huggingface-vision-trainer

Prepare, train, evaluate, and publish computer-vision and vision-language models with Hugging Face datasets, Transformers, PEFT, and Jobs.

Skill by metaspartan

IndexedMIT★ 32

Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/metaspartan-huggingface-vision-trainer

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Prepare, train, evaluate, and publish computer-vision and vision-language models with Hugging Face datasets, Transformers, PEFT, and Jobs.

Use cases: ml

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools: No tools found
Secrets: None asked for
Install scripts: None found

Network hosts it talks to 1

codeload.github.com · found in install command

Files it touches

~/.claude/skills/huggingface-vision-trainer

What's inside

Skill name: huggingface-vision-trainer
Description: Prepare, train, evaluate, and publish computer-vision and vision-language models with Hugging Face datasets, Transformers, PEFT, and Jobs.
License: MIT
Instructions: 20 lines in SKILL.md · view on GitHub

Files (1)

SKILL.md

SKILL.md (first lines)

# Hugging Face Vision Trainer

Use this workflow for image classification, object detection, segmentation, depth estimation, and vision-language fine-tuning.

## Workflow

1. Inspect the dataset card, license, image features, labels, splits, corrupt records, class balance, dimensions, and annotation coordinate format.
2. Visualize a small representative sample before transforming labels or augmenting images.
3. Choose a model and processor that explicitly support the task and license requirements.
4. Preserve label mappings, preprocessing settings, image normalization, and augmentation seeds in the training configuration.
5. Split by source or subject when random row splitting could leak near-duplicate images.
6. Select task-appropriate metrics such as accuracy/F1, mAP, IoU, or VQA-style scores.
7. Run a small batch through preprocessing, forward pass, loss, and metric computation before launching remote training.
8. Save the processor with the model and publish representative evaluation outputs in the model card.

For vision-language models, ensure image tokens are not truncated and verify the model's chat template. Use PEFT when it materially reduces memory without invalidating the task.

Remote GPU work follows the `huggingface-jobs` confirmation and monitoring workflow. Do not upload private images, biometric data, or sensitive annotations without explicit authorization.

From metaspartan/cybara, MIT.

Signals

Stars: 32
Forks: 9
Open issues: 0
Last commit (this folder): Jul 30, 2026 (3 months ago)
Last release: Oct 24, 2026 (5 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/metaspartan/cybara
Website: cybara.ai
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: GitHub github-graphql

More from metaspartan

Cite this page

huggingface-vision-trainer skill. Agent Mart. https://agentmart.dev/skills/metaspartan-huggingface-vision-trainer (last updated 2026-10-24)