MenbereHailu/Amharic-E-commerce-Data-Extractor

Transform messy Telegram posts into a smart FinTech engine that reveals which vendors are the best candidates for a loan. # 10_Academy_AI_Mastery_kifiya_Week_5 10_Academy_AI_Mastery_kifiya_Week_5 # Machine Learning (specifically LLM) Project ## Overview This project focuses on fine-tuning LLM’s for Amharic Named Entity Recognition (NER) system that extracts key business entities such as product names, prices, and Locations, from text, images, and documents shared across these Telegram channels. The extracted data will be used to populate EthioMart's centralised database, making it a comprehensive e-commerce hub. ## Project Structure - `src/`: Source extracted data from telegram for the project. - `notebooks/`: Jupyter notebooks for experimentation. - `tests/`: Unit tests for project validation. - `scripts/`: Utility scripts for running tasks, like codes for data scrapping, loading, preparation, transformation, and LLM. ## Setup 1. Clone the repository. 2. Install dependencies: `pip install -r requirements.txt`. 3. Run tests: `python -m unittest discover -s tests`. ## Usage Run the main script: ```bash python Notebooks/main.ipynb