Welcome to WaterSoftHack —a recurring two-week hackathon for water science students and researchers. As part of WaterSoftHack initiative, we created --WaterSoft Python Package--a curated training package focused on modern coding, CI, and machine learning modeling for water science. WaterSoft modules follows FAIR (Findable, Accessible, Interoperable, Reusable) CI practices. In this project, we devoted significant effort to teaching workflow interoperability and reproducibility—including data acquisition, preprocessing, model installation, execution, deployment, and reuse.
This repository serves as a landing page and introduction to the WaterSoft Package, developed as part of the WaterSoftHack training program. Please acknowledge use below citation and cknolwedge our effort when you use these freely accessable modules**.
watersofthack. (n.d.). WaterSoft Python Package. GitHub. https://github.com/watersofthack/WaterSoft/blob/main/README.md
WaterSoft Sustainability Plan
Project Leads: The PIs Provide strategic direction, oversee scientific and technical integrity, approve major releases, and ensure alignment with WaterSoftHack training objectives and NSF deliverables.
Core Developers: Our team maintain core WaterSoft modules, review and merge pull requests, manage software releases, and ensure code quality, documentation, and reproducibility.
Training and Curriculum Leads: Our team develop, update, and maintain WaterSoftHack training materials, tutorials, and example workflows; align content with evolving CI and water science needs.
Community Contributors: Community members can submit issues, propose enhancements, contribute code, workflows, or documentation, and participate in GitHub Discussions.
The repository is divided into three main sections:
The Data Analytics section features HydroSuite, a collection of web-based tools developed by the Hydroinformatics Lab at Tulane University. These tools enable:
- Development of hydrologic web applications
- Fast in-browser analysis
- Enhancement and extension of existing tools
The objective is to demonstrate the use of web-based systems to improve current hydrological workflows. For complete documentation, click here.
The Machine Learning section provides a series of tutorials focused on time series processing for water science-related datasets.These codes and materials are developed by Clemson Hydroinformatics Research Group. Topics include:
- Data acquisition from multiple sources
- Data exploration, cleaning, and preparation
- Feeding processed data into machine learning models
It covers a range of models, from ARIMA to Transformers, with step-by-step guides and detailed explanations.
Explore more here.
(Coming soon)
This section will include tutorials and tools for deploying data analytics and machine learning models in cloud environments.