A package for predicting the distribution of dust particles in protoplanetary disk based off our paper Multi-Output Random Forest Regression to Emulatethe Earliest Stages of Planet Formation by André Zazzera, Kevin Hoffman, and Jae Sung. Our work for this capstone project can be found in this repository.
Astrodust can be installed from PyPI via pip:
pip install astrodust
The package requires two pretrained models, a random forest regression model and XGBoost classifier. These can be downloaded beforehand from Zenodo and placed in the current working directory in a models directory. Otherwise the package will prompt to automatically download them when the DustModel is instaniated.
The documentation for the package is located here. A demonstration code notebook is also available, or can be viewed online here.
Particle sizes for each bin for the input and output are included as a reference in our wiki.