---
title: README
emoji: 👀
colorFrom: blue
colorTo: indigo
sdk: static
pinned: false
---

# ViTAMIn-O: Democratizing computer vision-based machine learning for stem cell research 

Welcome to the official model hub for **ViTAMIn-O**. This organization serves as an open-science database for customized, vision-based deep learning models specifically trained for organoid cross-sectional classification, longitudinal prediction, downstream model evaluation, and concept discovery.

ColabViTAMIn-O [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1nDXBvSziGndkVcpUH9jxRbZpwX-2MMDU?usp=sharing)

## About ColabViTAMIn-O
ViTAMIn-O is a code-free, Colab-based infrastructure designed to encourage wet-lab scientists to use deep learning in their workflows. Without prior coding experience, researchers can seamlessly upload imaging datasets, fine-tune state-of-the-art vision foundation models, and discover morphologically-grounded concepts using the OrgACE pipeline.

## The ViTAMIn-O Model Hub
This hub hosts models trained by the community. You can explore the models via our **Collections**, which categorize models by organoid type (e.g., Ductal, Intestinal, Cerebral).

### How to Contribute
We encourage motivated researchers to share their specialized models.
A quick guide:

1. Open the [ViTAMIn-O Colab Dashboard](https://colab.research.google.com/drive/1nDXBvSziGndkVcpUH9jxRbZpwX-2MMDU?usp=sharing).
2. Train your custom model using your dataset.
3. Check the "Upload to Hugging Face" box under 2. Task Settings.
4. Input your personal HF token and the target Repo name.
5. Once your model is on your personal profile, request to add it to one of our collections, we will catch up within 48 hours!

## Important
If you use the ViTAMIn-O framework or any shared models hosted in this hub, please cite the corresponding paper where available.