LLMDocParser

A package for parsing PDFs and analyzing their content using LLMs.

This package is an improvement based on the concept of gptpdf.

Method

gptpdf uses PyMuPDF to parse PDFs, identifying both text and non-text regions. It then merges or filters the text regions based on certain rules, and inputs the final results into a multimodal model for parsing. This method is particularly effective.

Based on this concept, I made some minor improvements.

Main Process

Using a layout analysis model, each page of the PDF is parsed to identify the type of each region, which includes Text, Title, Figure, Figure caption, Table, Table caption, Header, Footer, Reference, and Equation. The coordinates of each region are also obtained.

Layout Analysis Result Example:

[{'header': ((101, 66, 436, 102), 0)},
 {'header': ((1038, 81, 1088, 95), 1)},
 {'title': ((106, 215, 947, 284), 2)},
 {'text': ((101, 319, 835, 390), 3)},
 {'text': ((100, 565, 579, 933), 4)},
 {'text': ((100, 967, 573, 1025), 5)},
 {'text': ((121, 1055, 276, 1091), 6)},
 {'reference': ((101, 1124, 562, 1429), 7)},
 {'text': ((610, 565, 1089, 930), 8)},
 {'text': ((613, 976, 1006, 1045), 9)},
 {'title': ((612, 1114, 726, 1129), 10)},
 {'text': ((611, 1165, 1089, 1431), 11)},
 {'title': ((1011, 1471, 1084, 1492), 12)}]

This result includes the type, coordinates, and reading order of each region. By using this result, more precise rules can be set to parse the PDF.

Finally, input the images of the corresponding regions into a multimodal model, such as GPT-4o or Qwen-VL, to directly obtain text blocks that are friendly to RAG solutions.

img_path	type	page_no	filename	content	filepath
{absolute_path}/page_1_title.png	Title	1	attention is all you need	[Text Block 1]	{file_absolute_path}
{absolute_path}/page_1_text.png	Text	1	attention is all you need	[Text Block 2]	{file_absolute_path}
{absolute_path}/page_2_figure.png	Figure	2	attention is all you need	[Text Block 3]	{file_absolute_path}
{absolute_path}/page_2_figure_caption.png	Figure caption	2	attention is all you need	[Text Block 4]	{file_absolute_path}
{absolute_path}/page_3_table.png	Table	3	attention is all you need	[Text Block 5]	{file_absolute_path}
{absolute_path}/page_3_table_caption.png	Table caption	3	attention is all you need	[Text Block 6]	{file_absolute_path}
{absolute_path}/page_1_header.png	Header	1	attention is all you need	[Text Block 7]	{file_absolute_path}
{absolute_path}/page_2_footer.png	Footer	2	attention is all you need	[Text Block 8]	{file_absolute_path}
{absolute_path}/page_3_reference.png	Reference	3	attention is all you need	[Text Block 9]	{file_absolute_path}
{absolute_path}/page_1_equation.png	Equation	1	attention is all you need	[Text Block 10]	{file_absolute_path}

See more in llm_parser.py main function.

Installation

pip install llmdocparser

Installation from Source

To install this project from source, follow these steps:

Clone the Repository:

First, clone the repository to your local machine. Open your terminal and run the following commands:
```
git clone https://github.com/lazyFrogLOL/llmdocparser.git
cd llmdocparser
```
Install Dependencies:

This project uses Poetry for dependency management. Make sure you have Poetry installed. If not, you can follow the instructions in the Poetry Installation Guide.

Once Poetry is installed, run the following command in the project's root directory to install the dependencies:
```
poetry install
```
This will read the pyproject.toml file and install all the required dependencies for the project.

Usage

from llmdocparser.llm_parser import get_image_content

content, cost = get_image_content(
    llm_type="azure",
    pdf_path="path/to/your/pdf",
    output_dir="path/to/output/directory",
    max_concurrency=5,
    azure_deployment="azure-gpt-4o",
    azure_endpoint="your_azure_endpoint",
    api_key="your_api_key",
    api_version="your_api_version"
)
print(content)
print(cost)

Parameters

llm_type: str

The options are azure, openai, dashscope.
pdf_path: str

Path to the PDF file.
output_dir: str

Output directory to store all parsed images.
max_concurrency: int

Number of GPT parsing worker threads. Batch calling details: Batch Support

If using Azure, the azure_deployment and azure_endpoint parameters need to be passed; otherwise, only the API key needs to be provided.

base_url: str

OpenAI Compatible Server url. Detail: OpenAI-Compatible Server

Cost

Using the 'Attention Is All You Need' paper for analysis, the model chosen is GPT-4o, costing as follows:

Total Tokens: 44063
Prompt Tokens: 33812
Completion Tokens: 10251
Total Cost (USD): $0.322825

Average cost per page: $0.0215

AI-Jie01/llmdocparser