How to use AniruddhaAI/sanskrit-e5-small-retrieval with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("AniruddhaAI/sanskrit-e5-small-retrieval") sentences = [ "query: The two princes Vinda and Anuvinda, both fierce bowmen, for the sake of their friend Duryodhana, rushed upon Virata the ruler of the Matsyas, heedless of their lives and holding their weapons over head.", "passage: Invite them with plenty of seats and welcome them; for they remaining unseated, who is capable of taking his seat?", "passage: The Sankhya system, the Aranyaka-Veda, and the Pancharatra scriptures, are all identical and form parts of one whole. This is the religion of those who are devoted whole. mindedly to Narayana—the religion that has Narayana for its Soul.", "passage: विन्दानुविन्दावावन्त्यौविराटं मस्यमार्च्छताम्। प्राणांस्त्यक्त्वा महेष्वासौ मित्रार्थेऽभ्युद्यतायुधौ॥" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4]
