Result Types¶

When calling compute, results are packaged into these types (or returned as builtin Python types when appropriate, such as for Dict, Int, or Datetime).

Note that these are just simple containers, and aren’t interoparable with their corresponding Proxytypes. For instance, you can’t do wf.Image.from_id("foo") + my_image_result, where my_image_result is an ImageResult instance.

Classes:

`ImageResult`(ndarray, properties, bandinfo, …)	Result of calling `compute` on an `Image`.
`ImageCollectionResult`(ndarray, properties, …)	Result of calling `compute` on an `ImageCollection`.
`GeometryResult`(type, coordinates[, crs])	Result of calling `compute` on a `Geometry`.
`GeometryCollectionResult`(type, geometries[, crs])	Result of calling `compute` on a `GeometryCollection`.
`FeatureResult`(geometry, properties)	Result of calling `compute` on a `Feature`.
`FeatureCollectionResult`(features)	Result of calling `compute` on a `FeatureCollection`.

class ImageResult(ndarray, properties, bandinfo, geocontext)[source]¶

Result of calling compute on an Image.

Examples

>>> from descarteslabs.workflows import Image
>>> my_img = Image.from_id("sentinel-2:L1C:2019-05-04_13SDV_99_S2B_v1")
>>> my_img.compute(my_geoctx) # my_geoctx is an arbitrary geocontext for 'my_img' 
ImageResult:
  * ndarray: MaskedArray<shape=(27, 512, 512), dtype=float64>
  * properties: 'absolute_orbit', 'acquired', 'archived', 'area', ...
  * bandinfo: 'coastal-aerosol', 'blue', 'green', 'red', ...
  * geocontext: 'geometry', 'key', 'resolution', 'tilesize', ...

ndarray¶

3-dimensional array of image data, in order (band, y, x)

Type:	numpy.ndarray

properties¶

dict of metadata about the Image.

Type:	dict[str, any]

bandinfo¶

OrderedDict of metadata about each band. The order corresponds to the bands in the ndarray.

Type:	OrderedDict[str, dict[str, any]]

geocontext¶

GeoContext over which computation was done.

Type:	dict

class ImageCollectionResult(ndarray, properties, bandinfo, geocontext)[source]¶

Result of calling compute on an ImageCollection.

Examples

>>> from descarteslabs.workflows import ImageCollection
>>> my_col = ImageCollection.from_id("landsat:LC08:01:RT:TOAR",
...        start_datetime="2017-01-01",
...        end_datetime="2017-05-30")
>>> my_col.compute(my_geoctx) # my_geoctx is an arbitrary geocontext for 'my_col' 
ImageCollectionResult of length 2:
  * ndarray: MaskedArray<shape=(2, 27, 512, 512), dtype=float64>
  * properties: 2 items
  * bandinfo: 'coastal-aerosol', 'blue', 'green', 'red', ...
  * geocontext: 'geometry', 'key', 'resolution', 'tilesize', ...

ndarray¶

4-dimensional array of image data, in order (image, band, y, x)

Type:	numpy.ndarray

properties¶

List of metadata dicts about each Image in the ImageCollection. The order corresponds to axis 0 of the ndarray.

Type:	list[dict[str, any]]

bandinfo¶

OrderedDict of metadata about each band. The order corresponds to the bands (axis 1) in the ndarray.

Type:	OrderedDict[str, dict[str, any]]

geocontext¶

GeoContext over which computation was done.

Type:	dict

class GeometryResult(type, coordinates, crs=None)[source]¶

Result of calling compute on a Geometry.

Examples

>>> from descarteslabs.workflows import Geometry
>>> my_geom = Geometry(type="Point", coordinates=[1, 2])
>>> my_geom.compute() 
GeometryResult(type=Point, coordinates=[1, 2])

type¶

The type of geometry. One of “Point”, “MultiPoint”, “LineString”, “MultiLineString”, “Polygon”, “MultiPolygon”, or “LinearRing”.

Type:	str

coordinates¶

Coordinates for the geometry, in WGS84 lat-lon (EPSG:4326). May be a list of floats, a list of lists of floats, or a list of lists of lists of floats depending on the type.

Type:	list

__geo_interface__¶

GeoJSON representation of the Geometry, following the Python __geo_interface__ convention.

Type:	dict

shapely¶

The GeometryResult as a shapely shape. Raises ImportError if the shapely package is not installed.

Type:	shapely.geometry.BaseGeometry

class GeometryCollectionResult(type, geometries, crs=None)[source]¶

Result of calling compute on a GeometryCollection.

Examples

>>> from descarteslabs.workflows import Geometry, GeometryCollection
>>> my_geom = Geometry(type="Point", coordinates=[1, 2])
>>> my_gc = GeometryCollection(type="GeometryCollection", geometries=[my_geom, my_geom, my_geom])
>>> my_gc.compute() 
GeometryCollectionResult(type=GeometryCollection,
        geometries=(
            GeometryResult(type=Point, coordinates=[1, 2]),
            GeometryResult(type=Point, coordinates=[1, 2]),
            GeometryResult(type=Point, coordinates=[1, 2])))

type¶

The type of geometry. Always “GeometryCollection”.

Type:	str

geometries¶

List of GeometryResult objects in the collection.

Type:	list[GeometryResult]

__geo_interface__¶

GeoJSON representation of the GeometryCollection, following the Python __geo_interface__ convention.

Type:	dict

shapely¶

The GeometryCollection as a shapely shape. Raises ImportError if the shapely package is not installed.

Type:	shapely.geometry.GeometryCollection

class FeatureResult(geometry, properties)[source]¶

Result of calling compute on a Feature.

Examples

>>> from descarteslabs.workflows import Geometry, Feature
>>> my_geom = Geometry(type="Point", coordinates=[1, 2])
>>> my_feat = Feature(geometry=my_geom, properties={"foo": "bar"})
>>> my_feat.compute() 
FeatureResult(geometry=GeometryResult(type=Point, coordinates=[1, 2]), properties={'foo': 'bar'})

geometry¶

The GeometryResult of the feature.

Type:	GeometryResult

properties¶

Properties associated with the FeatureResult.

Type:	dict

__geo_interface__¶

GeoJSON representation of the FeatureResult, following the Python _geo_interface__ convention.

Type:	dict

shapely¶

The geometry of the FeatureResult as a shapely shape. Raises ImportError if the shapely package is not installed.

Type:	shapely.geometry.BaseGeometry

class FeatureCollectionResult(features)[source]¶

Result of calling compute on a FeatureCollection.

Examples

>>> from descarteslabs.workflows import Geometry, Feature, FeatureCollection
>>> my_geom = Geometry(type="Point", coordinates=[1, 2])
>>> my_feat = Feature(geometry=my_geom, properties={"foo": "bar"})
>>> my_fc = FeatureCollection(features=[my_feat, my_feat, my_feat])
>>> my_fc.compute() 
FeatureCollectionResult(features=(
    FeatureResult(geometry=GeometryResult(type=Point, coordinates=[1, 2]), properties={'foo': 'bar'}),
    FeatureResult(geometry=GeometryResult(type=Point, coordinates=[1, 2]), properties={'foo': 'bar'}),
    FeatureResult(geometry=GeometryResult(type=Point, coordinates=[1, 2]), properties={'foo': 'bar'})))

features¶

List of FeatureResult result objects in the collection.

Type:	list[FeatureResult]

__geo_interface__¶

GeoJSON representation of the FeatureCollectionResult, following the Python __geo_interface__ convention.

Type:	dict