Financial Document Analyzer (Fixed)

Overview

This project analyzes financial documents (PDF, DOCX) to extract key values and summaries. It was originally buggy, but has been debugged and improved.

Fixes

• Fixed broken imports and missing functions
• Fixed file handling (PyPDF2 issues)
• Improved prompts for LLM
• Added error handling for robustness

Bonus Features

• Added Celery + Redis queue worker for concurrent analysis
• Added SQLite + SQLAlchemy database for storing results

Usage

pip install -r requirements.txt
python financial_analyzer.py

To run Celery worker:

celery -A tasks worker --loglevel=info

API

• run_analysis(file_path, use_llm=False): Run analysis on a document
• Returns: (report.pdf, metrics_dict, summary, llm_summary)

Outputs

• report.pdf → Analysis report
• analysis.db → Database storing documents and analyses