I-GUIDE Summer School: GeoAI
About Organization
Objective
Our primary objective during the summer school was to benchmark and migrate models to I-GUIDE, specifically focusing on the newly released GeoAI Foundation Model by IBM and NASA. Then, we implemented our model.
Tools & Technologies
I-GUIDE Platform, Jupyter Notebooks, TensorFlow, Pytorch
Project Details
This project originated during the 2023 I-GUIDE Summer School on Convergence Science in Action, where I had the privilege of participating. The program was generously funded by the National Science Foundation (NSF) and took place in Boulder, Colorado from August 7th to August 11th, 2023. The school was organized by a distinguished group of professionals, including Dr. Mohan Ramamurthy, Dr. Anand Padmanabhan, Dr. Eric Shook, Dr. Diana Sinton, and Dr. Shaowen Wang.
I was in Team 6 (GeoAI, Team Awesome). There are three phases for our team project.
- Model Migration: We began by migrating the GeoAI Foundation Model to I-GUIDE and ran their demo to assess its compatibility and functionality.
- Testing and Validation: Subsequently, we conducted extensive testing on the model, specifically exploring flood, crop, and burn scar detection demos to ensure that it was working accurately.
- Custom Demo Development: Leveraging our ambition and enthusiasm, we took it a step further and developed a custom demo of our own. This involved the implementation of the Normalized Difference Building Index (NDBI) using the foundation model.
This project not only served as an opportunity to collaborate with experts in the field but also allowed us to contribute to the broader scope of GeoAI and convergence science. The knowledge and skills acquired during the summer school will undoubtedly play a pivotal role in advancing our careers and fostering interdisciplinary connections. We are excited about the potential impact of our project and look forward to further developments in this rapidly evolving field.