|
--- |
|
license: apache-2.0 |
|
pipeline_tag: image-feature-extraction |
|
tags: |
|
- medical |
|
- pathology |
|
--- |
|
|
|
Hibou-L - a Foundational Vision Transformer for digital pathology pretrained on a 1.2B image private dataset using DINOv2 framework. |
|
|
|
For more information and examples of usage visit [https://github.com/HistAI/hibou](https://github.com/HistAI/hibou) and read the [paper](https://arxiv.org/abs/2406.05074). |
|
|
|
Hibou-B is also available at [https://huggingface.co/histai/hibou-b](https://huggingface.co/histai/hibou-b) |
|
|
|
Basic usage: |
|
```python |
|
from transformers import AutoImageProcessor, AutoModel |
|
|
|
processor = AutoImageProcessor.from_pretrained("histai/hibou-L", trust_remote_code=True) |
|
model = AutoModel.from_pretrained("histai/hibou-L", 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. |
