Instructions to use microsoft/trocr-base-handwritten with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use microsoft/trocr-base-handwritten with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="microsoft/trocr-base-handwritten")# Load model directly from transformers import AutoTokenizer, AutoModelForImageTextToText tokenizer = AutoTokenizer.from_pretrained("microsoft/trocr-base-handwritten") model = AutoModelForImageTextToText.from_pretrained("microsoft/trocr-base-handwritten") - Notebooks
- Google Colab
- Kaggle
how to fine-tune it on multiple language handwritten text
#9
by GardensOfBabylon29 - opened
i am trying to it run on multiple languages , but couldn't get better results , any idea , how to implement it ?
any luck ? i think the best approch would be to just train the tokenizer
how?? i am trying to train it on RTL langages like persian , so pre-train then train .
Let me know if there is any solution to be found I am also working on RTL Languages currently like Urdu and Arabic @Cherryblade29
@Cherryblade29 can we get on this together like help each other out. If you dont mind.
I am trying for english language but getting the bad result. any idea or suggestion ?