A TrOCR model for Persian. This is just a preview to verify the model works in Hezar. This model will be trained on more and better datasets soon!
from hezar.models import Model model = Model.load("trocr-base-fa-v1") text = model.predict(["image.jpg"]) print(text)
