prospectivity prediction based on GIS shapefiles
This is a pretty simple prospectivity prediction package that predicts the prospectivity of an element based on how close it is to different bedrock units.
Overview of Package
The package utilizes 3 main classes.
- handles exploring and prepping a shapefile and determining the project boundary. It can read in a single shapefile (using geopandas) and then you can use it to select the polygons of interest.
- uses the previously initialized PrepShapes() class to help you setup the raster template used by the predictor to predict prospectivity at a specified location.
- the predictor class that predicts the likelihood based on distance to shapes of interest and the weighting schema architect. The predictor class uses the previously initialized PrepShapes() and RasterTemplate() classes. Only one prediction architect is currently implemented.
Prediction Weighting Schema
Currently this package uses a pseudo variogram style weighting schema with the following model (for location i and as an example 2 distances)
Generating a prospectivity heat map
Using the package you can generate a heat map with liklihood of finding the element (based on a distance from 2 or more shape categories). The predictor generates prediction values ranging from 0, being least likely (i.e. 0%), to 1, being most likely (i.e. 100%). As an example here is a heat map generated from the included dataset.
An example.ipynb is included in the repo to demostate how to use the package.
the included dataset is from the British Columbia Geologica Survey.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size prospectpredictor-0.1.1-py3-none-any.whl (14.1 kB)||File type Wheel||Python version py3||Upload date||Hashes View|
|Filename, size prospectpredictor-0.1.1.tar.gz (13.5 kB)||File type Source||Python version None||Upload date||Hashes View|
Hashes for prospectpredictor-0.1.1-py3-none-any.whl