We have a GP Service that returns a FeatureSet.
The GP Service can be used using arcgis.geoprocessing.import_toolbox() and creates an arcgis.features.FeatureSet.
The arcgis.geoanalytics.summarize_data.reconstruct_tracks function requires an input layer (not a FeatureSet or a FeatureCollection). This is weird, as the corresponding REST endpoint does have support for FeatureCollections.
So is it possible to make use the reconstruct_tracks with any kind of an in-memory feature structure? How?
The input features to reconstruct_tracks allows Feature Collection to be passed in as a Python dictionary. You should be able to get the dict from a FeatureSet using it's value property and construct a FeatureCollection from it and pass it in the reconstruct_tracks tool.
I'm working with Jens on this issue. The problem is that the reconstruct_tracks expects input features that have a time set on the feature source and I don't know how to do this on a FeatureCollection.
The following example would fail with error below:
Any idea how to work around this?
Do the features in the FeatureSet have a time field? What does a feature in the FeatureSet look like (json)?
Reconstruct tracks uses the time to reconstruct the tracks. Another way to create FeatureSets from a set of GPS coordinates will be to use gis.content.import_data as in the Watson integration example at arcgis-python-api/ArcGIS Python API for Analysts and Data Scientists.ipynb at master · Esri/arcgis-p...