skills.rest
eddielueng
eddieluengCOMMUNITY
Explore all skills by this author →
GitHub

weights-and-biases

Track ML experiments with W&B logging, sweeps, and artifact registry.

1|Updated Apr 14, 2026
Search Tags:#experiment tracking#model registry#real-time metrics#ml experiments#hyperparameter sweeps#wandb artifacts#collaborative mlops
One-click install
npx skills add https://github.com/eddielueng/hermes-agent-zh --skill weights-and-biases-eddielueng
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: weights-and-biases
Source: https://github.com/eddielueng/hermes-agent-zh/tree/main/skills/mlops/evaluation/weights-and-biases
Command: npx skills add https://github.com/eddielueng/hermes-agent-zh --skill weights-and-biases-eddielueng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires wandb, and includes references (resource) components.

What problem does it solve?

Weights & Biases (W&B) eliminates scattered experiment tracking by centralizing metrics, configurations, artifacts, and model registry workflows so teams can compare results and reproduce outcomes.

Core Features & Use Cases

  • Experiment tracking (projects & runs): record configs, metrics, run IDs, and shareable run URLs for fast comparison across experiments.
  • Real-time visualization (metrics & custom plots): log scalars, images, tables, histograms, and charts to understand training dynamics as they happen.
  • Hyperparameter optimization (sweeps): run automated sweeps (grid/random/bayesian) to find better learning rates and training configurations efficiently.
  • Artifacts & model registry: version datasets and models with lineage, aliases (latest/best/production), and staged promotion for collaboration and deployment.
  • Framework integrations: work with popular training stacks such as Hugging Face Transformers, PyTorch Lightning, Keras/TensorFlow, and PyTorch native loops.

Quick Start

Install W&B with pip, log in with your API key using wandb login, then initialize a run via wandb.init(project="your-project") and log metrics with wandb.log during your training loop.

Frequently Asked Questions about weights-and-biases

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track ML experiments and compare training runs automatically?▼

You can track ML experiments by initializing a W&B run with wandb.init, logging metrics with wandb.log, and recording configs to automatically compare training runs and reproduce outcomes.

Can I use wandb with Hugging Face Transformers or PyTorch Lightning for model training?▼

Yes, wandb integrates with Hugging Face Transformers, PyTorch Lightning, and Keras/TensorFlow to provide automated reporting and real-time visualization during model training.

What is the best way to run hyperparameter sweeps for deep learning models?▼

Run automated hyperparameter sweeps using grid, random, or Bayesian search to efficiently find better learning rates and training configurations for your deep learning models.

How do I version datasets and manage a model registry for deployment?▼

Version datasets and models using wandb artifacts to track lineage, assign aliases like latest or production, and enable staged model promotion for deployment and collaboration.

Do I need to install the wandb package to log real-time metrics and custom plots?▼

Yes, you must pip install wandb and log in with your API key to log scalars, images, tables, and custom plots for real-time metric visualization during training.

Related Skills

Explore More Data & Analytics Skills→
bigbiobigbio
11

sdrf:setup

Install and configure SDRF annotation dependencies via Conda, Pip, or uv.

Official
Intermediate
ovachieverovachiever
52

weights-and-biases

Automate ML experiment tracking and model registry management via the wandb API.

Community
Advanced
Mekann2904Mekann2904

research-ml-deep

Integrate PyTorch Lightning, Transformers, and PyTorch Geometric for deep learning tasks.

Community
Advanced
DifanaDAPDifanaDAP

weights-and-biases

Automate ML experiment tracking and hyperparameter sweeps via the Python API.

Community
Advanced