使用pandas和BERT将余弦相似度公式从一个数据帧循环到另一个数据帧

import pandas as pd
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity

model = SentenceTransformer('bert-base-nli-mean-tokens')

embeddings1 = model.encode(df1['文本列'].tolist())
embeddings2 = model.encode(df2['文本列'].tolist())

similarity_matrix = cosine_similarity(embeddings1, embeddings2)

df1['相似度列'] = similarity_matrix.tolist()

import pandas as pd
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity

# 加载BERT模型
model = SentenceTransformer('bert-base-nli-mean-tokens')

# 准备数据
df1 = pd.DataFrame({'文本列': ['文本1', '文本2', '文本3']})
df2 = pd.DataFrame({'文本列': ['文本4', '文本5', '文本6']})

# 对文本数据进行向量化
embeddings1 = model.encode(df1['文本列'].tolist())
embeddings2 = model.encode(df2['文本列'].tolist())

# 计算余弦相似度
similarity_matrix = cosine_similarity(embeddings1, embeddings2)

# 将余弦相似度矩阵添加到df1或df2
df1['相似度列'] = similarity_matrix.tolist()

print(df1)