BGE-M3 Financial Matryoshka

Primero instala la biblioteca de Sentence Transformers:

pip install -U sentence-transformers

Luego puedes cargar este modelo y ejecutar la inferencia:

from sentence_transformers import SentenceTransformer

# Descargar del Hub de 🤗
model = SentenceTransformer("haophancs/bge-m3-financial-matryoshka")
# Ejecutar inferencia
sentences = [
    'As of January 28, 2024 the net carrying value of our inventories was $1.3 billion, which included provisions for obsolete and damaged inventory of $139.7 million.',
    "What is the status of the company's inventory as of January 28, 2024, in terms of its valuation and provisions for obsolescence?",
    'What is the relationship between the ESG goals and the long-term growth strategy?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Obtener las puntuaciones de similitud para las incrustaciones
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]