LEALLA-large

Usando el modelo:

import torch
from transformers import BertModel, BertTokenizerFast

tokenizer = BertTokenizerFast.from_pretrained("setu4993/LEALLA-large")
model = BertModel.from_pretrained("setu4993/LEALLA-large")
model = model.eval()

english_sentences = [
"dog",
"Puppies are nice.",
"I enjoy taking long walks along the beach with my dog.",
]
english_inputs = tokenizer(english_sentences, return_tensors="pt", padding=True)

with torch.no_grad():
  english_outputs = model(**english_inputs)

# Para obtener los embeddings de oración, use la salida del pooler:
english_embeddings = english_outputs.pooler_output

# Salida para otros idiomas:
italian_sentences = [
"cane",
"I cuccioli sono carini.",
"Mi piace fare lunghe passeggiate lungo la spiaggia con il mio cane.",
]
japanese_sentences = ["犬", "子犬はいいです", "私は犬と一緒にビーチを散歩するのが好きです"]
italian_inputs = tokenizer(italian_sentences, return_tensors="pt", padding=True)
japanese_inputs = tokenizer(japanese_sentences, return_tensors="pt", padding=True)

with torch.no_grad():
  italian_outputs = model(**italian_inputs)
  japanese_outputs = model(**japanese_inputs)

italian_embeddings = italian_outputs.pooler_output
japanese_embeddings = japanese_outputs.pooler_output

# Para la similitud entre oraciones, se recomienda una norma L2 antes de calcular la similitud:
import torch.nn.functional as F

def similarity(embeddings_1, embeddings_2):
  normalized_embeddings_1 = F.normalize(embeddings_1, p=2)
  normalized_embeddings_2 = F.normalize(embeddings_2, p=2)
  return torch.matmul(
    normalized_embeddings_1, normalized_embeddings_2.transpose(0, 1)
  )

print(similarity(english_embeddings, italian_embeddings))
print(similarity(english_embeddings, japanese_embeddings))
print(similarity(italian_embeddings, japanese_embeddings))