archive_feature_process() | FeatureStore Archive Method | Teradata Package for Python - archive_feature_process() - Teradata Package for Python

Use the archive_feature_process() method to archive the feature process with the given process_id.

• An archived feature process is not available for any further processing.
• Archived FeatureProcess can be viewed using FeatureStore.list_feature_processes(archived=True) method.

Required Parameter

process_id

Specifies the ID of the FeatureProcess to archive from repository.

Example setup

>>> load_example_data('dataframe', ['sales'])

Create a teradataml DataFrame.

>>> from teradataml import DataFrame, FeatureProcess, FeatureStore
>>> df = DataFrame("sales")

Create FeatureStore for repo 'repo'.

>>> fs = FeatureStore("repo", data_domain='sales')

Repo repo does not exist. Run FeatureStore.setup() to create the repo and setup FeatureStore.

Set up FeatureStore for this repository.

>>> fs.setup()

True

Run FeatureProcess to ingest features.

>>> from teradataml import FeatureProcess
>>> fp = FeatureProcess(repo='repo',
...                     data_domain='sales',
...                     object=df,
...                     entity='accounts',
...                     features=['Jan', 'Feb', 'Mar', 'Apr'])
>>> fp.run()

Process '2a014f2d-6b71-11f0-aeda-f020ffe7fe09' started.
Process '2a014f2d-6b71-11f0-aeda-f020ffe7fe09' completed.

List the available feature processes.

>>> fs.list_feature_processes()

                                     description data_domain       process_type data_source entity_id       feature_names feature_ids                     valid_start                       valid_end
process_id                                                                                                                                                                                          
2a014f2d-6b71-11f0-aeda-f020ffe7fe09                   sales  denormalized view     "sales"  accounts  Apr, Feb, Jan, Mar        None  2025-07-28 05:10:34.760000+00:  9999-12-31 23:59:59.999999+00:

Example: Archive the FeatureProcess with process_id '2a014f2d-6b71-11f0-aeda-f020ffe7fe09'

>>> fs.archive_feature_process("2a014f2d-6b71-11f0-aeda-f020ffe7fe09")

Feature 'Jan' is archived from table 'FS_T_a38baff6_821b_3bb7_0850_827fe5372e31'.
Feature 'Jan' is archived from metadata.
Feature 'Feb' is archived from table 'FS_T_6003dc24_375e_7fd6_46f0_eeb868305c4a'.
Feature 'Feb' is archived from metadata.
Feature 'Mar' is archived from table 'FS_T_a38baff6_821b_3bb7_0850_827fe5372e31'.
Feature 'Mar' is archived from metadata.
Feature 'Apr' is archived from table 'FS_T_a38baff6_821b_3bb7_0850_827fe5372e31'.
Feature 'Apr' is archived from metadata.
FeatureProcess with process id '2a014f2d-6b71-11f0-aeda-f020ffe7fe09' is archived.
True