Home › Glossary › Systems › Checkpoint

Intermediate · Systems

Checkpoint

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

TL;DR. A saved state of a machine learning model's parameters, allowing training to be paused and resumed or enabling model deployment.

Technical Definition

A saved state of a machine learning model's parameters, allowing training to be paused and resumed or enabling model deployment.

How it works

A checkpoint is a snapshot of a model's state, capturing the values of its parameters at a specific point in time during or after training. This allows users to save progress, enabling them to resume training later from where they left off, or to load a trained model for inference without restarting the training process. Checkpoints are crucial for long training runs and for model replication.

Related Concepts

  • Inference — Using a trained model to make predictions on new data — the deployment phase of machine learning.
  • Model Deployment — The process of making a trained machine learning model available for use in a production environment.

Further Reading

  • Google ML Glossary