Tensorflow Implementation of the CGNN

Code provided to reproduce the results from the article "Learning Functional Causal Models with Generative Neural Networks"

Requirements: numpy scipy scikit-learn tensorflow joblib pandas

In order to run the CGNN and launch the experiments:

1. First install the CGNN package. Enter in the code directory. Run the command line "python setup.py install develop --user"

2. Launch the example python script for pairwise inference: "python run_GNN_pairwise_inference.py"

3. Launch the example python script for graph reconstruction from a skeleton: "python run_CGNN_graph.py"

4. Launch the example python script for graph reconstruction in presence of hidden variables: "python run_CGNN_graph_hidden_variables.py"

5. The complete datasets used in the article may be found at the following url:

• pairwise datasets : http://dx.doi.org/10.7910/DVN/3757KX
• graph datasets : http://dx.doi.org/10.7910/DVN/UZMB69

Fast Pytorch implementation of CGNN available in the CDT

A faster implementation of CGNN in pytorch in available in the CausalDiscoveryToolBox (CDT)

https://github.com/Diviyan-Kalainathan/CausalDiscoveryToolbox

arXiv paper of the CDT: https://arxiv.org/abs/1903.02278