skills.rest
Clay-HHK
Clay-HHKCOMMUNITY
Explore all skills by this author →
GitHub

weights-and-biases

Track and visualize ML experiments and artifacts with wandb.

2|Updated Apr 12, 2026
Search Tags:#mlops#artifacts#wandb#experiment-tracking#model-registry#hyperparameter-sweeps
One-click install
npx skills add https://github.com/Clay-HHK/claude-config --skill weights-and-biases-clay-hhk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: weights-and-biases
Source: https://github.com/Clay-HHK/claude-config/tree/main/skills/AI-research-SKILLs/13-mlops/weights-and-biases
Command: npx skills add https://github.com/Clay-HHK/claude-config --skill weights-and-biases-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Provide a unified, collaborative platform to automatically track machine learning experiments, capture configuration and metrics, version datasets and models, and visualize training in real time so teams can reproduce results and ship models faster.

Core Features & Use Cases

  • Experiment Tracking: Automatic logging of metrics, hyperparameters, system metrics, and training artifacts for each run.
  • Real-time Visualization & Comparison: Live dashboards to compare runs, view custom charts, histograms, and media.
  • Hyperparameter Sweeps: Integrated sweep configurations and agents for random, grid, or Bayesian optimization.
  • Artifacts & Model Registry: Versioned dataset and model artifacts with lineage, aliases, and promotion workflows for staging/production.
  • Framework Integrations: Examples and integrations for PyTorch, TensorFlow/Keras, PyTorch Lightning, HuggingFace, XGBoost, LightGBM, and fast.ai.
  • Use Case: Run large-scale sweeps to tune an image classification pipeline, track dataset versions, and promote the best model to a registry for deployment.

Quick Start

Initialize a W&B run for your project, set key hyperparameters in the run config, log metrics and artifacts during training with wandb.log, and finish the run to upload results.

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 machine learning experiments and visualize training runs in real time?▼

To track machine learning experiments and visualize training runs, initialize a run, set hyperparameters in the config, and use logging to capture metrics. This provides live dashboards to compare runs, view charts, and monitor system metrics automatically.

Does experiment tracking work with PyTorch, TensorFlow, and HuggingFace frameworks?▼

Yes, experiment tracking works with PyTorch, TensorFlow, PyTorch Lightning, HuggingFace, XGBoost, LightGBM, and fast.ai. It provides framework integrations to automatically log metrics, hyperparameters, and artifacts during training.

What is hyperparameter sweep orchestration and when do I need it for model tuning?▼

Hyperparameter sweep orchestration automates model tuning by running random, grid, or Bayesian searches across training runs. You need it to systematically explore hyperparameter configurations and identify the optimal model parameters.

Can I version datasets and manage a model registry for staging and production?▼

Yes, you can version datasets and manage a model registry using artifacts. This provides versioned dataset and model artifacts with lineage tracking, aliases, and promotion workflows for staging and production deployment.

Do I need an API key to log metrics and use the model registry?▼

Yes, you need the wandb Python package and an API key to log metrics, use the model registry, and upload results. This authentication is required to sync run data and artifacts to your collaborative platform.

Related Skills

Explore More Data & Analytics Skills→
KarlinskySKarlinskyS

weights-and-biases

Track and visualize ML experiments with automatic logging via Weights & Biases.

Community
Advanced
pluginagentmarketplacepluginagentmarketplace
1

ml-deployment

Deploy ML models as REST APIs with Docker containerization and monitoring.

Community
Advanced
ovachieverovachiever
52

ray-train

Orchestrate distributed training across clusters with Ray Train and Ray Tune.

Community
Advanced
DarkArty07DarkArty07
3

weights-and-biases

Track ML experiment metrics, hyperparameters, and model versions with W&B.

Community
Intermediate