sandhi

denormalized sanskrit sandhi corpus based on panini sutra

This repository provides a character-level model for Sanskrit text classification, predicting whether each character in a word is part of a Sandhi point (SP) or not (NSP). It includes preprocessing, training with an MLP model, and APIs (FastAPI) and a Telegram bot for easy predictions.

This HTML code creates a web page for a Sanskrit Sandhi tool, using HTML for layout, CSS for appearance, and JavaScript to process inputs, call the Gemini API for results, and display them.

A Python web project likely focused on text processing or natural language tasks, possibly related to linguistic "sandhi", with a development setup using Poetry and configurations for Project IDX or VS Code.

Sandhify (सन्धीराट्) — a rule-based, Paninian Sanskrit sandhi engine that deterministically applies Ashtadhyayi sutras to produce fully sandhified sentences, a boundary summary table, and a step-by-step prakriya (derivation) with exact sutra citations. It includes a Streamlit web UI, multi-script transliteration (Aksharamukha), and a Python API for

MoonBit port of Samsaadhanii's table-driven Sanskrit sandhi & morphological analysis — native CLI + WasmGC