Home / ultralytics / skills · skills/yolo-tuning/SKILL.md · GitHub

yolo-tuning skillA

yolo-tuning is agent-read markdown (skill) from ultralytics/skills: Use when improving or comparing Ultralytics YOLO models in Platform or code, or running hyperparameter search/autotraining — Platform experiment comparison, the systematic improvement playbook, model.tune() genetic evolution, Ray Tune, search spaces, and deciding whether tuning is worthwhile. For one training run and its arguments, see yolo-training..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# Improving models & hyperparameter tuning

## The improvement playbook (follow in order — tuning is the LAST step)

Hyperparameter tuning is expensive and usually not the bottleneck. Escalate in this
order, re-validating after each step:

1. **Fix the data** — check `confusion_matrix.png` and `train_batch*.jpg` for label
   noise; review the top false-negative/false-positive val images; add examples of
   failing classes and true-background images. Data quality beats every other lever.
2. **Train longer** — if val mAP was still rising at the end: more `epochs`, higher
   `patience`.
3. **Bigger input** — small objects or mAP50 ≫ mAP50-95: raise `imgsz` (640 → 960/1280).
4. **Bigger model** — underfitting (train and val both mediocre): n → s → m → l.
5. **Domain-matched augmentation** — aerial `degrees=180 flipud=0.5`, crowded scenes
   `copy_paste=0.3`/`mixup=0.1`, color-critical classes lower `hsv_h`
   (see yolo-training's `training-args.md`).
6. **Only now: hyperparameter tuning** — worth ~0.5–2 mAP when everything above is
   exhausted.

Decision signals: overfitting (val drops while train improves) → more data/aug or
…

Read the whole file at its exact version.

How to install

Latest version
mdr add ultralytics/skills/yolo-tuning@git:20260820.dcee0db
Exact content
mdr add ultralytics/skills/yolo-tuning@sha256:c29ed67cdfb35669

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_s6hvrnpiujd5qyfj.svg)](https://markdownregistry.com/a/art_s6hvrnpiujd5qyfj)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260820.dcee0db latest2026-08-20 dcee0db 5,209 BA view · diff
git:20260811.e35dbcc2026-08-11 e35dbcc 5,210 BA view · diff
git:20260811.47343d92026-08-11 47343d9 4,560 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (5209 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

ultralytics/skills · 25 stars · license AGPL-3.0 · pushed 2026-09-24 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_s6hvrnpiujd5qyfj
GET https://markdownregistry.com/api/v1/resolve?ref=ultralytics/skills/yolo-tuning
GET https://markdownregistry.com/api/v1/blob/c29ed67cdfb356692c053a4310e82ae115c9bfd7c5325f27964fd94c62cc8fef

Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

More from ultralytics/skills

AGENTS.md agents
ultralytics/skills · AGENTS.md
git:20260911.983f6a2 · audit A · 25 stars
CLAUDE.md claude
ultralytics/skills · CLAUDE.md
git:20260811.e5e8483 · audit B · 25 stars
platform-cli skill
ultralytics/skills · skills/platform-cli/SKILL.md · Use when operating Ultralytics Platform from a terminal or script with the ul cloud CLI (ul cloud RESOURCE OPERATION…
git:20260924.c2d2489 · audit A · 25 stars
yolo-datasets skill
ultralytics/skills · skills/yolo-datasets/SKILL.md · Use when uploading, annotating, building, converting, analyzing, or debugging datasets in Ultralytics Platform or local…
git:20260904.9c3b8ec · audit A · 25 stars
yolo-export skill
ultralytics/skills · skills/yolo-export/SKILL.md · Use when exporting or deploying Ultralytics YOLO models in Platform or code — the Platform Export tab and yolo…
git:20260904.9c3b8ec · audit A · 25 stars
yolo-inference skill
ultralytics/skills · skills/yolo-inference/SKILL.md · Use when testing, running, or deploying Ultralytics YOLO inference in Platform or code on images, video, webcams, or…
git:20260904.9c3b8ec · audit A · 25 stars
yolo-models skill
ultralytics/skills · skills/yolo-models/SKILL.md · Use when choosing or comparing Ultralytics models in Platform or code — picking a model family (YOLO26/YOLO11/YOLOv8…
git:20260904.9c3b8ec · audit A · 25 stars
yolo-training skill
ultralytics/skills · skills/yolo-training/SKILL.md · Use when training, fine-tuning, or validating Ultralytics YOLO models in Platform, cloud GPUs, or local code —…
git:20260904.9c3b8ec · audit A · 25 stars
yolo skill
ultralytics/skills · skills/yolo/SKILL.md · Use for ANY task involving Ultralytics Platform, the ultralytics Python package, yolo CLI, YOLO model weights (.pt)…
git:20260922.61a0f16 · audit A · 25 stars

Every file in ultralytics/skills

Browse by kind, by grade A, or by owner.