# Ml Hyperparameter Tuning

> Systematically finding the optimal model configuration to maximize performance.

- Skill: `jcorpac/ml-hyperparameter-tuning` (Agent Skill)
- Install (CLI): `npx skillmds@latest add jcorpac/ml-hyperparameter-tuning`
- Raw SKILL.md: https://api.skillmd.com/api/skills/jcorpac/ml-hyperparameter-tuning/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: jcorpac (https://skillmd.com/u/jcorpac)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/jcorpac/ml-hyperparameter-tuning

---


# ML Hyperparameter Tuning

Tuning is the difference between an "okay" model and a "great" one.

## The Search Space
Identify which parameters have the most impact (e.g., `n_estimators`, `learning_rate`, `max_depth`).

## Methods
- **Grid Search**: Exhaustive search over a fixed grid. Reliable but slow.
- **Random Search**: Random sampling of the space. Often better than grid search for fewer iterations.
- **Bayesian Optimization (Optuna)**: Uses previous trials to inform the next search. The state-of-the-art for ML tuning.

## Best Practices
- **Pruning**: Stop poor-performing trials to save time.
- **Nested Cross-Validation**: To avoid overfitting the hyperparameters themselves.