yoni2z/Amharic-E-commerce-Data-Extractor-Week4

# Amharic-E-commerce-Data-Extractor-Week4 Project for 10 Academy Week 4: Building an Amharic NER system for EthioMart. ## Setup 1. Clone the repository: `git clone ` 2. Install dependencies: `pip install -r requirements.txt` 3. Create `.env` file with Telegram API credentials. 4. Run scripts in `scripts/` for each task. ## Folder Structure - `data/`: Raw and processed data. - `scripts/`: Python scripts for tasks 1–6. - `notebooks/`: Jupyter notebooks for exploration. - `results/`: Model outputs and vendor scorecard. - `docs/`: Submission PDFs. ## Tasks 1. Data Ingestion: `scripts/data_ingestion/telegram_scraper.py` 2. Data Labeling: `scripts/data_labeling/label_data.py` 3. Model Fine-Tuning: `scripts/model_training/fine_tune_ner.py` 4. Model Comparison: `scripts/model_comparison/compare_models.py` 5. Model Interpretability: `scripts/model_interpretability/interpret_model.py` 6. Vendor Scorecard: `scripts/vendor_scorecard/vendor_scorecard.py`