13. Compute Historical Feature Values

Compute historical feature values¶

Historical feature values are needed to train and test Machine Learning models.

Let's take the feature list we just created and compute feature values for a given observation table.

In [1]:

import featurebyte as fb

# Set your profile to the tutorial environment
fb.use_profile("tutorial")

catalog_name = "Credit Default Dataset SDK Tutorial"
catalog = fb.Catalog.activate(catalog_name)

import featurebyte as fb # Set your profile to the tutorial environment fb.use_profile("tutorial") catalog_name = "Credit Default Dataset SDK Tutorial" catalog = fb.Catalog.activate(catalog_name)

18:04:49 | INFO     | SDK version: 3.3.1
18:04:49 | INFO     | No catalog activated.
18:04:49 | INFO     | Using profile: tutorial
18:04:49 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
18:04:49 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
18:04:49 | INFO     | SDK version: 3.3.1
18:04:49 | INFO     | No catalog activated.
18:04:50 | INFO     | Catalog activated: Credit Default Dataset SDK Tutorial
16:12:25 | INFO     | No catalog activated.
16:12:25 | INFO     | Catalog activated: Grocery Dataset Tutorial

List feature lists in Catalog¶

In [2]:

catalog.list_feature_lists()

catalog.list_feature_lists()

Out[2]:

	id	name	num_feature	status	deployed	readiness_frac	online_frac	tables	entities	primary_entity	created_at
0	69315caf133daa6ca04988cd	40 features for Credit Default	40	DRAFT	False	0.0	0.0	[NEW_APPLICATION, CLIENT_PROFILE, BUREAU, INST...	[New Application, Client]	[New Application]	2025-12-04T10:04:36.571000

Get Feature List from Catalog¶

In [3]:

feature_list_name = "40 features for Credit Default"
simple_feature_list = catalog.get_feature_list(feature_list_name)

feature_list_name = "40 features for Credit Default" simple_feature_list = catalog.get_feature_list(feature_list_name)

Loading Feature(s) |████████████████████████████████████████| 40/40 [100%] in 0.

Get an observation table¶

In [4]:

# List observation tables
catalog.list_observation_tables()

# List observation tables catalog.list_observation_tables()

Out[4]:

	id	name	type	shape	feature_store_name	created_at
0	69315bf8565d52c79a28f8e1	Applications up to March 2025	source_table	[307511, 3]	playground	2025-12-04T10:01:41.876000

In [5]:

# Get observation table: 'Applications up to March 2025'
training_observations = catalog.get_observation_table("Applications up to March 2025")

# Get observation table: 'Applications up to March 2025' training_observations = catalog.get_observation_table("Applications up to March 2025")

Compute historical features¶

In [6]:

# Create training data
training_data_table = simple_feature_list.compute_historical_feature_table(
    training_observations,
    historical_feature_table_name=f"{feature_list_name} - TRAIN",
)

# Create training data training_data_table = simple_feature_list.compute_historical_feature_table( training_observations, historical_feature_table_name=f"{feature_list_name} - TRAIN", )

Done! |████████████████████████████████████████| 100% in 2:12.6 (0.01%/s)       
Done! |████████████████████████████████████████| 100% in 36.4s (0.03%/s)        

In [7]:

# List historical feature tables from catalog
catalog.list_historical_feature_tables()

# List historical feature tables from catalog catalog.list_historical_feature_tables()

Out[7]:

	id	name	feature_store_name	observation_table_name	shape	created_at
0	69315cc4da8963295e76f9d5	40 features for Credit Default - TRAIN	playground	Applications up to March 2025	[307511, 43]	2025-12-04T10:07:01.157000

Concepts in this tutorial¶

• More on Historical feature serving

SDK reference for¶

• Historical feature table
• FeatureList.compute historical feature table()
• FeatureList.compute_historical_features() to compute directly a data frame