Skip to the content.

About

Click here for detailed papers on this work!

To download all cases via Kaggle API, download this bash script.

Summary of the data are avalable here!

Mission

BLASTNet aims to address gaps in open machine learning (ML) within the sciences, specifically fluid mechanics by providing researchers in reacting and non-reacting flow physics communities with (mostly) externally contributed open-source ML resources.

These contributions now include (i) 4.8 TB of high-fidelity simulation datasets that have been processed in a convenient format for ML applications, (ii) >13,000 lines of code that aid the training and evaluating of these models, (iii) >100 pre-trained weights in flow physics problems, and (iv) regular workshop events that disseminate ML for flow physics via seminars and competitions.

Application

This data is useful for fluid flows in a wide range of ML applications tied to automotive, propulsion, energy, and the environment. Specifically, scientific engineering tasks related to these domains may include turbulent closure modeling, spatio-temporal modeling, and inverse modeling.

Distribution

Our ML resources are shared via github and Kaggle. Specifically, code is shared via github, while data and models are shared via Kaggle.

To circumvent Kaggle storage constraints, we partition our data into a network of <100 GB subsets, with each subset containing a separate simulation configuration. This partitioned data can then be uploaded as separate datasets on Kaggle. To download all cases via Kaggle API, download this bash script. Summary of the data are avalable here!

Our network of datasets approach: approach

Contribute

Find out how to contribute to our project to join our growing list of authors.

Authors and Contributors

We thank the following people for contributing and curating this network-of-datasets:

Name Affilliation Date Joined
Wai Tong Chung Stanford University June 6 2022
Matthias Ihme Stanford University, SLAC National Laboratory June 6 2022
Ki Sung Jung Sandia National Laboratory June 6 2022
Jacqueline H. Chen Sandia National Laboratory June 6 2022
Jack Guo Stanford University June 6 2022
Davy Brouzet Stanford University June 6 2022
Mohsen Talei University of Melbourne June 6 2022
Bin Jiang University of Melbourne Nov 18 2022
Bruno Savard Polytechnique Montréal Jan 26 2023
Alexei Y. Poludnenko University of Connecticut June 7 2023
Bassem Akoush Stanford University June 7 2023
Pushan Sharma Stanford University June 7 2023
Alex Tamkin Stanford University June 7 2023
Qing Wang Google August 29 2023
Shantanu Shahane Google August 29 2023
Yifan Chen Google August 29 2023