Welcome to Malaya’s documentation!#


MIT License discord

Malaya-Graph is a Knowledge Graph Toolkit for Bahasa Malaysia, powered by Tensorflow and PyTorch.


Proper documentation is available at https://malaya-graph.readthedocs.io/

Installing from the PyPI#

$ pip install malaya-graph

It will automatically install all dependencies except for Tensorflow and PyTorch. So you can choose your own Tensorflow CPU / GPU version and PyTorch CPU / GPU version.

Only Python >= 3.6.0, Tensorflow >= 1.15.0, and PyTorch >= 1.10 are supported.

Development Release#

Install from master branch,

$ pip install git+https://github.com/huseinzol05/malaya-graph.git

We recommend to use virtualenv for development.

Documentation at https://malaya-graph.readthedocs.io/en/latest/


  • Text to Knowledge Graph, Generate knowledge graph from human sentences.

  • Is Triplet, Clasify a triplet is a correct head-type-tail relationship.

  • Knowledge Graph to Text, Generate human sentences from a knowledge graph.

  • Link Prediction, Zero-shot link classifier between a head and a tail.

  • Relation Prediction, Zero-shot head or tail classifier.


If you use our software for research, please cite:

@misc{Malaya, Knowledge Graph Toolkit for Bahasa Malaysia, powered by Tensorflow and PyTorch,
  author = {Husein, Zolkepli},
  title = {Malaya-Graph},
  year = {2022},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/huseinzol05/malaya-graph}}


Thanks to Mesolitica for private RTXs cloud to train Malaya-Graph models,