ML-Modell registrieren

See Extended Examples for complete configuration files and templates.

Implementieren MLflow Modellieren Registry for systematic model versioning, stage management, and deployment governance.

Wann verwenden

• Promoting a trained model from experimentation to production
• Managing multiple model versions across development stages
• Implementing model approval workflows for governance
• Tracking model lineage from training to deployment
• Rolling back to previous model versions
• Comparing deployed model versions for A/B testing
• Auditing model changes for compliance requirements

Eingaben

• Erforderlich: MLflow tracking server with Modellieren Registry enabled
• Erforderlich: Trained model logged with MLflow (from tracking runs)
• Erforderlich: Modellieren name for registry registration
• Optional: Approval workflow integration (email, Slack, Jira)
• Optional: CI/CD pipeline for automated promotion
• Optional: Modellieren validation metrics thresholds

Vorgehensweise

Schritt 1: Konfigurieren Modellieren Registry Backend

Einrichten MLflow Modellieren Registry with database backend (file-based registry not recommended for production).

# Start MLflow server with Model Registry support
mlflow server \
  --backend-store-uri postgresql://user:pass@localhost:5432/mlflow \
  --default-artifact-root s3://mlflow-artifacts/models \
  --host 0.0.0.0 \
  --port 5000

Python configuration:

# model_registry_config.py
import mlflow
from mlflow.tracking import MlflowClient

# Set tracking URI (must support Model Registry)
MLFLOW_TRACKING_URI = "http://mlflow-server.company.com:5000"
mlflow.set_tracking_uri(MLFLOW_TRACKING_URI)

# ... (see EXAMPLES.md for complete implementation)

Erwartet: Modellieren Registry UI tab appears in MLflow, search_registered_models() returns erfolgreich (even if empty), database contains registered_models table.

Bei Fehler: Verifizieren MLflow version ≥1.2 (Modellieren Registry introduced in 1.2), check database backend (SQLite not fully supported for Modellieren Registry), ensure --backend-store-uri points to database (not file://), verify database user has CREATE TABLE Berechtigungs, check MLflow server logs for migration errors.

Schritt 2: Registrieren Modellieren from Training Run

Registrieren a logged model to the Modellieren Registry with comprehensive metadata.

# register_model.py
import mlflow
from mlflow.tracking import MlflowClient
from model_registry_config import MLFLOW_TRACKING_URI

mlflow.set_tracking_uri(MLFLOW_TRACKING_URI)
client = MlflowClient()

# ... (see EXAMPLES.md for complete implementation)

Erwartet: New model version appears in Modellieren Registry UI, version includes description and tags, model artifacts are accessible via models:/<model-name>/<version> URI, model signature and input example are preserved.

Bei Fehler: Verifizieren run_id exists and has completed (client.get_run(run_id)), check model artifact path matches logged artifact (mlflow.search_runs() to inspect), ensure model was logged with proper framework flavor (mlflow.sklearn.log_model not mlflow.log_artifact), verify no special characters in model name (use hyphens not underscores), check artifact storage accessibility.

Schritt 3: Implementieren Stage Transitions with Validation

Move model versions durch stages (None → Staging → Production → Archived) with validation checks.

# stage_management.py
import mlflow
from mlflow.tracking import MlflowClient
from datetime import datetime

client = MlflowClient()

class ModelStageManager:
# ... (see EXAMPLES.md for complete implementation)

Erwartet: Modellieren version stage updates in registry, old versions archived automatisch, transition timestamps recorded in tags, rollback restores previous production version.

Bei Fehler: Check version exists and is in expected stage, verify archive_existing_versions flag behavior (may not archive if only one version), ensure database supports concurrent transactions for stage updates, check for stage transition locks (only one transition per version at a time), verify approval workflow integration.

Schritt 4: Implementieren Modellieren Aliasing and References

Use model aliases for stable deployment references (MLflow ≥2.0).

# model_aliases.py
from mlflow.tracking import MlflowClient

client = MlflowClient()

def set_model_alias(model_name, version, alias):
    """
    Set an alias for a model version (MLflow 2.0+).
# ... (see EXAMPLES.md for complete implementation)

Erwartet: Aliases appear in Modellieren Registry UI, loading models by alias works (models:/name@alias), updating alias sofort affects new loads, A/B test infrastructure functional.

Bei Fehler: Upgrade MLflow to ≥2.0 for native alias support, use tag-based fallback for older versions, verify alias naming (alphanumeric and hyphens only), check for alias conflicts (one alias per model version).

Schritt 5: Implementieren Modellieren Lineage Tracking

Verfolgen full lineage from data to deployment with comprehensive metadata.

# model_lineage.py
import mlflow
from mlflow.tracking import MlflowClient
import json

client = MlflowClient()

def enrich_model_metadata(model_name, version, lineage_data):
# ... (see EXAMPLES.md for complete implementation)

Erwartet: Modellieren version tags include comprehensive lineage information, get_model_lineage() returns full history, JSON report contains Datenquelle, training details, and deployment info.

Bei Fehler: Verifizieren tag values are strings (convert dicts to JSON), check tag key naming (no spaces or special chars), ensure lineage data captured waehrend training, verify run_id is valid and accessible.

Schritt 6: Automate Registry Operations with CI/CD

Integrieren model registration into CI/CD pipelines for automated promotion.

# .github/workflows/model_promotion.yml
name: Model Promotion Pipeline

on:
  workflow_dispatch:
    inputs:
      model_name:
        description: 'Model name to promote'
# ... (see EXAMPLES.md for complete implementation)

Python automation script:

# scripts/promote_model.py
import argparse
from stage_management import ModelStageManager

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("--model-name", required=True)
    parser.add_argument("--version", type=int, required=True)
# ... (see EXAMPLES.md for complete implementation)

Erwartet: GitHub Actions workflow triggers on manual dispatch, validation tests pass, model promoted to target stage, Slack notification sent, deployment pipeline triggered automatisch.

Bei Fehler: Check GitHub secrets configuration for MLFLOW_TRACKING_URI, verify network access from GitHub Actions to MLflow server (may need VPN or IP allowlist), ensure validation script has correct metric thresholds, check Slack webhook configuration, verify Python script executable Berechtigungs.

Validierung

• Modellieren Registry accessible and backend configured
• Models register erfolgreich from training runs
• Stage transitions work (None → Staging → Production → Archived)
• Validation checks enforce quality thresholds
• Modellieren aliases set and resolved korrekt
• Lineage metadata captured comprehensively
• Rollback functionality restores vorherige Versions
• CI/CD pipeline automates promotions
• Team notifications working for stage changes
• Modellieren URIs resolve korrekt in all stages

Haeufige Stolperfallen

• SQLite limitations: Modellieren Registry requires database backend (PostgreSQL/MySQL) for production - file-based registry causes concurrency issues
• Stage conflicts: Multiple versions in same stage cause confusion - use archive_existing_versions=True to auto-archive
• Missing run linkage: Registering models ohne run_id loses lineage - always register from MLflow runs, not raw files
• Alias confusion: Using stages as deployment targets stattdessen of aliases - stages are for workflow, aliases for deployment references
• Validation skipped: Promoting to Production ohne checks - implement mandatory validation in CI/CD pipeline
• No rollback plan: Production issues ohne rollback capability - maintain previous Production version in Archived stage
• Tag overload: Too many unstructured tags - standardize tag schema and naming conventions
• Manual processes: Human-driven promotions are error-prone and slow - automate with CI/CD and approval workflows
• Lost artifacts: Modellieren registered but artifacts deleted from storage - ensure artifact retention policies align with model lifecycle

Verwandte Skills

• track-ml-experiments - Log models to MLflow vor registering them
• deploy-ml-model-serving - Bereitstellen registered models to serving infrastructure
• run-ab-test-models - A/B test models using registry aliases
• orchestrate-ml-pipeline - Automate model training and registration
• version-ml-data - Version training data for model lineage