| """ | |
| LLM Mail Trainer - Finance Entity Extraction Package. | |
| A production-grade system for extracting financial entities from emails | |
| using fine-tuned LLMs on Apple Silicon (MLX). | |
| Features: | |
| - Multi-bank email parsing (HDFC, ICICI, SBI, Axis, Kotak) | |
| - UPI/NEFT/IMPS transaction detection | |
| - Merchant and category classification | |
| - REST API for inference | |
| - LoRA fine-tuning support | |
| Example: | |
| >>> from src.data import EntityExtractor | |
| >>> extractor = EntityExtractor() | |
| >>> result = extractor.extract("Rs.500 debited from account 1234") | |
| >>> print(result.amount) | |
| '500' | |
| Author: Ranjit Behera | |
| License: MIT | |
| Version: 0.3.0 | |
| """ | |
| __version__ = "0.3.0" | |
| __author__ = "Ranjit Behera" | |
| __email__ = "ranjit@example.com" | |
| __license__ = "MIT" | |
| # Package-level imports for convenience | |
| from src.data.extractor import EntityExtractor, FinancialEntity | |
| from src.data.classifier import EmailClassifier, ClassificationResult | |
| from src.data.parser import EmailParser | |
| __all__ = [ | |
| "EntityExtractor", | |
| "FinancialEntity", | |
| "EmailClassifier", | |
| "ClassificationResult", | |
| "EmailParser", | |
| "__version__", | |
| ] | |