feature-store-bundle 2.0.0b10

Feature Store for the Daipe AI Platform

Feature Store bundle

This package is distributed under the "DataSentics SW packages Terms of Use." See license

Feature store bundle allows you to store features with metadata.

Installation

poetry add feature-store-bundle

Getting started

1. Define entity and custom feature decorator

from featurestorebundle.entity.getter import get_entity
from featurestorebundle.feature.FeaturesStorage import FeaturesStorage
from featurestorebundle.notebook.decorator import feature_decorator_factory

entity = get_entity()
features_storage = FeaturesStorage(entity)

feature_decorator = feature_decorator_factory.create(entity, features_storage)

2. Use the feature decorator to save features as you create them

import daipe as dp

from pyspark.sql import functions as f
from pyspark.sql import DataFrame

@dp.transformation(dp.read_table("silver.tbl_loans"), display=True)
@feature_decorator(
    ("Age", "Client's age"),
    ("Gender", "Client's gender"),
    ("WorkExperience", "Client's work experience"),
    category="personal",
)
def client_personal_features(df: DataFrame):
    return (
        df.select("UserName", "Age", "Gender", "WorkExperience")
        .groupBy("UserName")
        .agg(
            f.max("Age").alias("Age"),
            f.first("Gender").alias("Gender"),
            f.first("WorkExperience").alias("WorkExperience"),
        )
        .withColumn("timestamp", f.lit(today))
    )

3. Write/Merge all features in one go

import daipe as dp
from featurestorebundle.feature.writer.FeaturesWriter import FeaturesWriter

@dp.notebook_function()
def write_features(writer: FeaturesWriter):
    writer.write(features_storage)