hadrakey's picture
Training in progress, step 1000
e06b649 verified
|
raw
history blame
508 Bytes

Alphapen

This project aims to develop an OCR model for instantaneous text extraction from handwritten documents. The ultimate goal is to seamlessly integrate such a model into computers or mobile phones, allowing for the direct digitalization of handwritten documents using a proprietary pen manufactured by a startup company named Alphapen.

Fine-tuning the TrOCR model

python model.py --log_with wandb --push_to_hub True --hub_model_id hadrakey/alphapen_trocr