INTRODUCTION TO GEOSPATIAL DATA SCIENCE
This nearly two and a half hour pre-recorded course aims to give a comprehensive overview of getting started in geospatial data science using Python. The course covers the key concepts of spatial analytics, such as geometries and coordinate reference systems, different types of geometric data, such as road networks and building footprints, and map status and interactive visualizations. Attention - this is a fully hands-on Python course with a main focus on vector data.
The course includes 4 pre-recorded subtitled videos with and the entire code base discussed during class in live and revised versions.
Setting up
To fully enjoy this course, you need to install a Python version, preferably in a Jupyter Notebook environment. Additionally, you will need to install several Python libraries, the versions of which are listed in the requirements.txt file. https://jupyter.org
Geometries - The Building Blocks of Spatial Analytics
Onboarding to geometries - Shapely, imports, basic geometry typesGeometry operations - buffering, set operations, and others
────────────
GeoPandas in Practice - The Spatial Swiss Knife
- Onboarding to GeoPandas - imports, versions, sample data, Natural Earth, GeoDataFrames
- Simple functions and computations - creating new geo features, statistics, histograms, correlations
- Visualizing sample data with GeoPandas - basic statistics, color maps, log scaling
- CRS and map projections - local and global coordinate reference systems
────────────
Collecting and Exploring Vector Data - OpenStreetMap
- OpenStreetMap
- Download data with OSMNx - polygons, footprints, POIs, road networks, computational exercises
- Combined map visualization - multiple layers, base maps, deriving complex urban features
- Interactive visualizations - Folium
────────────
Geospatial Features and Urban Analytics
- Download all districts for Budapest - downloading, storing, parsing, and merging data
- Urban feature engineering - road networks, density, and distance-based metrics
- Automate feature generation