After I spend the first few weeks mainly working on learning hte bafkground info neeeded to work with this kind of geo spatial data, week 3-5 was spent creating the first GeoAI module that teschers can use to learn and eventually teach GeoAI in schools. The module is a Google Colab, so it is part text and part code. The text explains all the background info needed: machine learning. GeoAI, Geospatial date, etc. The code blocks walks through an example of stream detection using UNET archecture.

Week one and two went very well. I spent most of the first week getting settled in Gainesville. I met Professor Jiang for the first time in person and we were abloe to talk about this summer’s project in greater depth. I spent the following week reviewing the background information to Professor Jiang’s research on using GeoAI for mapping coastal storms. My eventual goal is create some kind of learning tool for high school teachers, some kind of module or notebook, which teachers the ML methods and strategies employed in Professor Jiang’s research. I learned about the different kinds of geospatial data (raster and vector), reading this data in the form of satelittle images, preoproccess this data, and use machine learning models to map and predicts the data. The specific model we will be using is UNET archeticture (a type of neural network). Although I have taken a machine learning class at my home college, this has been new and interesting becseu I never used machine learning on images!