Official implementation of the HPB-ViT architecture

This repository is still under development. If you have any questions or issues feel free to submit an issue or mail me at luca.zedda@unica.it

A usage example is provided in the test_HPB_ViT_model.py file. Currently, we are supporting automatic backbone integration using the Huggingface transformer library

@InProceedings{10.1007/978-3-031-43153-1_19,
author="Zedda, Luca
and Loddo, Andrea
and Di Ruberto, Cecilia",
editor="Foresti, Gian Luca
and Fusiello, Andrea
and Hancock, Edwin",
title="Hierarchical Pretrained Backbone Vision Transformer for Image Classification in Histopathology",
booktitle="Image Analysis and Processing -- ICIAP 2023",
year="2023",
publisher="Springer Nature Switzerland",
address="Cham",
pages="223--234",
isbn="978-3-031-43153-1"
}

This project is released under the [MIT]

Most of the code for the Transformer architecture basic modules is taken from the https://github.com/lucidrains/vit-pytorch repository, shout out to Phil Wang and his amazing work!