Spaces:
Running
on
T4
Apply for community grant: Personal project (gpu)
Project Description
The Swedish National Archives is about to release an end-to-end Handwritten Text Recognition (HTR) pipeline, specifically engineered for analyzing Swedish text documents from the 1600s to the 1800s. This HTR pipeline aims to provide a versatile solution for text-region and text-line segmentation, and subsequently, for recognizing the segmented text. We are committed to continually updating these models to ensure their efficacy and adaptability as more data becomes available. However, new features and packaging of the code will be done separately and are planned to be integrated into the Hugging Face Hub.
Financial Situation
As part of the HUMinfra research infrastructure, our broader objective is to democratize HTR technology. We aim to open-source our models, simplify HTR inferences, and provide comprehensive tools for easier evaluation. Our project is primarily led by researchers and data scientists who are passionate about text recognition and its implications. However, we are currently constrained by not having the hardware to host it ourselves. We are also financially constrained as part of a governmental institution. It's for this reason that we are seeking community grant support.
Hi @Gabriel , we have assigned a gpu to this space. Note that GPU Grants are provided temporarily and might be removed after some time if the usage is very low.
To learn more about GPUs in Spaces, please check out https://huggingface.co/docs/hub/spaces-gpus