GSOC 2019 Ideas Page
This is just a starting point. Your own ideas are most welcome. Please questions on gitter
We are under the NumFocus umbrella, so please check their offical gsoc page for information on how to apply. We are not requiring a pull request to apply.
GeoPandas allows users to work with geospatial data in Python by providing geospatial operations in Pandas and a high-level interface to multiple geometries to shapely. It also facilitates exploratory visualization of geospatial data via descartes and matplotlib.
The visualization interface in geopandas is currently on the rough side; it generates simple polygon based maps that are inconsistent in terms of supporting matplotlib features such as color mapping and color bars.
| Intensity | Involves | Mentors |
|---|---|---|
| Intermediate | Python, GIS | @story645, @jorisvandenbossche, @phobson |
Technical details This work will likely mostly be a matter of diagnosing failure points in the geopandas imshow and integrating support for features supported by matplotlib. For example:
- fixing support for color mapping
- adding support for colorbar arguments
- customized legend markers for distinguishing points, lines and polygons.
- support choropleth with column with missing values
- other new plotting features
This project is in collaboration with geopandas, and it’s expected that tasks will also be solicited from them. There is also room for collaboration with Pysal
Open Source Development Experience This project involves experience with Python, understanding of matplotlib’s API, and understanding of geospatial data analysis and visualization.