File size: 2,478 Bytes
cd38f82 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 |
---
license: mit
language:
- en
pipeline_tag: image-to-text
tags:
- gregg-shorthand
- handwriting-recognition
- ocr
- historical-documents
- stenography
library_name: pytorch
datasets:
- a0a7/Gregg-1916
metrics:
- accuracy
---
# Gregg Shorthand Recognition Model
This model recognizes Gregg shorthand notation from images and converts it to readable text.
## Model Description
- **Model Type**: Image-to-Text recognition
- **Architecture**: CNN-LSTM with advanced pattern recognition
- **Training Data**: Gregg shorthand samples
- **Language**: English
- **License**: MIT
## Intended Use
This model is designed to:
- Recognize Gregg shorthand from scanned documents
- Convert historical stenographic notes to digital text
- Assist in digitizing shorthand archives
- Support stenography education and research
## How to Use
### Using the Hugging Face Transformers library
```python
from transformers import pipeline
from PIL import Image
# Load the pipeline
pipe = pipeline("image-to-text", model="a0a7/gregg-recognition")
# Load an image
image = Image.open("path/to/shorthand/image.png")
# Generate text
result = pipe(image)
print(result[0]['generated_text'])
```
### Using the original package
```python
from gregg_recognition import GreggRecognition
# Initialize the recognizer
recognizer = GreggRecognition(model_type="image_to_text")
# Recognize text from image
result = recognizer.recognize("path/to/image.png")
print(result)
```
### Command Line Interface
```bash
# Install the package
pip install gregg-recognition
# Use the CLI
gregg-recognize path/to/image.png --verbose
```
## Model Performance
The model uses advanced pattern recognition techniques optimized for Gregg shorthand notation.
## Training Details
- **Framework**: PyTorch
- **Optimizer**: Adam
- **Architecture**: Custom CNN-LSTM with pattern database
- **Input Resolution**: 256x256 pixels
- **Preprocessing**: Grayscale conversion, normalization
## Limitations
- Optimized specifically for Gregg shorthand notation
- Performance may vary with image quality
- Best results with clear, high-contrast images
## Citation
If you use this model in your research, please cite:
```bibtex
@misc{gregg-recognition,
title={Gregg Shorthand Recognition Model},
author={Your Name},
year={2025},
url={https://huggingface.co/a0a7/gregg-recognition}
}
```
## Contact
For questions or issues, please open an issue on the [GitHub repository](https://github.com/a0a7/GreggRecognition). |