Updated assign land use classes:

- number of categories now dynamic
- included a category function for aggregation

New release

measprocess/geospatial/process.py

@@ -4,7 +4,7 @@ from shapely.geometry import LineString, Point, Polygon, box
 from shapely.ops import unary_union
 from tqdm import tqdm
 import pandas as pd
-from typing import Dict
+from typing import Dict, Union, Callable
 
 def project_onto_streets(
     point_series: gpd.GeoSeries,
@@ -206,13 +206,16 @@ def delete_indoor_meas(
 
     return outdoor_meas_gdf.geometry
 
+
 def assign_landuse_classes(
     point_geodataframe: gpd.GeoDataFrame,
     polygon_geodataframe: gpd.GeoDataFrame,
     cut_polygons_by_buffer: bool = False,
     buffer_size: int = 0,
+    category_function : Union[None, Callable] = lambda land_use_code: "Urban" if land_use_code < 200 else "Rural",
     epsg: str = "EPSG:31287",
-) -> Dict:
+    n_categories: int = 5
+) -> gpd.GeoDataFrame:
     '''
     Returns a DataFrame with the landuse class distribution in the area defined by the buffer_size around the passed locations
     Landuseclasses are defined according to the Corine Land Cover Project:
@@ -230,7 +233,9 @@ def assign_landuse_classes(
     :param polygon_geodataframe: A GeoDataFrame with the landuse class polygons. Can be retrieved from
     :param cut_polygons_by_buffer: If True the polygon_geodataframe will be cut by the total bounds of the point series. Can result in a speedup.
     :param buffer_size: The buffer_size in meters used to construct the area around points for which the distribution of classes will be computed.
+    :param category_function: If not None a additional dataframe is returned which aggregates land_use classes to the defined categories
     :param epsg: The projection used for the computation of the area. Defaults to EPSG:31287 for Austria.
+    :param n_categories: Maximum number of top categories sorted by the respective percentage values to be considered.
 
     :return: A GeoDataFrame with columns describing the percentage of the area assigned to the respective land use class.
     '''
@@ -273,8 +278,22 @@ def assign_landuse_classes(
         for key, value in zip(codes, percentages):
             code_percent_dict[key] = code_percent_dict.get(key, 0) + value
         code_percent_dict = dict(sorted(code_percent_dict.items(), key=lambda x: x[1], reverse =True))
-        first5pairs = {k: dict(code_percent_dict)[k] for k in list(dict(code_percent_dict))[:5]}
+        firstnpairs = {int(k): dict(code_percent_dict)[k] for k in list(dict(code_percent_dict))[:n_categories]}
+
+        results[i] = firstnpairs
+
+    if category_function is not None:
+        processed = {}
+        for key, result in results.items():
+            processed[key] = {}
+            for land_use, area_percentage in result.items():
+                category = category_function(land_use)
+                processed[key][category] = processed[key].get(category, 0) + area_percentage
+
+        processed_result = pd.DataFrame(processed).transpose()
+        processed_result.replace(np.nan, 0, inplace=True)
 
-        results[i] = first5pairs
+        return results, processed_result
 
-    return results
+    else:
+        return results

setup.py

@@ -12,7 +12,7 @@ README = (HERE / "README.md").read_text()
 # This call to setup() does all the work
 setup(
     name="measprocess",
-    version="0.6.6",
+    version="0.6.7",
     description="Collection of measurement processing tools",
     long_description=README,
     long_description_content_type="text/markdown",