LenaVolkova/Arm-address-extractor
This repository contains tools, data, and models for parsing, annotating, and structuring address information from raw Armenian text (such as public utility maintenance announcements).
A specialized toolset for parsing and structuring Armenian addresses from raw text, featuring a Flask annotation interface, PostgreSQL database with curated dataset, and LoRA fine-tuned Llama-3.1-8B model for automated extraction.
A pipeline for Armenian Named Entity Recognition (NER) and network analysis. Downloads Armenian text data, preprocess...
A Python package for collecting, processing, and normalizing a Western Armenian language corpus. It provides a full E...
ArmGPT is a student-friendly implementation of a GPT language model specifically trained on Armenian text. It's built...
A comprehensive pipeline for digitizing Classical Armenian (Grabar) texts from scanned PDFs into a searchable, transl...
AzgIntel is an AI-powered NLP pipeline for Armenian text classification using a BERT-based multilingual model. It pro...
An open-source OCR model trained for Armenian text recognition, supporting Classical, Western, and Eastern Armenian. ...