Bayesian Optimization for kinetic Monte Carlo simulation of Free Radical Polymerization (BO-KMC)

This project is a space exploration framework that integrates Bayesian optimization and dynamic Monte Carlo simulation. It aims to find suitable sets of reaction parameters for free radical polymerization to achieve lower dispersity.

Installation Steps

1. Install Dependencies

This project is specifically designed to run in a Linux environment. Installing the KMC program on a Windows environment might result in errors. Ensure that the following dependencies and their versions are installed in your environment:

• python==3.11
• numpy==1.26.0
• pandas==2.0.3
• pytorch=2.0.1
• gpytorch==1.6.0
• botorch==0.6.0 (for detailed installation guide on botorch, please see https://github.com/pytorch/botorch)

2. Clone the Project

git clone https://github.com/gooster1997/BO-KMC.git
cd BO-KMC

3. Install KMC Functions

Before running the main program, it's necessary to install the KMC calculation files using pypa/build. Follow these steps:

python -m build
cd dist
pip install YOUR_FILE_NAME.whl

Replace YOUR_FILE_NAME.whl with the actual name of the generated .whl file in the dist directory. Afterward, the installation should be completed.

4. Running the Project

Run the BO_demo.ipynb file to execute this project.

Contribution

This project credits Yu Gu, Tianyi Gao, and Mao Chen from the Department of Macromolecules, Fudan University. If you have any suggestions or find a bug, feel free to raise an issue or contact ygu19@fudan.edu.cn or chenmao@fudan.edu.cn.