/fast_reorg_energy_prediction

Data and implementation of ChIRo and SchNet for "Accelerating Organic Electronic Materials Design with Low-Cost Molecular Reorganization Energy Predictions"

Primary LanguageJupyter Notebook

Reorganization Energy Predictions with Graph Neural Networks Informed by Low-Cost Conformers

This repository includes:

  • Curated QM9 dataset with their vertical IP, vertical EA, reorganization energy in ./data
  • Code for training and evaluating the Modified ChIRo model in ./ChIRo
  • Code for training and evaluating the SchNet implemented with SchNetPack in ./SchNet

Requirments to Run ChIRo and SchNet

  • Python = 3.8.13
  • Pytorch = 1.11.0
  • PyG = 2.0.4
  • RDKit = 2022.03.2
  • scikit-learn = 1.1.1
  • NumPy = 1.22.3
  • Pandas = 1.4.3
  • SciPy = 1.5.3
  • SymPy = 1.10.1
  • schnetpack = 1.0.0