Source code for paper "CoLAKE: Contextualized Language and Knowledge Embedding". If you have any problem about reproducing the experiments, please feel free to contact us or propose an issue.

Prepare your environment

We recommend to create a new environment.

conda create --name colake python=3.7
source activate colake

CoLAKE is implemented based on fastNLP and huggingface's transformers, and uses fitlog to record the experiments.

git clone https://github.com/fastnlp/fastNLP.git
cd fastNLP/ & python setup.py install
git clone https://github.com/fastnlp/fitlog.git
cd fitlog/ & python setup.py install
pip install transformers==2.11
pip install sklearn

To re-train CoLAKE, you may need mixed CPU-GPU training to handle the large number of entities. Our implementation is based on KVStore provided by DGL. In addition, to reproduce the experiments on link prediction, you may also need DGL-KE.

pip install dgl==0.4.3
pip install dglke

Reproduce the experiments

1. Download the model and entity embeddings

Download the pre-trained CoLAKE model and embeddings for more than 3M entities. To reproduce the experiments on LAMA and LAMA-UHN, you only need to download the model. You can use the download_gdrive.py in this repo to directly download files from Google Drive to your server:

mkdir model
python download_gdrive.py 1MEGcmJUBXOyxKaK6K88fZFyj_IbH9U5b ./model/model.bin
python download_gdrive.py 1_FG9mpTrOnxV2NolXlu1n2ihgSZFXHnI ./model/entities.npy

Alternatively, you can use gdown:

pip install gdown
gdown https://drive.google.com/uc?id=1MEGcmJUBXOyxKaK6K88fZFyj_IbH9U5b
gdown https://drive.google.com/uc?id=1_FG9mpTrOnxV2NolXlu1n2ihgSZFXHnI

2. Run the experiments

Download the datasets for the experiments in the paper: Google Drive.

python download_gdrive.py 1UNXICdkB5JbRyS5WTq6QNX4ndpMlNob6 ./data.tar.gz
tar -xzvf data.tar.gz
cd finetune/

FewRel

python run_re.py --debug --gpu 0

Open Entity

python run_typing.py --debug --gpu 0

LAMA and LAMA-UHN

cd ../lama/
python eval_lama.py

Re-train CoLAKE

1. Download the data

Download the latest wiki dump (XML format):

wget -c https://dumps.wikimedia.org/enwiki/latest/enwiki-latest-pages-articles.xml.bz2

Download the knowledge graph (Wikidata5M):

wget -c https://www.dropbox.com/s/6sbhm0rwo4l73jq/wikidata5m_transductive.tar.gz?dl=1
tar -xzvf wikidata5m_transductive.tar.gz

Download the Wikidata5M entity & relation aliases:

wget -c https://www.dropbox.com/s/lnbhc8yuhit4wm5/wikidata5m_alias.tar.gz?dl=1
tar -xzvf wikidata5m_alias.tar.gz

2. Preprocess the data

Preprocess wiki dump:

mkdir pretrain_data
# process xml-format wiki dump
python preprocess/WikiExtractor.py enwiki-latest-pages-articles.xml.bz2 -o pretrain_data/output -l --min_text_length 100 --filter_disambig_pages -it abbr,b,big --processes 4
# Modify anchors
python preprocess/extract.py 4
python preprocess/gen_data.py 4
# Count entity & relation frequency and generate vocabs
python statistic.py

3. Train CoLAKE

Initialize entity and relation embeddings with the average of RoBERTa BPE embedding of entity and relation aliases:

cd pretrain/
python init_ent_rel.py

Train CoLAKE with mixed CPU-GPU:

./run_pretrain.sh

Cite

If you use the code and model, please cite this paper:

@inproceedings{sun2020colake,
  author = {Tianxiang Sun and Yunfan Shao and Xipeng Qiu and Qipeng Guo and Yaru Hu and Xuanjing Huang and Zheng Zhang},
  title = {CoLAKE: Contextualized Language and Knowledge Embedding},
  booktitle = {Proceedings of the 28th International Conference on Computational Linguistics, {COLING}},
  year = {2020}
}

Acknowledgments

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

txsun1997 / CoLAKE

Programming Languages

Labels

Projects that are alternatives of or similar to CoLAKE

CoLAKE