features | Dataset Property | Teradata Package for Python - features - Teradata Package for Python

Use the features property to return feature objects associated with this dataset.

Example setup

>>> from teradataml import load_example_data, FeatureStore
>>> load_example_data('dataframe', 'sales')
>>> df = DataFrame("sales")

Create a feature store.

>>> fs = FeatureStore(repo='vfs_v1', data_domain='sales')
>>> fs.setup()

Repo vfs_v1 does not exist. Run FeatureStore.setup() to create the repo and setup FeatureStore.
True

Create a FeatureProcess object and ingest features on existing repo 'vfs_v1'.

>>> from teradataml import FeatureProcess
>>> fp = FeatureProcess(repo="vfs_v1",
...                     data_domain='sales',
...                     object=df,
...                     entity="accounts",
...                     features=["Jan", "Feb", "Mar", "Apr"])
>>> fp.run()

Process 'eadf3787-4ad4-11f0-8afd-f020ffe7fe09' started.
Process 'eadf3787-4ad4-11f0-8afd-f020ffe7fe09' completed.

Build a dataset.

>>> dc = DatasetCatalog(repo='vfs_v1', data_domain='sales')
>>> dataset = dc.build_dataset(entity='accounts',
...                            selected_features = {
...                             'Jan': 'eadf3787-4ad4-11f0-8afd-f020ffe7fe09',
...                             'Feb': 'eadf3787-4ad4-11f0-8afd-f020ffe7fe09'},
...                            view_name='ds_jan_feb',
...                            description='Dataset with Jan and Feb features')

List available datasets.

>>> dc.list_datasets()

                                     data_domain        name entity_name                        description                      valid_start                       valid_end
id                                                                                                                                                                        
abbde025-83b3-4cd8-bb72-57c40ba68f49       sales  ds_jan_feb    accounts  Dataset with Jan and Feb features   2025-06-12 12:06:15.572420+00:  9999-12-31 23:59:59.999999+00:

Use one of the dataset IDs to create a dataset object.

>>> ds = Dataset(repo='vfs_v1',
...              id='abbde025-83b3-4cd8-bb72-57c40ba68f49',
...              data_domain='sales')

Example: Retrieve the features

>>> ds.features

[Feature(name=Jan), Feature(name=Feb)]