A collaborative framework of logic rule-based model and neural network-based model for Temporal Knowledge Graph Completion.
Our codebase is partially built upon the following open-source projects. We sincerely thank the authors for their excellent work and for sharing their code:
- TLogic - Used the rule generation code for the logical rule module.
- ECEformer - Used as neural network module.
- Ubuntu 22.04 / 24.04
- Python >= 3.8
- CUDA >= 11.6 (if using GPU acceleration)
- java runtime environment >= 11
1). Clone this repository and install the required dependencies via requirements.txt:
pip install -r requirements.txt2). Follow our fork of ECEformer to install the TKGE package.
3). Download the jar archive Apache Commons CLI (version >= 1.9.0) and put it into the mln directory.
4). The source code of mln/mln.jar see here.
sh exp_tece.shDetailed configuration please refer to the content of exp_tece.sh.