Experiments overview

​

How it works

1. Create a W&B Run.
2. Store a dictionary of hyperparameters, such as learning rate or model type, into your configuration (wandb.Run.config).
3. Log metrics (wandb.Run.log()) over time in a training loop, such as accuracy and loss.
4. Save outputs of a run, like the model weights or a table of predictions.

# Start a run.
#
# When this block exits, it waits for logged data to finish uploading.
# If an exception is raised, the run is marked failed.
with wandb.init(entity="", project="my-project-name") as run:
  # Save mode inputs and hyperparameters.
  run.config.learning_rate = 0.01

  # Run your experiment code.
  for epoch in range(num_epochs):
    # Do some training...

    # Log metrics over time to visualize model performance.
    run.log({"loss": loss})

  # Upload model outputs as artifacts.
  run.log_artifact(model)

​

Get started

• Read the W&B Quickstart for a step-by-step outline of the W&B Python SDK commands you could use to create, track, and use a dataset artifact.
• Explore this chapter to learn how to:
  • Create an experiment
  • Configure experiments
  • Log data from experiments
  • View results from experiments
• Explore the W&B Python Library within the W&B API Reference Guide.

​

Best practices and tips