Parquet example

Python

from featurestore import Client, ParquetFile

# Initialise feature store client
client = Client("ip:port")
client.auth.login()

# Set project specifics
project = client.projects.create("demo")

# Create the parquet source
parquet = ParquetFile("<path to parquet file>")
parquet_schema = client.extract_schema_from_source(parquet)

# Register the feature set
my_feature_set = project.feature_sets.register(parquet_schema, "feature_set_name", primary_key=["key_name"])

# Ingest to cache
my_feature_set.ingest(parquet)

# Retrieve feature set
ref = my_feature_set.retrieve()
ref.download()

Scala

import ai.h2o.featurestore.Client
import ai.h2o.featurestore.core.sources.ParquetFile

// Initialise feature store client
val client = Client("url")
client.auth.login()

// Set project specifics
val project = client.projects.create("demo")

// Create the parquet source
val parquet = ParquetFile("<path to parquet file>")
val parquetSchema = client.extractSchemaFromSource(parquet)

// Register the feature set
val myFeatureSet = project.featureSets.register(parquetSchema, "featureSetName", primaryKey=Seq("keyName"))

// Ingest to cache
myFeatureSet.ingest(parquet)

// Retrieve feature set
val ref = myFeatureSet.retrieve()
ref.download()

---

Feedback

• Submit and view feedback for this page
• Send feedback about H2O Feature Store to cloud-feedback@h2o.ai