Working with Flowmachine

In some scenarios, the aggregates provided through FlowAPI will not be sufficient to meet your analytical needs. When this happens it is possible to use Flowmachine directly, which allows you to work with individual level data, build or customize metrics, and run arbitrary SQL.

Just like Flowclient, you can install Flowmachine using pip:

pip install flowmachine

See the mobile data usage example for a case which uses Flowmachine directly to produce a metric.

As with the other worked examples, you can try this using the quick start setup which includes JupyterLab. Alternatively, the Jupyter notebooks for this examples can be found here.