multimodal-parsers
PDF processing pipeline: removes headers/footers, converts to markdown, and generates image captions using MLX VLM.
Installation
pip install multimodal-parsers
Dependencies
The package automatically installs:
- Pillow
- mlx-vlm
- pymupdf
- scikit-learn
- numpy
- marker-pdf
Additionally, you may need to install:
pip install "unstructured[pdf]"
Usage
After installation, use the multimodal-parsers command:
multimodal-parsers <input_dir> <output_dir>
Example
multimodal-parsers Database/Private/Files Database/Private/Files/output
What it does
- Removes headers and footers from PDF files using clustering algorithms
- Converts PDFs to markdown using marker-pdf
- Generates image captions using MLX VLM (InternVL3-1B-4bit)
- Outputs final markdown files with captioned images
Development
git clone https://github.com/thuuyen98/PIER-QA
cd PIER-QA
pip install -e ".[dev]"
License
MIT License
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
multimodal_parsers-0.1.6.tar.gz
(20.3 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file multimodal_parsers-0.1.6.tar.gz.
File metadata
- Download URL: multimodal_parsers-0.1.6.tar.gz
- Upload date:
- Size: 20.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.10.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
560b6c12b6dda55a04079af145fe1eb7558a614b9eba774218a3103e2b9a7e1c
|
|
| MD5 |
cec8cb6e282b67d6bba411096acd1beb
|
|
| BLAKE2b-256 |
7f689289d146e1cf1d415333dee90ea76382f6b30e05a2b173e5d61457a9b651
|
File details
Details for the file multimodal_parsers-0.1.6-py3-none-any.whl.
File metadata
- Download URL: multimodal_parsers-0.1.6-py3-none-any.whl
- Upload date:
- Size: 15.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.10.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e63a492a108eeda843065063b05555b53feb17178ee3c2c6ca2ca95e76582d2c
|
|
| MD5 |
a4f76e7b3efec75365a175354674aac3
|
|
| BLAKE2b-256 |
bdd5328cf0f83dc2c617523085eea2b7583be24caccb7f83015f9cf9f10b21d2
|
