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

16:53:36 | WARNING  | Service endpoint is inaccessible: http://featurebyte-server:8088/

Activate Catalog¶

In [2]:

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

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

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

16:53:36 | INFO     | Using profile: tutorial
16:53:36 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
16:53:36 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
16:53:36 | INFO     | SDK version: 2.1.0.dev113
16:53:36 | INFO     | No catalog activated.
16:53:36 | INFO     | Catalog activated: Credit Default 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	67c2c8de53e241b5a68dd616	51 features for Credit Default	51	DRAFT	False	0.0	0.0	[NEW_APPLICATION, PRIOR_APPLICATIONS, CONSUMER...	[New Application, Client]	[New Application]	2025-03-01T08:44:22.006000

Get a feature list from Catalog¶

In [4]:

feature_list_name = "51 features for Credit Default"
simple_feature_list = catalog.get_feature_list(feature_list_name)

feature_list_name = "51 features for Credit Default" simple_feature_list = catalog.get_feature_list(feature_list_name)

Loading Feature(s) |████████████████████████████████████████| 51/51 [100%] in 0.

Deploy feature list¶

In [5]:

deployment_name = "Early Stage Loan Application: 51 features for Credit Default"
# Create a deployment
deployment = simple_feature_list.deploy(
  deployment_name=deployment_name,
  make_production_ready=True,
)

deployment_name = "Early Stage Loan Application: 51 features for Credit Default" # Create a deployment deployment = simple_feature_list.deploy( deployment_name=deployment_name, make_production_ready=True, )

Done! |████████████████████████████████████████| 100% in 9.1s (0.11%/s)         |████████████████████████████████████████| ▆▄▂ 100% in 8s (~0s, 0.1%/
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 8:45.4 (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	67c2cb1a3be1dbfd5a465cee	Early Stage Loan Application: 51 features for ...	51 features for Credit Default	V250301	51	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/67c2cb1a3be1dbfd5a465cee/online_features",
        headers={"Content-Type": "application/json", "active-catalog-id": "67c2c726799a655db130206e", "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([{"NEW_APPLICATION_ID": 100002}])

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: 67c2c726799a655db130206e' \
    -H 'Authorization: Bearer alZJY8HvLxoi_emU_eukghMpSMzxL0eQ4ufUSWxvsLk' \
    -d '{"entity_serving_names": [{"NEW_APPLICATION_ID": 100002}]}' \
    https://tutorials.featurebyte.com/api/v1/deployment/67c2cb1a3be1dbfd5a465cee/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]:

	BIRTHDATE	CODE_GENDER	INCOME_TYPE	EDUCATION_TYPE	OCCUPATION_TYPE	ORGANIZATION_TYPE	APPLICATION_ID	CLIENT_ID	AMT_CREDIT	AMT_ANNUITY	AMT_GOODS_VALUE	REGION_POPULATION_RELATIVE	APPLICATION_TIME	DAYS_EMPLOYED	DAYS_REGISTRATION	DAYS_LAST_PHONE_CHANGE	FLOORSMAX_MEDI	LANDAREA_MEDI	FLAG_DOCUMENT_3	AMT_REQ_CREDIT_BUREAU_QRT
0	1988-07-27	F	Working	Incomplete higher	Laborers	Industry: type 12	100245	100245	364896.0	20947.5	315000.0	0.016612	2025-02-28 17:58:39	-1397	-5725.0	-2264.0	0.1667	0.0773	1	0.0
1	1971-12-15	M	Working	Secondary / secondary special	Drivers	Self-employed	100754	100754	364500.0	13221.0	364500.0	0.007330	2025-02-28 15:44:54	-553	-9402.0	-1714.0	0.3333	0.1034	1	1.0
2	1989-10-25	F	Working	Higher education	Core staff	Kindergarten	102315	102315	234000.0	12825.0	234000.0	0.018209	2025-02-28 09:40:13	-1465	-6899.0	-475.0	0.3333	0.0207	0	1.0
3	1987-06-29	F	Working	Higher education	NaN	Other	104044	104044	675000.0	21775.5	675000.0	0.003541	2025-02-28 14:27:50	-1905	-4475.0	-537.0	0.1667	0.0267	1	1.0
4	1984-10-29	F	Commercial associate	Higher education	Laborers	Trade: type 7	109015	109015	790830.0	62613.0	675000.0	0.001276	2025-02-28 11:44:47	-3087	-8855.0	-2447.0	0.1667	0.0964	1	1.0
5	1982-01-26	F	Working	Secondary / secondary special	Core staff	Self-employed	112271	112271	1141686.0	41139.0	922500.0	0.005144	2025-02-28 10:06:41	-861	-2457.0	0.0	0.1667	0.0911	1	0.0
6	1981-07-24	M	Working	Secondary / secondary special	Drivers	Business Entity Type 3	112973	112973	450000.0	28759.5	450000.0	0.035792	2025-02-28 14:29:59	-1435	-1487.0	-1994.0	0.0000	0.0246	0	1.0
7	1971-05-18	M	Working	Secondary / secondary special	Laborers	Business Entity Type 3	116726	116726	337500.0	17361.0	337500.0	0.007020	2025-02-28 10:12:28	-5046	-6519.0	-799.0	NaN	NaN	0	0.0
8	1984-03-18	F	State servant	Secondary / secondary special	Laborers	Other	119771	119771	280332.0	25839.0	234000.0	0.018850	2025-02-28 14:43:49	-1505	-878.0	-592.0	NaN	NaN	1	1.0
9	1984-12-08	M	Working	Secondary / secondary special	Laborers	Business Entity Type 2	128213	128213	104256.0	12501.0	90000.0	0.006305	2025-03-01 05:03:04	-2399	-8720.0	-1883.0	0.1667	0.0165	1	0.0

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 0x329245b70>
 {
   'name': 'New Application',
   'created_at': '2025-03-01T08:38:05.280000',
   'updated_at': '2025-03-01T08:38:07.461000',
   'description': None,
   'serving_names': [
     'NEW_APPLICATION_ID'
   ],
   'catalog_name': 'Credit Default 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=["APPLICATION_ID"],
    columns_rename_mapping={
        "APPLICATION_ID": "NEW_APPLICATION_ID",
    }
)

# 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=["APPLICATION_ID"], columns_rename_mapping={ "APPLICATION_ID": "NEW_APPLICATION_ID", } )

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 20250301:0902'

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	67c2cd363be1dbfd5a465cef	Latest Applications the latest 24 hours at 202...	view	[141, 1]	playground	2025-03-01T09:02:51.026000

Compute batch feature table¶

In [18]:

# Get deployment and batch request table
deployment = catalog.get_deployment("Early Stage Loan Application: 51 features for Credit Default")
batch_request_table = catalog.get_batch_request_table(batch_request_table.name)

# Get deployment and batch request table deployment = catalog.get_deployment("Early Stage Loan Application: 51 features for Credit Default") 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 1:12.9 (0.01%/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	67c2cd403be1dbfd5a465cf0	51 features for Credit Default with Latest App...	playground	Latest Applications the latest 24 hours at 202...	[141, 52]	2025-03-01T09:04:04.573000

In [21]:

# Convert to pandas 
batch_features.to_pandas()

# Convert to pandas batch_features.to_pandas()

Downloading table |████████████████████████████████████████| 141/141 [100%] in 0

Out[21]:

	NEW_APPLICATION_ID	NEW_APPLICATION_AMT_ANNUITY	NEW_APPLICATION_AMT_ANNUITY_To_AMT_CREDIT	NEW_APPLICATION_AMT_ANNUITY_To_AMT_GOODS_VALUE	NEW_APPLICATION_AMT_CREDIT	NEW_APPLICATION_AMT_GOODS_VALUE_To_AMT_CREDIT	NEW_APPLICATION_AMT_REQ_CREDIT_BUREAU_QRT	NEW_APPLICATION_CODE_GENDER	NEW_APPLICATION_Credit-Goods_Gap	NEW_APPLICATION_DAYS_EMPLOYED	...	CLIENT_Max_of_Approved_Status_Prior_Applications_AMT_ANNUITYs_104w	CLIENT_Min_of_Cash_loans_Contract_type_Prior_Applications_CNT_PAYMENTs_104w	CLIENT_Max_of_Prior_Applications_AMT_CREDIT_To_AMT_APPLICATIONs_104w	CLIENT_Max_of_Prior_Applications_Application-Credit_Gaps_104w	CLIENT_Max_of_Prior_Applications_CNT_PAYMENTs_104w	CLIENT_Pct_of_Prior_Applications_is_Refused_Status_52w	CLIENT_Pct_of_Prior_Applications_is_Refused_Status_Cash_loans_Contract_type_104w	CLIENT_Sum_of_Approved_Status_Prior_Applications_AMT_ANNUITYs_To_Sum_of_Approved_Status_Prior_Applications_AMT_CREDITs_104w	CLIENT_Time_Since_Latest_Prior_Application_DECISION_DATE_104w	CLIENT_vs_OVERALL_Count_of_Approved_Status_Cash_loans_Contract_type_Prior_Applications_by_Prior_Application_PRODUCT_COMBINATION_104w
0	100245	20947.5	0.057407	0.066500	364896.0	0.863260	0.0	F	49896.0	-1397.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1	100754	13221.0	0.036272	0.036272	364500.0	1.000000	1.0	M	0.0	-553.0	...	24928.065	60.0	1.396000	0.0	60.0	0.250000	0.000000	0.029394	73.470966	0.757184
2	102315	12825.0	0.054808	0.054808	234000.0	1.000000	1.0	F	0.0	-1465.0	...	NaN	60.0	1.198000	0.0	60.0	0.333333	0.277778	NaN	196.724219	NaN
3	104044	21775.5	0.032260	0.032260	675000.0	1.000000	1.0	F	0.0	-1905.0	...	10880.100	48.0	1.000000	0.0	48.0	0.000000	0.000000	0.109890	277.482818	NaN
4	109015	62613.0	0.079174	0.092760	790830.0	0.853534	1.0	F	115830.0	-3087.0	...	16673.940	NaN	1.066007	-9090.0	10.0	0.000000	0.000000	0.113580	260.596047	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
136	441014	18382.5	0.041263	0.041263	445500.0	1.000000	NaN	M	0.0	-261.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
137	442468	11875.5	0.044536	0.058644	266652.0	0.759417	0.0	F	64152.0	-1543.0	...	37550.700	36.0	1.118802	0.0	36.0	0.000000	0.000000	0.068231	155.709138	0.625046
138	450233	18918.0	0.042040	0.042040	450000.0	1.000000	1.0	F	0.0	-5855.0	...	1758.600	NaN	0.799835	4360.5	12.0	NaN	0.000000	0.100930	471.477587	NaN
139	453893	18261.0	0.079569	0.079569	229500.0	1.000000	0.0	F	0.0	-786.0	...	17653.275	24.0	1.211196	112500.0	24.0	NaN	0.000000	0.065458	393.620341	0.625046
140	453971	24840.0	0.079420	0.092000	312768.0	0.863260	1.0	M	42768.0	-3574.0	...	19646.685	36.0	1.138600	0.0	36.0	0.500000	0.200000	0.038345	99.617540	0.757184

141 rows × 52 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 |████████████████████████████████████████| 141/141 [100%] in 0

Out[22]:

PosixPath('BATCH_FEATURE_TABLE_67c2cd40306bec3def2f7249.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 24.4s (0.04%/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	67c2cb1a3be1dbfd5a465cee	Early Stage Loan Application: 51 features for ...	51 features for Credit Default	V250301	51	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 [ ]: