Beginner · Fundamentals

Model

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A mathematical representation learned from data that maps inputs to outputs.

Technical Definition

A mathematical representation learned from data that maps inputs to outputs.

How it works

A model is the trained artifact produced by a machine learning algorithm. It encodes patterns discovered in training data as parameters (weights) and a structure (architecture) that together transform inputs into predictions. Models range from simple linear regressions to billion-parameter neural networks. Once trained, a model can be saved, shared, deployed behind an API, or fine-tuned for new tasks.

Related Concepts

• Inference — Using a trained model to make predictions on new data — the deployment phase of machine learning.
• Algorithm — A finite, well-defined sequence of steps used to learn from data or solve a computational problem.
• Training — The process of adjusting a model's parameters so it learns patterns from labeled or unlabeled data.