Can I use this for real-time feature engineering?

Yes, the skill includes patterns for both batch training pipelines and real-time feature processing combined with stream processing.

How does this skill handle data versioning?

It implements best practices for data lineage and versioning using tools like DVC to ensure reproducibility across the entire pipeline.

What tools does this MLOps skill support?

It provides guidance for major orchestration tools like Apache Airflow, Dagster, and Kubeflow, as well as tracking tools like MLflow and Weights & Biases.

Does it help with model monitoring?

Absolutely. It includes strategies for monitoring production traffic, detecting model drift, and setting up automated rollback triggers.

ML Pipeline & MLOps Workflow

Name: ML Pipeline & MLOps Workflow
Author: drgaciw

bydrgaciw

データサイエンスとML

Orchestrates end-to-end MLOps pipelines from data ingestion and preparation to model training, validation, and production deployment.

概要

This skill provides a comprehensive framework for building and automating production-grade machine learning pipelines. It guides users through the entire ML lifecycle, including designing Directed Acyclic Graphs (DAGs), implementing data validation with tools like Great Expectations, managing experiment tracking via MLflow, and executing advanced deployment strategies like canary or blue-green releases. It is an essential resource for data scientists and ML engineers looking to move from experimental notebooks to reproducible, scalable, and monitored production systems.

主な機能

0 GitHub stars
Advanced deployment strategies including canary and blue-green rollouts
End-to-end DAG orchestration for tools like Airflow, Dagster, and Kubeflow
Automated data validation and feature engineering pipeline patterns
Comprehensive model monitoring and performance regression detection
Integrated experiment tracking and model registry management

ユースケース

Setting up continuous training systems triggered by data or model drift
Implementing reproducible training workflows with integrated data versioning
Automating the transition from data exploration to production-ready ML pipelines

概要

主な機能

0 GitHub stars
Advanced deployment strategies including canary and blue-green rollouts
End-to-end DAG orchestration for tools like Airflow, Dagster, and Kubeflow
Automated data validation and feature engineering pipeline patterns
Comprehensive model monitoring and performance regression detection
Integrated experiment tracking and model registry management

ユースケース

Setting up continuous training systems triggered by data or model drift
Implementing reproducible training workflows with integrated data versioning
Automating the transition from data exploration to production-ready ML pipelines