/SGMP_code

NeurIPS 2021 paper 'Representation Learning on Spatial Networks' code

Primary LanguagePython

Representation Learning on Spatial Networks

This repository is the official implementation of Representation Learning on Spatial Networks.

Training

Examples of running the models:

**1. training the SGMP model on BACE dataset:

  • bash run_SGMP_BACE.sh

**2. training the SGMP model with sampling spanning trees on BACE dataset:

  • bash run_SGMP_st_BACE.sh

**3. training the PointNet benchmark model on BACE dataset:

  • bash run_PointNet_BACE.sh

**4. training the SGMP model on QM9 dataset for target 0 ($\mu$):

  • bash run_SGMP_QM9.sh

**5. training the SGMP model with sampling spanning trees on QM9 dataset for target 0 ($\mu$):

  • bash run_SGMP_st_QM9.sh

**6. training the PointNet benchmark model on QM9 dataset for target 0 ($\mu$):

  • bash run_PointNet_QM9.sh

**7. generate synthetic dataset

  • cd data
  • python build_synthetic_data.py

Evaluation

The evaluation will be given in the ./results