Weights and Biases
- Category
- AI and LLMs
- Docs outline
- 12 sections
Official docs outline
docs.wandb.ai- W&B overview
What Weights and Biases provides for experiment tracking, artifacts, sweeps, and reports.
- Quickstart
Installing, logging in with an API key, and creating a first run.
- Runs and experiment tracking
Initializing runs, config, summary metrics, and organizing runs into projects.
- Logging metrics and media
Logging scalars, media, tables, and system metrics with wandb.log.
- Configuration and hyperparameters
Recording hyperparameters and run settings with wandb.config.
- Artifacts
Versioning datasets and models and tracking lineage between runs.
- Sweeps
Hyperparameter search via sweep configuration, controllers, and agents.
- Reports and dashboards
Interactive reports, panels, and workspaces for comparing runs.
- Registry
Promoting artifacts to a central registry for model lifecycle governance.
- Tables
Logging and querying tabular and media data for analysis.
- Framework integrations
Integrations with PyTorch, Keras, Hugging Face, and others.
- Weave for LLM applications
Tracing, evaluating, and monitoring LLM applications.
Each section links to the official page. Titles and summaries by Extra Credit; the docs belong to their authors.
