句子相似度计算

---

https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2

这里使用预训练的 nreimers/MiniLM-L6-H384-uncased 模型，并在 1B 句对数据集上微调。

---

如果你使用 sentence-transformers

pip install -U sentence-transformers

---

可以这样使用模型

import os ,sys from sentence_transformers import SentenceTransformer, utilmodel = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')def count(sent0, sent1):embeddings = model.encode([sent0, sent1])# print(embeddings)emb0 = embeddings[0]emb1 = embeddings[1]cosine_score = util.cos_sim(emb0, emb1) # tensor([[0.4046]])return float(cosine_score) # 0.4046

---

如果不使用 sentence-transformers

首先，将输入 传给 transformer 模型，然后使用 在上下文化的单词嵌入之上 应用正确的池操作。

from transformers import AutoTokenizer, AutoModel
import torch
import torch.nn.functional as F#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):token_embeddings = model_output[0] #First element of model_output contains all token embeddingsinput_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/all-MiniLM-L6-v2')
model = AutoModel.from_pretrained('sentence-transformers/all-MiniLM-L6-v2')# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')# Compute token embeddings
with torch.no_grad():model_output = model(**encoded_input)# Perform pooling
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])# Normalize embeddings
sentence_embeddings = F.normalize(sentence_embeddings, p=2, dim=1)print("Sentence embeddings:")
print(sentence_embeddings)

---