https://arxiv.org/abs/2406.05074

This repository contains the code to run the Hibou-B model locally. For inquiries about accessing Hibou-L, please contact us at [email protected].

The easiest way to use the Hibou-B model is through the HuggingFace repository. Run the following code to get started:

from transformers import AutoImageProcessor, AutoModel

processor = AutoImageProcessor.from_pretrained("histai/hibou-b", trust_remote_code=True)
model = AutoModel.from_pretrained("histai/hibou-b", trust_remote_code=True)

We use a customized implementation of the DINOv2 architecture from the transformers library to add support for registers, which requires the trust_remote_code=True flag.

If you prefer to use the model without the transformers library, follow these steps:

1. Install the requirements and the package:

   git clone https://github.com/HistAI/hibou.git
   cd hibou
   pip install -r requirements.txt && pip install -e .

2. Download the model weights:

   Hibou-B Weights

3. Load the model with the following code:

   from hibou import build_model
   
   model = build_model("weights-path")

For more information, refer to the example.ipynb notebook.

Table: Linear probing benchmarks reporting top-1 accuracy.

Metrics for Virchow and RudolfV are derived from the respective papers, as these models are not open-sourced.

This repository is licensed under the Apache License, Version 2.0. See the LICENSE file for the full license text.

We would like to thank the authors of the DINOv2 repository, upon which this project is built. The original repository can be found here.

Feel free to reach out at [email protected] if you have any questions or need further assistance!

If you use our work, please cite:

@misc{nechaev2024hibou,
    title={Hibou: A Family of Foundational Vision Transformers for Pathology},
    author={Dmitry Nechaev and Alexey Pchelnikov and Ekaterina Ivanova},
    year={2024},
    eprint={2406.05074},
    archivePrefix={arXiv},
    primaryClass={eess.IV}
}