Contact Info
198 West 21th Street, Suite 721
New York, NY 10010
youremail@yourdomain.com
+88 (0) 101 0000 000
Follow Us

Pyinterpolate – Open Source for spatial prediction

We use Open Source on a daily basis. QGIS, Python packages and JavaScript libraries are foundation of our work. With Open Source you are able to develop everything: from machine learning models and data processing pipelines through web and mobile application. That’s why we actively participate in Open Source development.

Data Lions sponsored development of pyinterpolate package. Pyinterpolate is designed as the Python library for geostatistics. It’s role is to provide access to spatial statistics tools used in a wide range of studies. If you’re:

  • GIS expert,
  • geologist,
  • mining engineer,
  • ecologist,
  • public health specialist,
  • data scientist,

Then this package may be useful for you. You could use it for:

  • spatial interpolation and spatial prediction,
  • alone or with machine learning libraries,
  • for point and areal datasets.

Package was tested in commercial and research projects:

  • Tick-borne Disease Detector (prediction of areas of infection risk), research project funded by the European Space Agency,
  • commercial project related to the prediction of demand for specific flu medications,
  • commercial project related to the large-scale infrastructure maintenance.

Pyinterpolate allows you to perform:

  • Ordinary Kriging and Simple Kriging (spatial interpolation from points),
  • Centroid-based Kriging of Polygons (spatial interpolation from blocks and areas),
  • Area-to-area and Area-to-point Poisson Kriging of Polygons (spatial interpolation and data deconvolution from areas to points).

The example how Area-to-point Poisson Kriging works with epidemiological data. Top figure shows disease rates over areas and bottom figure shows exact population at risk.

Before deconvolution – our variable is aggregated over large area, it’s hard to decide how to deal with it:

After deconvolution – our variable is deconvoluted based on the relation to the process at a finer grid:

Package is still actively developed by our team members. You may download it from pip here: https://pypi.org/project/pyinterpolate/

Post a Comment