MLflow
MLflow is an open-source platform for managing the machine-learning lifecycle, including experiment tracking, model and artifact management, and model deployment.
Machine Learning Platform Engineering
MLflow is an open-source platform for managing the machine learning lifecycle, including experiment tracking, model packaging, registry, and deployment.
- How it is used
- Tracks experiments and manages model versions from exploratory training through selection and inference.
- How deep it goes
- Works through local setup, run logging, metric-based registration, pipeline validation, and serving commands; contrasts two deployment paths.
- Examples
- Uses income-classifier runs, movie-recommender training and validation, registry queries, and inference-service commands.
- What you can do after
- Operate/configure MLflow to track runs, select and register a model, and serve a registered version alone or with BentoML.
Read how Machine Learning Platform Engineering explains MLflow →
Generative AI in Action
MLflow is an open-source platform for managing machine-learning experimentation, reproducibility, model packaging, deployment, and lifecycle operations.
- How it is used
- Supports experiment management and production observability for machine-learning and LLM applications.
- How deep it goes
- Explains core components and demonstrates tracking configuration and metrics logging, without detailing deployment internals.
- Examples
- Uses Docker Compose, an MLflow dashboard, and an OpenAI chat-completion example that records prompts, responses, latency, and token counts.
- What you can do after
- Operate and configure MLflow tracking for LLM requests, experiments, artifacts, and operational metrics.
Designing Deep Learning Systems: A guide for software engineers
MLflow is an open-source platform for managing the machine learning lifecycle, including experiment tracking, artifact management, model registration, and deployment.
- How it is used
- Supports experiment comparison, troubleshooting, reproducibility, deployment, artifact management, and centralized model lifecycle operations.
- How deep it goes
- Explains component responsibilities, remote-server architecture, storage boundaries, artifact-upload choices, and client operations, without implementation internals.
- Examples
- Demonstrates run logging, SQL metadata storage, object-storage artifacts, tracking-server proxying, and model-version retrieval.
- What you can do after
- Operate MLflow to record runs, retrieve metadata, manage artifacts, compare experiments, and govern model versions.
Read how Designing Deep Learning Systems: A guide for software engineers explains MLflow →
Related concepts
- experiment trackingused by · Experiment tracking is the systematic recording and organization of experiment configurations, metrics, artifacts, and results to support inspection, comparison, or reproducibility.
- Weights and Biasesalternative to · Weights & Biases is a machine-learning development platform for tracking experiments and managing runs, metrics, artifacts, models, and related workflows.
- model registryhas part · A model registry is a system for storing, organizing, versioning, and managing machine-learning models and their associated metadata throughout their lifecycle.
- platformkind of · A platform is a foundational computing system, environment, or online service that provides capabilities or infrastructure for building, running, delivering, or interacting with applications and services.
- MLflow clientused by · The MLflow client is a programmatic interface for tracking machine-learning runs, managing metadata and artifacts, and accessing registered models.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.