wandb

Overview

Weights & Biases (wandb) tracks ML experiments with rich visualizations, hyperparameter sweeps, dataset versioning, model registry, and collaborative dashboards. Industry standard for experiment tracking across ML teams.

Installation

uv pip install wandb
wandb login  # authenticate with API key

Experiment Tracking

import wandb

wandb.init(project="my_project", config={
    "learning_rate": 0.001,
    "batch_size": 32,
    "architecture": "transformer",
})
for epoch in range(10):
    loss = train_one_epoch()
    wandb.log({"train_loss": loss, "val_loss": val_loss, "epoch": epoch})
wandb.finish()

Hyperparameter Sweep

sweep_config = {
    "method": "bayes",
    "metric": {"name": "val_loss", "goal": "minimize"},
    "parameters": {"lr": {"min": 1e-5, "max": 1e-2}},
}
sweep_id = wandb.sweep(sweep_config, project="my_project")
wandb.agent(sweep_id, function=train_function, count=20)

References

• W&B docs
• W&B GitHub