This repository contains code and data for reproducing Interpretable models for extrapolation in scientific machine learning. No ongoing maintenance or support of this content should be expected.
- download this folder and open a terminal inside it
- create a new python environment:
python3 -m venv venv
- activate the python environment:
. venv/bin/activate
- update pip in the python environment:
python -m pip install -U pip
- install dependencies in the python environment:
pip install -r requirements.txt
- start the jupyter server:
jupyter lab
- open
workflow.ipynb
in jupyter and run it
workflow.ipynb
: juypter notebook which reproduces the paperutils.py
: python functions imported byworkflow.ipynb
data
dataset_config.csv
: specifies the name, target property, size, and source of each datasetfig
: figures generated byworkflow.ipynb
and used in the paperraw
: raw data files in CSV format
Distribution A Approved for Public Release, Distribution Unlimited