Paper: TransChem: Data-based prediction of redox potentials via introducing chemical features into the transformer architecture

• Getting Started

  • Prerequisites

  • Prediction

  • Train TransChem on your own dataset

• Acknowledgement

• LICENSE

1. Clone this repo

git clone https://github.com/xianHUJINGJING/TransChem.git
cd example

1. Prepare model weight

Download model_xxx.pt (160M), put it under './model_weight' directory.

1. Install dependencies.

This code requires Python 3.7, PyTorch, and a few other Python libraries. We recommend creating conda environment and installing all the dependencies as follows:

# create conda env
conda create --name transchem python=3.7
# activate env
conda actiavte transchem
# install pytorch with CUDA 11.0
conda install pytorch torchvision torchaudio cudatoolkit=11.0 -c pytorch
# install other python packages
pip install pickle rdkit sklearn random numpy pandas

The PyTorch version we tested is 1.9.0.

cd example

To train TransChem on your own dataset, please prepare your dataset annotations following the format

Training can be launched by running the following command:

python train.py

For more configurable options, please checkout our config file config.py.

The annotation files are under CC BY-NC-SA 4.0 license, see ./data/LICENSE. All the code are under MIT license, see LICENSE.