Instructions to use PaddlePaddle/PaddleOCR-VL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PaddleOCR
How to use PaddlePaddle/PaddleOCR-VL with PaddleOCR:
# See https://www.paddleocr.ai/latest/version3.x/pipeline_usage/PaddleOCR-VL.html to installation from paddleocr import PaddleOCRVL pipeline = PaddleOCRVL(pipeline_version="v1") output = pipeline.predict("path/to/document_image.png") for res in output: res.print() res.save_to_json(save_path="output") res.save_to_markdown(save_path="output") - Notebooks
- Google Colab
- Kaggle
Enable download counts
Congrats on the release! ๐๐ป
As the Hub best practice it'd be great to host different weights in different repositories.
This unlocks two things:
- If you expose transformers custom code weights on repo-level, the download counts will be enabled as well.
- This also enables "Use this model" snippet with transformers.
- You can host PaddlePaddle weights in a different repository, we can work on counting their downloads too afterwards ๐๐ป๐ค
Thank you for your valuable suggestion. We highly appreciate it and will take this issue into serious consideration.
@merve Thanks for your suggestions! We agree with you and think that is the best practice.
Therefore, I have moved the PaddleOCR-VL-0.9B model relevant files to the root directory of the repository and removed the PaddleOCR-VL-0.9B subdirectory.
As for the PP-DocLayoutV2 model currently present in the PaddlePaddle/PaddleOCR-VL repository, we will remove it at an appropriate time in the future. After that, the PP-DocLayoutV2 model will only be available in the PaddlePaddle/PP-DocLayoutV2 repository.