This is the implementation of the Neural Logic Inductive Learning model (NLIL) proposed in the ICLR 2020 paper: Learn to Explain Efficiently via Neural Logic Inductive Learning. The Transformer implementation is based on this repo.

• python 3.6+
• pytorch 1.1.0+
• numpy
• tqdm

You can run knowledge completion task on WN18 and FB15K with provided scripts

bash run_wn.sh
bash run_fb.sh

First, download the scene-graph dataset from the official site (click "Download Scene Graphs")

https://cs.stanford.edu/people/dorarad/gqa/download.html

Extract the files, and run the following script to generate the dataset

bash preprocess.sh path/to/the/sgraph/folder

Now you can run object classification with

bash run_gqa.sh

@inproceedings{
    yang2020learn,
    title={Learn to Explain Efficiently via Neural Logic Inductive Learning},
    author={Yuan Yang and Le Song},
    booktitle={International Conference on Learning Representations},
    year={2020},
    url={https://openreview.net/forum?id=SJlh8CEYDB}
}