PI: ZIHENG SUN, GEORGE MASON UNIVERSITY
Geoweaver is a web system allowing users to easily compose and execute full-stack deep learning workflows by taking advantage of online spatial data facilities, high-performance computation platforms, and open-source deep learning libraries. It is a perfect alternative to SSH client (e.g., Putty), FTP client, and scientific workflow software.
Through the ESIP Lab project, the team succeeded in:
- turning large-scale distributed deep network into manageable modernized workflows;
- boosting higher utilization ratio of the existing cyberinfrastructures by separating scientists from tedious technical details;
- enhancing the frequency and accuracy of classified land cover land use maps for agricultural purposes;
- enabling the tracking of provenance by recording the execution logs in structured tables to evaluate the quality of the result maps;
- prove the effectiveness of operationally using large-scale distributed deep learning models in classifying Landsat image time series
Ziheng Sun, et al, Crop Mapping by Deep Neural Network and Cyberinfrastructure, ISPRS Journal of Photogrammetry and Remote Sensing, under review.
Ziheng Sun, et al, Advanced Cyberinfrastructure for Intercomparison and Validation of Climate Models, Environmental Modelling and Software, under review.
Ziheng Sun, et al, Advanced Geospatial Cyberinfrastructure for Pulling GIScience into AI Phase, in preparation.
- Presentations and posters on Figshare.
- 2018 InnoCentive Challenge Winner
VIEW IT ON GITHUB