Model registry

Use W&B Models as a central system of record for your best models, standardized and organized in a model registry across projects and teams.

With W&B Models, you can:

• Bookmark your best model versions for each machine learning task.
• Move model versions through its ML lifecycle; from staging to production.
• Track and audit the history of changes to production models.

How it works

Track and manage your trained models with a few simple steps.

1. Log a model version: In your training script, add a few lines of code to save the model files as an artifact to W&B.
2. Compare performance: Check live charts to compare the metrics and sample predictions from model training and validation. Identify which model version performed the best.
3. Link to registry: Bookmark the best model version by linking it to a registered model, either programmatically in Python or interactively in the W&B UI.

The following code snippet demonstrates how to log and link a model to the Model Registry:

import wandb
import random

# Start a new W&B run
with wandb.init(project="models_quickstart") as run:
    # Simulate logging model metrics
    run.log({"acc": random.random()})

    # Create a simulated model file
    with open("my_model.h5", "w") as f:
        f.write("Model: " + str(random.random()))

    # Save the dummy model to W&B
    best_model = wandb.Artifact(f"model_{run.id}", type="model")
    best_model.add_file("my_model.h5")
    run.log_artifact(best_model)

    # Link the model to the Model Registry
    run.link_artifact(best_model, "model-registry/My Registered Model")

    run.finish()

4. Test and deploy your model: Transition model versions through customizable workflows stages, such as staging and production.

How to get started

Depending on your use case, explore the following resources to get started with W&B Models:

• Check out the Model Registry Quickstart YouTube video.
• Read the models walkthrough for a step-by-step outline of the W&B Python SDK commands you could use to create, track, and use a dataset artifact.
• Review this report on how the Model Registry fits into your ML workflow and the benefits of using one for model management.
• Explore this chapter to learn how about:
  • Role based access controls (RBAC).
  • How to automate model testing and deployment.
  • Use tags to organize registered models.
  • Set up Slack notifications when a new model version is linked to a model registry.