GNN_MF_GE

Generalization Error of Graph Neural Network in the Mean-field Regime.

Environment Setup

The codebase is implemented in Python 3.7. package versions used for development are below.

networkx                        2.6.3
numpy                           1.20.3
scipy                           1.7.1
argparse                        1.1.0
torch                           1.10.1
pyg                             2.0.3

Folder structure

./execution/ stores files that can be executed to generate outputs. For vast number of experiments, we use GNU parallel, which can be downloaded in command line and make it executable via:

wget http://git.savannah.gnu.org/cgit/parallel.git/plain/src/parallel
chmod 755 ./parallel

./joblog/ stores job logs from parallel. You might need to create it by

mkdir joblog

./Output/ stores raw outputs (ignored by Git) from parallel. You might need to create it by

mkdir Output

./data/ stores processed data sets.
./result_arrays/ stores results for different data sets. Each data set has a separate subfolder.
./exp/ stores trained models and logs.

Reproduce results

First, get into the ./execution/ folder:

cd execution

To reproduce the results to be executed on GPU-0 for machine a.

bash a0.sh

Note that if you are operating on CPU, you may delete the commands ``CUDA_VISIBLE_DEVICES=xx". You can also set you own number of parallel jobs, not necessarily following the j numbers in the .sh files, or use other GPU numbers.

Direct execution with training files

Below are various options to try:

Creating a GCN model using the mean-readout function for the ER-4 data set.

python ./run_exp.py --model_type GCN --dataset ER-4 --pooling_method mean

Creating an MPGNN model using the sum-readout function for the PROTEINS data set and run for 50 epochs.

python ./run_exp.py --model_type MPGNN --max_epochs 50 --dataset PROTEINS --pooling_method sum

SherylHYX/GNN_MF_GE

GNN_MF_GE

Environment Setup

Folder structure

Reproduce results

Direct execution with training files