LogoLogo
v0.21-branch
v0.21-branch
  • Introduction
  • Community
  • Roadmap
  • Changelog
  • Getting started
    • Quickstart
    • Concepts
      • Overview
      • Data source
      • Entity
      • Feature view
      • Feature retrieval
      • Point-in-time joins
      • Dataset
    • Architecture
      • Overview
      • Feature repository
      • Registry
      • Offline store
      • Online store
      • Provider
    • Learning by example
    • Third party integrations
    • FAQ
  • Tutorials
    • Overview
    • Driver ranking
    • Fraud detection on GCP
    • Real-time credit scoring on AWS
    • Driver stats on Snowflake
    • Validating historical features with Great Expectations
  • How-to Guides
    • Running Feast with Snowflake/GCP/AWS
      • Install Feast
      • Create a feature repository
      • Deploy a feature store
      • Build a training dataset
      • Load data into the online store
      • Read features from the online store
    • Running Feast in production
    • Deploying a Java feature server on Kubernetes
    • Upgrading from Feast 0.9
    • Adding a custom provider
    • Adding a new online store
    • Adding a new offline store
    • Adding or reusing tests
  • Reference
    • Data sources
      • File
      • Snowflake
      • BigQuery
      • Redshift
      • Push
      • Spark (contrib)
      • PostgreSQL (contrib)
    • Offline stores
      • File
      • Snowflake
      • BigQuery
      • Redshift
      • Spark (contrib)
      • PostgreSQL (contrib)
    • Online stores
      • SQLite
      • Redis
      • Datastore
      • DynamoDB
      • PostgreSQL (contrib)
    • Providers
      • Local
      • Google Cloud Platform
      • Amazon Web Services
    • Feature repository
      • feature_store.yaml
      • .feastignore
    • Feature servers
      • Python feature server
      • Go-based feature retrieval
    • [Alpha] Web UI
    • [Alpha] Data quality monitoring
    • [Alpha] On demand feature view
    • [Alpha] AWS Lambda feature server
    • Feast CLI reference
    • Python API reference
    • Usage
  • Project
    • Contribution process
    • Development guide
    • Versioning policy
    • Release process
    • Feast 0.9 vs Feast 0.10+
Powered by GitBook
On this page
  • Overview
  • CLI
  • Example
  • Registering transformations
  • Feature retrieval

Was this helpful?

Edit on GitHub
Export as PDF
  1. Reference

[Alpha] On demand feature view

Previous[Alpha] Data quality monitoringNext[Alpha] AWS Lambda feature server

Last updated 2 years ago

Was this helpful?

Warning: This is an experimental feature. It's intended for early testing and feedback, and could change without warnings in future releases.

To enable this feature, run feast alpha enable on_demand_transforms

Overview

On demand feature views allows users to use existing features and request time data (features only available at request time) to transform and create new features. Users define python transformation logic which is executed in both historical retrieval and online retrieval paths.

Currently, these transformations are executed locally. Future milestones include building a Feature Transformation Server for executing transformations at higher scale.

CLI

There are new CLI commands:

  • feast on-demand-feature-views list lists all registered on demand feature view after feast apply is run

  • feast on-demand-feature-views describe [NAME] describes the definition of an on demand feature view

Example

See for an example on how to use on demand feature views.

Registering transformations

We register RequestSource inputs and the transform in on_demand_feature_view:

from feast import Field, RequestSource
from feast.types import Float64, Int64
import pandas as pd

# Define a request data source which encodes features / information only
# available at request time (e.g. part of the user initiated HTTP request)
input_request = RequestSource(
    name="vals_to_add",
    schema=[
        Field(name='val_to_add', dtype=Int64),
        Field(name='val_to_add_2', dtype=Int64)
    ]
)

# Use the input data and feature view features to create new features
@on_demand_feature_view(
   sources=[
       driver_hourly_stats_view,
       input_request
   ],
   schema=[
     Field(name='conv_rate_plus_val1', dtype=Float64),
     Field(name='conv_rate_plus_val2', dtype=Float64)
   ]
)
def transformed_conv_rate(features_df: pd.DataFrame) -> pd.DataFrame:
    df = pd.DataFrame()
    df['conv_rate_plus_val1'] = (features_df['conv_rate'] + features_df['val_to_add'])
    df['conv_rate_plus_val2'] = (features_df['conv_rate'] + features_df['val_to_add_2'])
    return df

Feature retrieval

The on demand feature view's name is the function name (i.e. transformed_conv_rate).

And then to retrieve historical or online features, we can call this in a feature service or reference individual features:

training_df = store.get_historical_features(
    entity_df=entity_df,
    features=[
        "driver_hourly_stats:conv_rate",
        "driver_hourly_stats:acc_rate",
        "driver_hourly_stats:avg_daily_trips",
        "transformed_conv_rate:conv_rate_plus_val1",
        "transformed_conv_rate:conv_rate_plus_val2",
    ],
).to_df()
https://github.com/feast-dev/on-demand-feature-views-demo