Machine Learning Lifecycle Made Easy with MLflow
summary
Describes using MLflow’s Tracking, Models, and Projects to record and reproduce experiments: concrete demo on the sklearn diabetes dataset trains an SGD regressor, logs hyperparameters (alpha, l1_ratio, penalty), metrics (RMSE and R²), and EDA plots as artifacts into an mlruns directory, launches mlflow ui localhost:5000 to inspect runs, logs a model named sgd_model, and serves it with mlflow serve on a port to return JSON predictions — the end-to-end example runs and returns predictions.
AI-written summary · about
chapters (8)
- 0:00Introduction and speakers
- 1:08ML lifecycle challenges overview
- 8:09Introducing MLflow solution
- 13:24MLflow tracking deep dive
- 16:18MLflow models for deployment
- 18:45MLflow projects for reproducibility
- 20:55Live demo of MLflow tracking
- 35:03Conclusions and resources
related talks
31mMLflow: Platform for Complete Machine Learning Lifecycle
57mManaging Machine Learning Lifecycle with MLflow
33mOperationalizing Model Lifecycle with MLFlow
30mMLflow: an Open Source Tool for Managing the ML Project Lifecycle
01h 04mGo with the flow–Track your machine learning lifecycle
17mIntroducing mlflow
01h 25mIntroduction to MLOps with MLflow
37mPlatform for Complete Machine Learning Lifecycle