

Aug, 2018

Prediction Demo

Model Framework

Quick Start

Requirements

Introduction of Three Functions

1. Named Entity Recognition

2. Relation Extraction

3. Attribute Extraction

Notebook Tutorial

Tips

To do

Citation

Developers

← Metadata

Owner

Metadata

DeepKE DeepKE copied to clipboard

Metadata

Table of Contents

What's New

Aug, 2022

June, 2022

May, 2022

Jan, 2022

Dec, 2021

Nov, 2021

Oct, 2021

Aug, 2019

Aug, 2018

Prediction Demo

Model Framework

Quick Start

Requirements

Introduction of Three Functions

1. Named Entity Recognition

2. Relation Extraction

3. Attribute Extraction

Notebook Tutorial

Tips

To do

Citation

Developers

← Metadata

Owner

Metadata

DeepKE
DeepKE copied to clipboard