Vector geometry toolkit¶

Build and reshape FeatureCollections:

• from_bbox / from_records — construct collections from a box or from dict rows.
• with_coordinates / with_centroid — attach x/y or centroid columns.
• explode — split multi-part geometries into single parts.
• concat — stack two collections.

Setup¶

In [1]:

%matplotlib inline

import tempfile
from pathlib import Path

import numpy as np

DATA = Path('../../../examples/data')
WORK = Path(tempfile.mkdtemp(prefix='pyramids-t3-'))
DATA.is_dir(), WORK.is_dir()

%matplotlib inline import tempfile from pathlib import Path import numpy as np DATA = Path('../../../examples/data') WORK = Path(tempfile.mkdtemp(prefix='pyramids-t3-')) DATA.is_dir(), WORK.is_dir()

Out[1]:

(True, True)

In [2]:

from shapely.geometry import Point

from pyramids.feature import FeatureCollection

from shapely.geometry import Point from pyramids.feature import FeatureCollection

2026-07-11 14:42:28 | INFO | pyramids.base.config | Logging is configured.

Construct — from_bbox and from_records¶

In [3]:

box = FeatureCollection.from_bbox((0.0, 0.0, 1.0, 1.0), epsg=4326)
pts = FeatureCollection.from_records(
    [{'id': 1, 'geometry': Point(0, 0)}, {'id': 2, 'geometry': Point(1, 1)}], crs=4326
)
(len(box), box.epsg), (len(pts), list(pts.columns))

box = FeatureCollection.from_bbox((0.0, 0.0, 1.0, 1.0), epsg=4326) pts = FeatureCollection.from_records( [{'id': 1, 'geometry': Point(0, 0)}, {'id': 2, 'geometry': Point(1, 1)}], crs=4326 ) (len(box), box.epsg), (len(pts), list(pts.columns))

Out[3]:

((1, 4326), (2, ['id', 'geometry']))

Derive columns — with_coordinates / with_centroid¶

In [4]:

polys = FeatureCollection.read_file(DATA / 'coello_polygons.geojson')
coords = polys.with_coordinates()
centroids = polys.with_centroid()
[c for c in coords.columns if c in ('x', 'y')], type(centroids).__name__

polys = FeatureCollection.read_file(DATA / 'coello_polygons.geojson') coords = polys.with_coordinates() centroids = polys.with_centroid() [c for c in coords.columns if c in ('x', 'y')], type(centroids).__name__

Out[4]:

(['x', 'y'], 'FeatureCollection')

Plot the source polygons, then their centroids — the same features reduced to a single representative point each.

In [5]:

polys.plot()
centroids.plot()

polys.plot() centroids.plot()

Out[5]:

<Axes: >

Reshape — explode and concat¶

In [6]:

single_parts = polys.explode()
stacked = polys.concat(polys)
(len(polys), '-> explode ->', len(single_parts)), ('concat ->', len(stacked))

single_parts = polys.explode() stacked = polys.concat(polys) (len(polys), '-> explode ->', len(single_parts)), ('concat ->', len(stacked))

Out[6]:

((4, '-> explode ->', 4), ('concat ->', 8))

Plot the exploded single-part geometries — each multi-part feature is now split into its own row.

In [7]:

single_parts.plot()

single_parts.plot()

Out[7]:

<Axes: >

Notes¶

• A FeatureCollection is a geopandas.GeoDataFrame subclass, so all of geopandas works too.
• See also: Rasterize ↔ vectorize, Vector formats.