Geoconda is a collection of python packages that facilitate the development of python scripts for geoinformatics applications. It includes following python packages:
- ArcGIS Python API - provides simple and efficient tools for sophisticated vector and raster analysis, geocoding, map making, routing and directions.
- boto3 - for working files in S3 storage, for example Allas. Example.
- cartopy - for map plotting.
- dask - provides advanced parallelism for analytics, enabling performance at scale, including Dask-ML and Dask JupyterLab extension
- descartes - use Shapely or GeoJSON-like geometric objects as matplotlib paths and patches.
- fiona - reads and writes spatial data files.
- gdal - reads and writes spatial data files, and GDAL/OGR data manipulation tools.
- geoalchemy2 - provides extensions to SQLAlchemy for working with spatial databases, primarily PostGIS.
- igraph - for fast routing.
- geopandas - GeoPandas extends the datatypes used by pandas.
- networkx - for the creation, manipulation, and study of the structure, dynamics, and functions of complex networks.
- pyproj - performs cartographic transformations and geodetic computations.
- osmnx - download spatial geometries and construct, project, visualize, and analyze street networks from OpenStreetMap's APIs.
- pysal - spatial analysis functions.
- pdal - for lidar data
- rasterio - access to geospatial raster data.
- rasterstats - for summarizing geospatial raster datasets based on vector geometries. It includes functions for zonal statistics and interpolated point queries.
- rtree - spatial indexing and search.
- sentinelsat - downloading Sentinel images
- shapely - manipulation and analysis of geometric objects in the Cartesian plane.
- scikit-learn - machine learning for Python.
- skimage - algorithms for image processing.
- swiftclient, keystoneclient - for working with SWIFT storage, for example Allas.
- xarray - for multidimensional raster data.
- And many more, for retrieving the full list in Puhti use:
Additionally geoconda includes:
- spyder - Scientific Python Development Environment with graphical interface (similar to RStudio for R).
- GDAL/OGR commandline tools 3.0.4 in geoconda-3.8 and 3.0.2 in geoconda-3.7
- PDAL 2.1.0 in geoconda-3.8 and 2.0.1 in geoconda-3.7
- QGIS 3.14 in geoconda-3.8 and 3.10 in geoconda-3.7
- LasTools 20171231
- ncview for visualizing netcdf files
- proj4, geos and many more, see
/appl/soft/geo/geoconda/miniconda3/envs/geoconda-3.7/binfor full list.
If you want to use Spyder, QGIS, ncview or other tools with graphical user interfaces, you should connect to Puhti using NoMachine and start an interactive session for best performance
Python has multiple packages for parallel computing, for example multiprocessing, joblib and dask. In our Puhti Python examples there are examples how to utilize these different parallelisation libraries.
(If you think that some important GIS package for Python is missing from here, you can ask for installation from email@example.com.)
geoconda module is available in Puhti:
- 3.8 (version number is the same as Python version)
For using Python packages and other tools listed above, you can initialize them with:
module load geoconda
By default the latest geoconda module is loaded. If you want a specific version you can specify the version number of geoconda:
module load geoconda/[VERSION]
For using the Spyder IDE give:
To check the exact packages and versions included in the loaded module:
Adding more Python packages to GeoConda
You can add more Python packages to Geoconda for your own use with
pip, for example:
pip install [newPythonPackageName] --user
The packages are by default installed to your home directory under
.local/lib/python3.7/site-packages. If you would like to change the installation folder define
PYTHONUSERBASE environmentvariable with new installation location first:
You should use the same export command then always also before using, inc in the batch job files.
Using Allas from Python
There are two Python libraries installed in Geoconda that can interact with Allas. Swiftclient uses the swift protocol and boto3 uses S3 protocol. You can find CSC examples how to use both here. With large quantities of data in Allas, virtual rasters should be considered. More information on how to create and use virtual rasters can be found here.
License and citing
All packages are licensed under various free and open source licenses (FOSS), see the linked pages above for exact details. In your publications please acknowledge also oGIIR and CSC, for example “The authors wish to acknowledge for computational resources CSC – IT Center for Science, Finland (urn:nbn:fi:research-infras-2016072531) and the Open Geospatial Information Infrastructure for Research (oGIIR, urn:nbn:fi:research-infras-2016072513).”
- CSC Python parallelisation examples
- Python spatial libraries
- Essential Python Geospatial Libraries
- Geoprocessing with Python using Open Source GIS
- GeoExamples, a lot of examples of using Python for spatial analysis
- Automating GIS processes course materials, where most of the exercises are done using Python (University of Helsinki)
- Geohack Week materials
- Multiprocessing Basics