15. Deploy and Serve a Feature List

Deploy and Serve a Feature List for Model Predictions¶

Once you have a feature list ready, it's essential to serve these features effectively to make real-time or batch predictions using your machine learning model.

In this section, we'll take the feature list we previously crafted and explore two primary ways to serve its values:

REST API: Ideal for real-time predictions where you need instantaneous results. For instance, in applications where user interactions require immediate feedback based on model predictions.

Batch Processing: Best suited for scenarios where you have a bulk of data and don't require instant results.

Important Note for FeatureByte Enterprise Users¶

In Catalogs with Approval Flow enabled, moving features to production-ready status involves a comprehensive approval process.

This includes several evaluations, such as checking the feature's compliance with default cleaning operations and the feature job setting of its source tables. It also involves confirming the status of these tables and backtesting the feature job setting to prevent future training-serving inconsistencies. Additionally, essential details of the feature, particularly its feature definition file, are shared and subjected to a thorough review.

In [1]:

import featurebyte as fb
from datetime import datetime

import featurebyte as fb from datetime import datetime

14:13:36 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:13:36 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.

Activate Catalog¶

In [2]:

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

catalog_name = "Loan Applications Dataset SDK Tutorial"
catalog = fb.Catalog.activate(catalog_name)

# Set your profile to the tutorial environment fb.use_profile("tutorial") catalog_name = "Loan Applications Dataset SDK Tutorial" catalog = fb.Catalog.activate(catalog_name)

14:02:26 | INFO     | Using profile: tutorial
INFO    :featurebyte:Using profile: tutorial
14:02:27 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
INFO    :featurebyte:Using configuration file at: /Users/gxav/.featurebyte/config.yaml
14:02:27 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
INFO    :featurebyte:Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
14:02:27 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:02:27 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.
14:02:27 | INFO     | Catalog activated: Loan Applications Dataset SDK Tutorial
INFO    :featurebyte.api.catalog:Catalog activated: Loan Applications Dataset SDK Tutorial
16:13:11 | WARNING  | Remote SDK version (1.1.0.dev7) is different from local (1.1.0.dev1). Update local SDK to avoid unexpected behavior.
16:13:11 | INFO     | No catalog activated.
16:13:11 | INFO     | Catalog activated: Grocery Dataset Tutorial

List feature lists in Catalog¶

In [3]:

catalog.list_feature_lists()

catalog.list_feature_lists()

Out[3]:

	id	name	num_feature	status	deployed	readiness_frac	online_frac	tables	entities	primary_entity	created_at
0	683d3c6ea3c52cb2862eea26	40 features for Loan Applications	40	DRAFT	False	0.0	0.0	[NEW_APPLICATION, CLIENT_PROFILE, BUREAU, INST...	[New Application, Client]	[New Application]	2025-06-02T05:53:57.964000

Get a feature list from Catalog¶

In [4]:

feature_list_name = "40 features for Loan Applications"
simple_feature_list = catalog.get_feature_list(feature_list_name)

feature_list_name = "40 features for Loan Applications" simple_feature_list = catalog.get_feature_list(feature_list_name)

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

Deploy feature list¶

In [5]:

deployment_name = "Loan Application: simple feature list with 40 features"
# Create a deployment
deployment = simple_feature_list.deploy(
  deployment_name=deployment_name,
  make_production_ready=True,
  use_case_name="Loan Default by client",
)

deployment_name = "Loan Application: simple feature list with 40 features" # Create a deployment deployment = simple_feature_list.deploy( deployment_name=deployment_name, make_production_ready=True, use_case_name="Loan Default by client", )

Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         

In [6]:

# Enable deployment
deployment.enable()

# Enable deployment deployment.enable()

Done! |████████████████████████████████████████| 100% in 5:38.7 (0.00%/s)       
Done! |████████████████████████████████████████| 100% in 1:49.2 (0.01%/s)       

In [7]:

# Check that the deployment is enabled
catalog.list_deployments()

# Check that the deployment is enabled catalog.list_deployments()

Out[7]:

	id	name	feature_list_name	feature_list_version	num_feature	enabled
0	683d3e7a52b8c1c6652ed315	Loan Application: simple feature list with 40 ...	40 features for Loan Applications	V250602	40	True

In [8]:

# Check status of the feature list
print("simple_feature_list.status:", simple_feature_list.status)

# Check status of the feature list print("simple_feature_list.status:", simple_feature_list.status)

simple_feature_list.status: DEPLOYED

Get template for online serving¶

In [9]:

# Get a python template for consuming the feature serving API
deployment.get_online_serving_code(language="python")

# Get a python template for consuming the feature serving API deployment.get_online_serving_code(language="python")

Out[9]:

from typing import Any, Dict

import pandas as pd
import requests


def request_features(entity_serving_names: Dict[str, Any]) -> pd.DataFrame:
    """
    Send POST request to online serving endpoint

    Parameters
    ----------
    entity_serving_names: Dict[str, Any]
        Entity serving name values to used for serving request

    Returns
    -------
    pd.DataFrame
    """
    response = requests.post(
        url="https://tutorials.featurebyte.com/api/v1/deployment/683d3e7a52b8c1c6652ed315/online_features",
        headers={"Content-Type": "application/json", "active-catalog-id": "683d3958cd5508b7e42fde58", "Authorization": "Bearer alZJY8HvLxoi_emU_eukghMpSMzxL0eQ4ufUSWxvsLk"},
        json={"entity_serving_names": entity_serving_names},
    )
    assert response.status_code == 200, response.json()
    return pd.DataFrame.from_dict(response.json()["features"])


request_features([{"SK_ID_CURR": 367827}])

In [10]:

# Get shell script
deployment.get_online_serving_code(language="sh")

# Get shell script deployment.get_online_serving_code(language="sh")

Out[10]:

#!/bin/sh

curl -X POST \
    -H 'Content-Type: application/json' \
    -H 'active-catalog-id: 683d3958cd5508b7e42fde58' \
    -H 'Authorization: Bearer alZJY8HvLxoi_emU_eukghMpSMzxL0eQ4ufUSWxvsLk' \
    -d '{"entity_serving_names": [{"SK_ID_CURR": 367827}]}' \
    https://tutorials.featurebyte.com/api/v1/deployment/683d3e7a52b8c1c6652ed315/online_features

Create batch request table¶

Let's run a batch request for the last 24 hours applications.

In [11]:

# Get view of new applications
application_view = catalog.get_view("NEW_APPLICATION")

# Get view of new applications application_view = catalog.get_view("NEW_APPLICATION")

In [12]:

from datetime import datetime, timedelta, timezone
import pandas as pd
date_time_now_utc = datetime.now(timezone.utc)
now_less_24h = date_time_now_utc - timedelta(hours=24)
cond = application_view.application_time > pd.Timestamp(now_less_24h)
recent_application_view = application_view[cond]

from datetime import datetime, timedelta, timezone import pandas as pd date_time_now_utc = datetime.now(timezone.utc) now_less_24h = date_time_now_utc - timedelta(hours=24) cond = application_view.application_time > pd.Timestamp(now_less_24h) recent_application_view = application_view[cond]

In [13]:

recent_application_view.preview()

recent_application_view.preview()

Out[13]:

	SK_ID_CURR	ClientID	CONTRACT_TYPE	AMT_INCOME_TOTAL	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_PRICE	TYPE_SUITE	REGION_POPULATION_RELATIVE	application_time	...	FLAG_DOCUMENT_18	FLAG_DOCUMENT_19	FLAG_DOCUMENT_20	FLAG_DOCUMENT_21	AMT_REQ_CREDIT_BUREAU_HOUR	AMT_REQ_CREDIT_BUREAU_DAY	AMT_REQ_CREDIT_BUREAU_WEEK	AMT_REQ_CREDIT_BUREAU_MON	AMT_REQ_CREDIT_BUREAU_QRT	AMT_REQ_CREDIT_BUREAU_YEAR
0	370675	370675	Cash loans	225000.0	248760.0	28134.0	225000.0	Unaccompanied	0.025164	2025-06-02 03:21:48	...	0	0	0	0	NaN	NaN	NaN	NaN	NaN	NaN
1	370878	370878	Cash loans	90000.0	485190.0	24903.0	405000.0	Unaccompanied	0.005144	2025-06-02 00:59:30	...	0	0	0	0	0.0	0.0	0.0	0.0	0.0	2.0
2	373328	373328	Cash loans	112500.0	900000.0	38133.0	900000.0	Unaccompanied	0.018801	2025-06-01 16:03:25	...	0	0	0	0	0.0	0.0	0.0	0.0	0.0	4.0
3	374568	374568	Cash loans	292500.0	690660.0	54697.5	589500.0	Unaccompanied	0.046220	2025-06-02 03:16:18	...	0	0	0	0	0.0	0.0	0.0	0.0	1.0	3.0
4	376354	376354	Cash loans	157500.0	254799.0	20259.0	193500.0	Unaccompanied	0.010032	2025-06-01 20:14:39	...	0	0	0	0	0.0	0.0	0.0	0.0	0.0	3.0
5	376494	376494	Cash loans	270000.0	202500.0	16128.0	202500.0	Unaccompanied	0.026392	2025-06-02 02:04:27	...	0	0	0	0	0.0	0.0	0.0	0.0	1.0	1.0
6	381735	381735	Cash loans	225000.0	450000.0	47254.5	450000.0	Unaccompanied	0.010006	2025-06-01 17:31:43	...	0	0	0	0	0.0	0.0	0.0	0.0	0.0	4.0
7	382140	382140	Cash loans	72000.0	450000.0	22977.0	450000.0	Unaccompanied	0.008068	2025-06-01 13:43:16	...	0	0	0	0	NaN	NaN	NaN	NaN	NaN	NaN
8	384065	384065	Revolving loans	270000.0	180000.0	9000.0	180000.0	Unaccompanied	0.009334	2025-06-01 19:56:58	...	0	0	0	0	0.0	0.0	0.0	0.0	1.0	2.0
9	388505	388505	Cash loans	216000.0	817560.0	30820.5	675000.0	Unaccompanied	0.030755	2025-06-01 18:30:07	...	0	0	0	0	0.0	0.0	0.0	0.0	0.0	0.0

10 rows × 111 columns

In [14]:

# Check primary entity of the deployed feature list to obtain its serving name
simple_feature_list.primary_entity

# Check primary entity of the deployed feature list to obtain its serving name simple_feature_list.primary_entity

Out[14]:

[<featurebyte.api.entity.Entity at 0x302c7da80>
 {
   'name': 'New Application',
   'created_at': '2025-06-02T05:47:02.488000',
   'updated_at': '2025-06-02T05:47:03.887000',
   'description': None,
   'serving_names': [
     'SK_ID_CURR'
   ],
   'catalog_name': 'Loan Applications Dataset SDK Tutorial'
 }]

In [15]:

# Create batch request table from the view
batch_request_table = recent_application_view.create_batch_request_table(
    name="Latest Applications the latest 24 hours at " + date_time_now_utc.strftime("%Y%m%d:%H%M"),
    columns=["SK_ID_CURR"],
    columns_rename_mapping={
        "SK_ID_CURR": "SK_ID_CURR",
    }
)

# Create batch request table from the view batch_request_table = recent_application_view.create_batch_request_table( name="Latest Applications the latest 24 hours at " + date_time_now_utc.strftime("%Y%m%d:%H%M"), columns=["SK_ID_CURR"], columns_rename_mapping={ "SK_ID_CURR": "SK_ID_CURR", } )

Done! |████████████████████████████████████████| 100% in 9.2s (0.11%/s)         
Done! |████████████████████████████████████████| 100% in 12.2s (0.08%/s)        

In [16]:

# Get name of the batch request table
batch_request_table.name

# Get name of the batch request table batch_request_table.name

Out[16]:

'Latest Applications the latest 24 hours at 20250602:0608'

In [17]:

# List batch request tables in catalog
catalog.list_batch_request_tables()

# List batch request tables in catalog catalog.list_batch_request_tables()

Out[17]:

	id	name	type	shape	feature_store_name	created_at
0	683d3fdd52b8c1c6652ed316	Latest Applications the latest 24 hours at 202...	view	[132, 1]	playground	2025-06-02T06:08:33.231000

Compute batch feature table¶

In [18]:

# Get deployment and batch request table
deployment = catalog.get_deployment(deployment_name)
batch_request_table = catalog.get_batch_request_table(batch_request_table.name)

# Get deployment and batch request table deployment = catalog.get_deployment(deployment_name) batch_request_table = catalog.get_batch_request_table(batch_request_table.name)

In [19]:

# Compute batch features
batch_features = deployment.compute_batch_feature_table(
    batch_request_table=batch_request_table,
    batch_feature_table_name = 
    f"{feature_list_name} with {batch_request_table.name}"
)

# Compute batch features batch_features = deployment.compute_batch_feature_table( batch_request_table=batch_request_table, batch_feature_table_name = f"{feature_list_name} with {batch_request_table.name}" )

Done! |████████████████████████████████████████| 100% in 58.2s (0.02%/s)        
Done! |████████████████████████████████████████| 100% in 12.1s (0.08%/s)        

In [20]:

# List observation tables
catalog.list_batch_feature_tables()

# List observation tables catalog.list_batch_feature_tables()

Out[20]:

	id	name	feature_store_name	batch_request_table_name	shape	created_at
0	683d3fe752b8c1c6652ed317	40 features for Loan Applications with Latest ...	playground	Latest Applications the latest 24 hours at 202...	[132, 41]	2025-06-02T06:09:32.045000

In [21]:

# Convert to pandas 
batch_features.to_pandas()

# Convert to pandas batch_features.to_pandas()

Downloading table |████████████████████████████████████████| 132/132 [100%] in 0

Out[21]:

	SK_ID_CURR	NEW_APPLICATION_EXT_SOURCE_2	NEW_APPLICATION_EXT_SOURCE_3	NEW_APPLICATION_EXT_SOURCE_1	NEW_APPLICATION_Credit-Goods_Difference	NEW_APPLICATION_AMT_ANNUITY_To_AMT_CREDIT	NEW_APPLICATION_DAYS_EMPLOYED	CLIENT_GENDER	CLIENT_Max_of_Active_Cr_active_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_104w	CLIENT_EDUCATION_TYPE	...	CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_PAYMENT_TYPE_24cMo	CLIENT_Time_To_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w	NEW_APPLICATION_FLAG_DOCUMENT_3	NEW_APPLICATION_FLOORSMAX_MODE	CLIENT_Max_of_Installments_Days_Difference_Actual_vs_Scheduleds_24cMo	CLIENT_Max_of_Installments_PriorApplication_AMT_ANNUITY_To_AMT_CREDITs_6cMo	CLIENT_Std_of_Credit_card_monthly_balance_records_CNT_DRAWINGS_ATM_CURRENTs_24cMo	CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_NFLAG_INSURED_ON_APPROVAL_52w	CLIENT_Std_of_BureauReportedCredits_Available_Credits_26w	CLIENT_Max_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_52w
0	370675	0.688276	NaN	0.653914	23760.0	0.113097	-1828.0	M	NaN	Secondary / secondary special	...	None	-29.883851	1	NaN	NaN	NaN	NaN	{\n "0": 1.345500000000000e+05\n}	NaN	NaN
1	373328	0.340961	0.622922	NaN	0.0	0.042370	-3056.0	F	0.180025	Secondary / secondary special	...	None	NaN	1	NaN	NaN	NaN	NaN	None	0.000000	NaN
2	374568	0.566644	0.236611	0.367467	101160.0	0.079196	-2452.0	M	0.823229	Higher education	...	None	-128.880031	1	NaN	NaN	NaN	NaN	{\n "0": 1.955970000000000e+05\n}	120068.584626	NaN
3	390276	0.596397	NaN	0.637626	0.0	0.064453	NaN	F	0.956667	Secondary / secondary special	...	None	-22.864047	1	0.4583	NaN	NaN	NaN	{\n "1": 3.575700000000000e+04\n}	8780.969124	NaN
4	398384	0.766377	0.362277	0.635817	0.0	0.079007	-517.0	F	1.000000	Higher education	...	None	-78.316686	1	0.0417	NaN	NaN	NaN	{\n "0": 4.784400000000000e+04,\n "__MISSING...	89383.302862	1.000000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
127	355417	0.772387	0.794629	0.783045	143271.0	0.035918	-6312.0	F	0.916667	Secondary / secondary special	...	None	NaN	1	0.1667	NaN	NaN	NaN	None	0.000000	0.916667
128	356857	0.510737	0.360613	NaN	95427.0	0.042536	NaN	F	0.884400	Secondary / secondary special	...	None	-216.530089	0	NaN	NaN	NaN	NaN	{\n "1": 1.935720000000000e+05,\n "__MISSING...	13372.114158	0.525827
129	357464	0.663915	0.144648	0.643336	0.0	0.063900	-1805.0	F	0.921416	Secondary / secondary special	...	None	NaN	1	0.1667	NaN	NaN	NaN	{\n "__MISSING__": 2.636865000000000e+05\n}	51613.575403	0.689758
130	358444	0.530524	0.483050	NaN	79596.0	0.079350	-1596.0	M	0.968565	Secondary / secondary special	...	None	-521.448075	1	NaN	NaN	NaN	NaN	{\n "__MISSING__": 0.000000000000000e+00\n}	15923.250000	0.968565
131	364068	0.074342	0.475850	0.843238	209088.0	0.040288	NaN	F	0.961179	Higher education	...	None	-88.804128	1	NaN	NaN	NaN	NaN	{\n "0": 1.514835000000000e+05,\n "__MISSING...	329985.617751	0.961179

132 rows × 41 columns

Downloading table |████████████████████████████████████████| 500/500 [100%] in 0

Out[21]:

	GROCERYCUSTOMERGUID	PRODUCTGROUP	CUSTOMER_Age_band	CUSTOMER_Latest_invoice_Amount	CUSTOMER_Count_of_invoice_14d	CUSTOMER_Avg_of_invoice_Amount_14d	CUSTOMER_Std_of_invoice_Amount_14d	CUSTOMER_Latest_invoice_Amount_Z_Score_to_invoice_Amount_28d	CUSTOMER_vs_OVERALL_item_TotalCost_across_product_ProductGroups_26w	CUSTOMER_x_PRODUCTGROUP_Sum_of_item_TotalCost_14d	CUSTOMER_x_PRODUCTGROUP_Time_Since_Latest_Timestamp
0	14f94723-4cae-4041-a11a-a044df1a5a5f	Fromages	60-64	14.28	NaN	NaN	NaN	1.000000	0.373275	NaN	1527.861052
1	d4559f7d-eb28-42c6-b47d-847de24952c2	Fromages	65-69	20.04	NaN	NaN	NaN	NaN	0.370545	NaN	2177.017718
2	1d363f8c-88f4-49eb-a107-4f1e2fbe6722	Fromages	25-29	73.16	NaN	NaN	NaN	NaN	0.781391	NaN	1118.766885
3	768e57df-7513-412b-a466-effd270ad6a6	Fromages	65-69	7.16	3.0	8.246667	4.395121	-0.644055	0.650126	NaN	1357.819107
4	48f6e550-ce63-457e-ad14-d511ee934661	Fromages	40-44	24.62	1.0	24.620000	0.000000	-0.219845	0.753619	NaN	475.681329
...	...	...	...	...	...	...	...	...	...	...	...
495	c1b3224a-5dbb-427d-87d3-3882c6bb1278	Fromages	50-54	6.69	NaN	NaN	NaN	NaN	0.104953	NaN	NaN
496	261d2979-0568-4544-8be0-32355049e565	Fromages	65-69	2.00	NaN	NaN	NaN	NaN	0.317982	NaN	NaN
497	825ee709-2f85-49e2-bec7-27bea06efe3a	Fromages	45-49	3.00	2.0	5.995000	2.995000	-0.990561	0.375129	NaN	NaN
498	ea68ce6b-b3c3-4ba5-a471-f11b1c5bd892	Fromages	45-49	4.50	NaN	NaN	NaN	NaN	NaN	NaN	NaN
499	b8b00613-2e48-4eed-8971-61718990a994	Fromages	45-49	10.00	2.0	6.000000	4.000000	1.264961	0.579119	NaN	NaN

500 rows × 11 columns

In [22]:

# download parquet file
batch_features.download()

# download parquet file batch_features.download()

Downloading table |████████████████████████████████████████| 132/132 [100%] in 0

Out[22]:

PosixPath('BATCH_FEATURE_TABLE_683d3fe72b2121b4644f2571.parquet')

Downloading table |████████████████████████████████████████| 500/500 [100%] in 0

Out[22]:

PosixPath('BATCH_FEATURE_TABLE_666959284c98806ac301afa0.parquet')

In [23]:

# delete if not needed any more
batch_features.delete()
batch_request_table.delete()

# delete if not needed any more batch_features.delete() batch_request_table.delete()

Done! |████████████████████████████████████████| 100% in 9.2s (0.11%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         

Manage Deployment¶

In [24]:

deployment = catalog.get_deployment(deployment_name)

deployment = catalog.get_deployment(deployment_name)

In [25]:

# get feature jobs status (this will produce meaningful results only once multiple jobs have been run)
# deployment.get_feature_jobs_status(job_history_window=12)

# get feature jobs status (this will produce meaningful results only once multiple jobs have been run) # deployment.get_feature_jobs_status(job_history_window=12)

In [26]:

# Disable deployment
deployment.disable()

# Disable deployment deployment.disable()

Done! |████████████████████████████████████████| 100% in 18.6s (0.05%/s)        
Done! |████████████████████████████████████████| 100% in 12.1s (0.08%/s)        

In [27]:

# The deployment is still part of the catalog but disabled
catalog.list_deployments()

# The deployment is still part of the catalog but disabled catalog.list_deployments()

Out[27]:

	id	name	feature_list_name	feature_list_version	num_feature	enabled
0	683d3e7a52b8c1c6652ed315	Loan Application: simple feature list with 40 ...	40 features for Loan Applications	V250602	40	False

Concepts in this tutorial¶

• Feature list deployment
• REST API and batch serving
• Feature jobs

SDK reference for¶

• Deployment
• Batch Request Table
• Batch Feature Table
• FeatureList.deploy()
• Deployment.enable()
• Deployment.get_online_serving_code()
• Deployment.compute batch feature_table()
• View.create_batch_request_table()
• SourceTable.create_batch_request_table()

In [ ]: