This repository contains PyTorch implementation of the submission of jmlr: Fitting Autoregressive Graph Generative Models through Maximum Likelihood Estimation

This is a minimum working version of the code used for the paper.

enviroment setup

pip install -r requirements.txt

Run build.sh script in the project's root directory for MMD computation.

./build.sh

To list the arguments, run the following command:

python main.py -h

To train the model on datasets with Rout and DAGG, run the following:

python main.py -dataset caveman_small

To evaluate the generated graph, run the following:

python main.py -task evaluate -load_model_path 'saved_model_path'