Architekturkomponenten in End-to-End-ML-Frameworks

End-to-End Machine Learning

Joshua Stapleton

Machine Learning Engineer

Feature Stores

Features

  • Feature Selection
  • Feature Engineering

Feature Store

  • Zentrales Repository für Features
  • Sichert Konsistenz, reduziert Duplikate
  • Ermöglicht Teilen und Auffinden
  • Standardisiert Transformationen und Berechnungen

ML-Lebenszyklus mit Feature Store

End-to-End Machine Learning

Feast

Feast

  • Beliebtes Tool zur Implementierung von Feature Stores
  • Vereinheitlicht Verwaltung, Speicherung, Serving und Discovery von ML-Features

Prinzipien

  • Features mit Feature-Sets definieren und registrieren
  • Feature-Sets: Gruppen verwandter Features + Metadaten

Beispiel: Herzkrankheits-Features

  • Patienten-Entität
  • Zugehörige Features (Cholesterin, Alter, Geschlecht)
End-to-End Machine Learning

Feast Feature Stores Teil 1

from feast import Field, Entity, ValueType, FeatureStore
from feast.data_source import FileSource

# Define the entity, which in this case is a patient, and features patient = Entity(name="patient", join_keys=["patient_id"])
chol = Field(name="chol", dtype=Float32) age = Field(name="age", dtype=Int32) ...
# Define the data source data_source = FileSource( path="/path_to_heart_disease_dataset.csv", event_timestamp_column="event_timestamp", created_timestamp_column="created")
End-to-End Machine Learning

Feast Feature Stores Teil 2

# ... continued
# Create a feature view of the data
heart_disease_fv = FeatureView(name="heart_disease", entities=[patient],
    schema=[cholesterol, ...], ttl=timedelta(days=1), input=data_source,)

# Create a FeatureStore object store = FeatureStore(repo_path=".")
# Register the FeatureView store.apply([patient, heart_disease_fv])
End-to-End Machine Learning

Modell-Registries

Modell-Registry

  • Versionskontrollsysteme
  • Verfolge verschiedene Modellversionen
  • Modelle annotieren
  • Performance über die Zeit tracken

Vorteile

  • Organisation
  • Transparenz
  • Reproduzierbarkeit

ML-Lebenszyklus mit Modell-Registry

End-to-End Machine Learning

Lass uns üben!

End-to-End Machine Learning

Preparing Video For Download...