Machine learning system design

As explained in Delivering Machine Learning Projects: From design to deployment

Machine Learning System Design is a methodology and guide for designing production machine learning systems across problem definition, data, modeling, and serving.

Machine Learning System Design is a production-focused methodology that organizes machine learning work around problem definition, data, modeling, and serving. It contrasts research implementations with operational requirements, including performance and compute constraints. section 1.3

Structure the work into four connected phases:

• Project setup: clarify goals, user experience, performance constraints, evaluation, personalization, people, compute, and infrastructure. section 1.3
• Data pipeline: address privacy, bias, storage, preprocessing, and data availability. section 1.3
• Modeling: handle model selection, training, debugging, hyperparameter tuning, and scaling. section 1.3
• Serving: evaluate the model and examine assumptions involved in running it in the field. section 1.3

Figure 1.4 ML project flow as described in Huyen and Hopper (2020). Figure redrawn and adapted by the author. Arrows represent dependency relations rather than workflow.

Use the project-flow diagram to navigate dependencies among setup, data, modeling, and serving; its arrows represent dependencies rather than workflow. section 1.3

How Delivering Machine Learning Projects: From design to deployment covers it

How deep it goes

Maps four phases and their major tasks, while stopping short of detailed algorithmic or implementation treatment.

Examples

Uses a technical-interview discussion to elicit project goals, constraints, evaluation, personalization, and infrastructure needs.

What you can do after

Apply it to organize the tasks and considerations required to create a production machine learning system.

How it connects

Connects agile, DevOps, MLOps, data mining, privacy, bias, evaluation, and technical debt.

liveBookmanning.com

Read Delivering Machine Learning Projects: From design to deployment on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Machine Learning System DesignThe practice of architecting, building, and evaluating software systems whose behavior depends on machine learning models and their surrounding data and operational components.

Compare how 2 books explain machine learning system design →

Related concepts

• project managementuses · Project management is the discipline of planning, coordinating, organizing, and controlling work and resources to achieve defined project objectives within constraints.
• mlopscontrasts with · MLOps is the discipline of applying engineering and operational practices to develop, deploy, monitor, maintain, and manage machine-learning systems throughout their life cycle.
• machine learninguses · Machine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.

This explainer was generated by Manning's knowledge graph from the text of Delivering Machine Learning Projects: From design to deployment. The section links point to the parts of the book it draws on.