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

weights-and-biases

Track ML experiments, run hyperparameter sweeps, and manage model artifacts with W&B.

1|Updated Mar 5, 2026
Search Tags:#mlops#pytorch#hyperparameter tuning#experiment tracking#wandb#model registry#sweeps
One-click install
npx skills add https://github.com/Clay-HHK/claude-skills --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-skills/tree/main/weights-and-biases
Command: npx skills add https://github.com/Clay-HHK/claude-skills --skill weights-and-biases-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Machine learning teams lose track of experiments, hyperparameters, and model versions when training runs are scattered across notebooks and terminals. This Skill provides structured guidance for logging metrics, comparing runs, optimizing hyperparameters, and versioning models with Weights & Biases. ## Core Features & Use Cases - Experiment Tracking: Automatically log metrics, configs, media, and system stats from PyTorch, TensorFlow, Keras, HuggingFace, and PyTorch Lightning training loops. - Hyperparameter Sweeps: Run grid, random, or Bayesian optimization searches with early termination and parallel agents across multiple GPUs. - Artifacts & Model Registry: Version datasets and models with lineage tracking, aliases, and promotion workflows from staging to production. - Use Case: A researcher fine-tuning BERT can launch a Bayesian sweep over learning rate and batch size, compare 50 runs in a real-time dashboard, and promote the best checkpoint to a shared model registry. ## Quick Start Ask the AI to instrument your PyTorch training script with wandb logging for loss, accuracy, and model checkpoints.

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 PyTorch training with Weights & Biases?▼

Call wandb.init with your project name and config, then use wandb.log inside your training loop to record loss, accuracy, and other metrics. Finish with wandb.finish and optionally save checkpoints via wandb.save or Artifacts.

How to run hyperparameter sweeps with wandb?▼

Define a sweep config with a search method (grid, random, or bayes), a target metric, and parameter distributions, then create it with wandb.sweep. Launch one or more agents with wandb.agent pointing to your training function to execute trials.

Does wandb integrate with HuggingFace Transformers?▼

Yes, set report_to="wandb" in TrainingArguments and the Trainer automatically logs metrics, evaluation results, and checkpoints to W&B. You can also add custom WandbCallback subclasses for additional logging.

What is the difference between wandb.save and Artifacts?▼

wandb.save uploads files associated with a single run, while Artifacts create versioned, deduplicated objects with lineage tracking across runs. Artifacts are recommended for datasets and models that need versioning and registry promotion.

Can I use wandb offline without internet access?▼

Yes, set WANDB_MODE=offline before initializing your run and all data is stored locally. Later, run wandb sync on the run directory to upload everything to the W&B servers.

Which sweep search method should I choose in wandb?▼

Use Bayesian optimization for expensive training runs where sample efficiency matters, random search for quick exploration of many parameters, and grid search only for a few discrete parameters. Bayesian search learns from previous trials to focus on promising regions.

Related Skills

Explore More Data & Analytics Skills→
VYRE-StudiosVYRE-Studios

weights-and-biases

Track ML experiments and model artifacts with automated logging and visualization.

Official
Intermediate
QingHuan-6QingHuan-6

llm-serving-auto-benchmark

Benchmark SGLang, vLLM, and TensorRT-LLM serving to produce Markdown/CSV summaries.

Community
Advanced
Zeyuyang-0420Zeyuyang-0420

pytorch-lightning

Build, train, and deploy PyTorch models with PyTorch Lightning.

Community
Advanced
tranhieutttranhieutt
70

mlops-engineer

Build scalable ML infrastructure for experiment tracking, pipelines, and registry automation.

Community
Advanced