Weights & Biases (W&B) Flashcards

๐Ÿ“Š Weights & Biases (W&B) Flashcards

Experiment Tracking and Model Management for ML Teams

๐Ÿ“Œ What is W&B?

Weights & Biases is a tool for tracking experiments, visualizing metrics, managing models, and collaborating on ML workflows.

โš™๏ธ Setup

Install via pip install wandb and use wandb.init() to begin logging an experiment.

๐Ÿ“ˆ Metrics Logging

Use wandb.log() to track training/validation metrics in real time across epochs and batches.

๐Ÿ” Hyperparameter Tracking

Automatically track and compare hyperparameters used in different runs for optimization and reproducibility.

๐Ÿ“‚ Project & Run Organization

Organize experiments by project and run with a web dashboard that supports filtering, comparisons, and notes.

๐ŸŽฏ W&B Sweeps

Automate hyperparameter optimization using W&B Sweeps with Bayesian search, grid, or random strategies.

๐Ÿง  Model Management

Use W&B Artifacts to version and manage datasets, models, and evaluation results collaboratively.

๐Ÿงช Dataset Versioning

Track and version datasets like Git for data using Artifacts, with full lineage and metadata.

๐Ÿ“บ Visualization Dashboards

Interactive plots, custom charts, and parallel coordinates for hyperparameter sweep analysis.

๐ŸŒ Team Collaboration

Share live reports, dashboards, and analysis with team members and stakeholders across projects.

โšก Auto-Logging

Out-of-the-box support for popular ML libraries like Keras, PyTorch Lightning, Sklearn, Hugging Face, XGBoost, and more.

๐Ÿ› ๏ธ Integration Ecosystem

Seamlessly integrates with Colab, Jupyter, GitHub, S3, GCP, AWS, and other parts of the ML stack.