|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- trocr |
|
|
- image-to-text |
|
|
--- |
|
|
# Typst Equation OCR Model |
|
|
A pretrained TrOCR model for Typst equation OCR tasks. |
|
|
|
|
|
## Usage |
|
|
### Using `transformers` |
|
|
```python |
|
|
from PIL import Image |
|
|
from transformers import TrOCRProcessor, VisionEncoderDecoderModel |
|
|
|
|
|
processor = TrOCRProcessor.from_pretrained("paran3xus/typst_eq_ocr") |
|
|
model = VisionEncoderDecoderModel.from_pretrained('paran3xus/typst_eq_ocr') |
|
|
|
|
|
image_fps = [ |
|
|
'testimg/1.png', |
|
|
] |
|
|
images = [Image.open(fp).convert('RGB') for fp in image_fps] |
|
|
pixel_values = processor(images=images, return_tensors="pt").pixel_values |
|
|
generated_ids = model.generate(pixel_values) |
|
|
generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True) |
|
|
[print(i) for i in generated_text] |
|
|
``` |
|
|
### Using Typress Client with Web Frontend |
|
|
Check: https://github.com/ParaN3xus/typress |
