Trocr Finetuned 20250422 125947

🚀 Fine-tuned TrOCR Model for Portuguese

This repository houses a fine - tuned TrOCR model specially trained for Optical Character Recognition (OCR) on Portuguese text. It builds upon the microsoft/trocr-base-printed model and has been further trained on a dataset of Portuguese text images.

✨ Features

• Specifically designed for OCR on Portuguese text.
• Combines a vision encoder and a text decoder for effective image - to - text conversion.

📦 Installation

1. Install Dependencies

bash pip install transformers datasets Pillow requests

💻 Usage Examples

Basic Usage

python from transformers import VisionEncoderDecoderModel, TrOCRProcessor from PIL import Image

model = VisionEncoderDecoderModel.from_pretrained("mazafard/trocr-finetuned_20250422_125947") 
processor = TrOCRProcessor.from_pretrained("mazafard/trocr-finetuned_20250422_125947") 

image = Image.open("path/to/your/image.png").convert("RGB")

pixel_values = processor(image, return_tensors="pt").pixel_values

# Generate prediction
generated_ids = model.generate(pixel_values) 
generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

print(generated_text)

📚 Documentation

Model Description

The model is a VisionEncoderDecoderModel from the Hugging Face Transformers library. It combines a vision encoder (to process images) and a text decoder (to generate text) for OCR tasks.

Intended Use

This model is intended for extracting text from images containing Portuguese text. It can be used for various applications, such as:

• Digitizing Portuguese books and documents.
• Automating data entry from Portuguese forms and invoices.
• Extracting information from Portuguese screenshots or scanned images.

Limitations

• The model may not perform well on handwritten text or text with unusual fonts or styles.
• It might make mistakes on complex layouts or images with low quality.

Training Details

• Dataset: mazafard/portugues_ocr_dataset_full
• Training Parameters:

training_args = TrainingArguments(
    output_dir="./trocr-finetuned",
    per_device_train_batch_size=56,
    num_train_epochs=3,
    save_steps=500,
    logging_steps=50,
    learning_rate=5e-5,
    gradient_accumulation_steps=2,
    fp16=True,
    save_total_limit=2,
    remove_unused_columns=False,
    dataloader_num_workers=2,
)

Evaluation

No evaluation details provided in the original document.

Acknowledgements

This model is based on the TrOCR model by Microsoft.

📄 License

This project is licensed under the MIT License.