Config

from tmlt.analytics import Config

class tmlt.analytics.Config(*args, **kwargs)

Bases: object

Configuration for Tumult Analytics.

class Features

Bases: object

Toggles for experimental features in Tumult Analytics.

Most users should not need to modify these. Any features that are disabled by default are experimental and should not be used in production. Their APIs may change at any time.

A particular feature can be enabled or disabled by using the methods on its FeatureFlag instance, for example:

Config().features.example_feature.enable()

The current list of available features is:

• auto_partition_selection: Automatically performs differentially private partition selection when passing a list of columns to groupby(). This requires evaluating the query with an ApproxDPBudget.