talmolabtalmolabOfficialยท3 Agent Skills Included

sleap

Track animal body poses in research videos with deep learning

Tracks the body positions of multiple animals in research videos using deep learning models. Removes the need for manual frame-by-frame labeling with a guided GUI, active learning, and fast GPU training. Helps neuroscience and biology researchers train models, run inference, and proofread results in one workflow.
npx skills add talmolab/sleap --all -g -y
Available:

Tells the AI agent to manage this project with uv, run tests and linting through specific commands, and consult the related sleap-io and sleap-nn repositories when working on data or training tasks.

All Skills in This Repository (3)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install SLEAP?โ–ผ

Run `npx skills add talmolab/sleap --all -g -y` in your terminal to install all skills in this suite globally.

What is SLEAP used for?โ–ผ

SLEAP is a deep learning tool that tracks the body poses of one or many animals in videos, widely used in neuroscience and behavior research.

Do I need a GPU to use SLEAP?โ–ผ

A GPU makes training much faster (15 to 60 minutes for a typical dataset), but SLEAP also supports CPU-only setups and remote training workflows.

Can I use SLEAP without coding experience?โ–ผ

Yes. SLEAP includes a full labeling and training GUI, so researchers can label data, train models, and review predictions without writing code.

How many labels does SLEAP need to train?โ–ผ

SLEAP's human-in-the-loop workflow delivers accurate predictions with very few labels, letting you train, correct errors, and retrain iteratively.

Related Repositories in Education & Research

View All in Education & Researchโ†’