- 2024/10/31 0.9.0 released. This is a major new version with extensive code refactoring, addressing numerous issues, improving performance, reducing hardware requirements, and enhancing usability:
- Refactored the sorting module code to use [layoutreader](https://github.com/ppaanngggg/layoutreader) for reading order sorting, ensuring high accuracy in various layouts.
- Refactored the paragraph concatenation module to achieve good results in cross-column, cross-page, cross-figure, and cross-table scenarios.
- Refactored the list and table of contents recognition functions, significantly improving the accuracy of list blocks and table of contents blocks, as well as the parsing of corresponding text paragraphs.
- Refactored the matching logic for figures, tables, and descriptive text, greatly enhancing the accuracy of matching captions and footnotes to figures and tables, and reducing the loss rate of descriptive text to zero.
- Added multi-language support for OCR, supporting detection and recognition of 84 languages.For the list of supported languages, see [OCR Language Support List](https://paddlepaddle.github.io/PaddleOCR/latest/en/ppocr/blog/multi_languages.html#5-support-languages-and-abbreviations).
- Added memory recycling logic and other memory optimization measures, significantly reducing memory usage. The memory requirement for enabling all acceleration features except table acceleration (layout/formula/OCR) has been reduced from 16GB to 8GB, and the memory requirement for enabling all acceleration features has been reduced from 24GB to 10GB.
- Optimized configuration file feature switches, adding an independent formula detection switch to significantly improve speed and parsing results when formula detection is not needed.
- Added the self-developed `doclayout_yolo` model, which speeds up processing by more than 10 times compared to the original solution while maintaining similar parsing effects, and can be freely switched with `layoutlmv3` via the configuration file.
- Upgraded formula parsing to `unimernet 0.2.1`, improving formula parsing accuracy while significantly reducing memory usage.
- 2024/09/27 Version 0.8.1 released, Fixed some bugs, and providing a [localized deployment version](projects/web_demo/README.md) of the [online demo](https://opendatalab.com/OpenSourceTools/Extractor/PDF/) and the [front-end interface](projects/web/README.md).
- 2024/09/27 Version 0.8.1 released, Fixed some bugs, and providing a [localized deployment version](projects/web_demo/README.md) of the [online demo](https://opendatalab.com/OpenSourceTools/Extractor/PDF/) and the [front-end interface](projects/web/README.md).
- 2024/09/09: Version 0.8.0 released, supporting fast deployment with Dockerfile, and launching demos on Huggingface and Modelscope.
- 2024/09/09: Version 0.8.0 released, supporting fast deployment with Dockerfile, and launching demos on Huggingface and Modelscope.
-[magic_pdf_parse_main.py More Detailed Processing Workflow](demo/magic_pdf_parse_main.py)
-[magic_pdf_parse_main.py More Detailed Processing Workflow](demo/magic_pdf_parse_main.py)
### Deploy Derived Projects
Derived projects include secondary development projects based on MinerU by project developers and community developers,
such as application interfaces based on Gradio, RAG based on llama, web demos similar to the official website, lightweight multi-GPU load balancing client/server ends, etc.
These projects may offer more features and a better user experience.
For specific deployment methods, please refer to the [Derived Project README](projects/README.md)
### Development Guide
### Development Guide
TODO
TODO
# TODO
# TODO
- [x] Semantic-based reading order
- 🗹 Reading order based on the model
- [ ] List recognition within the text
- 🗹 Recognition of `index` and `list` in the main text
- [ ] Code block recognition within the text
- 🗹 Table recognition
- [ ] Table of contents recognition
- ☐ Code block recognition in the main text
- [x] Table recognition
- ☐ [Chemical formula recognition](docs/chemical_knowledge_introduction/introduction.pdf)
-[ ] [Chemical formula recognition](docs/chemical_knowledge_introduction/introduction.pdf)
- ☐ Geometric shape recognition
- [ ] Geometric shape recognition
# Known Issues
# Known Issues
- Reading order is segmented based on rules, which can cause disordered sequences in some cases
- Reading order is determined by the model based on the spatial distribution of readable content, and may be out of order in some areas under extremely complex layouts.
- Vertical text is not supported
- Vertical text is not supported.
- Lists, code blocks, and table of contents are not yet supported in the layout model
- Tables of contents and lists are recognized through rules, and some uncommon list formats may not be recognized.
- Comic books, art books, elementary school textbooks, and exercise books are not well-parsed yet
- Only one level of headings is supported; hierarchical headings are not currently supported.
- Enabling OCR may produce better results in PDFs with a high density of formulas
- Code blocks are not yet supported in the layout model.
- If you are processing PDFs with a large number of formulas, it is strongly recommended to enable the OCR function. When using PyMuPDF to extract text, overlapping text lines can occur, leading to inaccurate formula insertion positions.
- Comic books, art albums, primary school textbooks, and exercises cannot be parsed well.
- Table recognition may result in row/column recognition errors in complex tables.
- OCR recognition may produce inaccurate characters in PDFs of lesser-known languages (e.g., diacritical marks in Latin script, easily confused characters in Arabic script).
- Some formulas may not render correctly in Markdown.
| NVIDIA-SMI 537.34 Driver Version: 537.34 CUDA Version: 12.2 |
| NVIDIA-SMI 537.34 Driver Version: 537.34 CUDA Version: 12.2 |
...
@@ -95,8 +97,6 @@ magic-pdf -p small_ocr.pdf
...
@@ -95,8 +97,6 @@ magic-pdf -p small_ocr.pdf
If your graphics card has at least **8GB** of VRAM, follow these steps to test CUDA acceleration:
If your graphics card has at least **8GB** of VRAM, follow these steps to test CUDA acceleration:
> ❗ Due to the extremely limited nature of 8GB VRAM for running this application, you need to close all other programs using VRAM to ensure that 8GB of VRAM is available when running this application.
1. Modify the value of `"device-mode"` in the `magic-pdf.json` configuration file located in your home directory.
1. Modify the value of `"device-mode"` in the `magic-pdf.json` configuration file located in your home directory.
@@ -60,8 +60,6 @@ Download a sample file from the repository and test it.
...
@@ -60,8 +60,6 @@ Download a sample file from the repository and test it.
If your graphics card has at least 8GB of VRAM, follow these steps to test CUDA-accelerated parsing performance.
If your graphics card has at least 8GB of VRAM, follow these steps to test CUDA-accelerated parsing performance.
> ❗ Due to the extremely limited nature of 8GB VRAM for running this application, you need to close all other programs using VRAM to ensure that 8GB of VRAM is available when running this application.
1.**Overwrite the installation of torch and torchvision** supporting CUDA.
1.**Overwrite the installation of torch and torchvision** supporting CUDA.
@@ -22,7 +22,9 @@ The configuration file can be found in the user directory, with the filename `ma
...
@@ -22,7 +22,9 @@ The configuration file can be found in the user directory, with the filename `ma
> Due to feedback from some users that downloading model files using git lfs was incomplete or resulted in corrupted model files, this method is no longer recommended.
> Due to feedback from some users that downloading model files using git lfs was incomplete or resulted in corrupted model files, this method is no longer recommended.
If you previously downloaded model files via git lfs, you can navigate to the previous download directory and use the `git pull` command to update the model.
When magic-pdf <= 0.8.1, if you have previously downloaded the model files via git lfs, you can navigate to the previous download directory and update the models using the `git pull` command.
> For versions 0.9.x and later, due to the repository change and the addition of the layout sorting model in PDF-Extract-Kit 1.0, the models cannot be updated using the `git pull` command. Instead, a Python script must be used for one-click updates.
## 2. Models downloaded via Hugging Face or Model Scope
## 2. Models downloaded via Hugging Face or Model Scope
