导入HuggingFace Sentence Transformers库耗时过长的原因咨询

嗨，我是刚接触HuggingFace的新手，如果这是个入门级问题还请见谅😅

我最近在试sentence-transformers的示例程序，发现from sentence_transformers import SentenceTransformer这条导入语句居然是个巨大的时间瓶颈——哪怕不是第一次运行它，耗时也长到离谱。

我有点搞不懂：既然虚拟环境里已经安装好这个库了，导入操作不该这么慢才对吧？

举个例子，我写了这段测试代码：

import time
start_time = time.time()
from sentence_transformers import SentenceTransformer
end_time = time.time()
print(f"Execution time: {end_time - start_time} seconds")
model = SentenceTransformer("multi-qa-mpnet-base-cos-v1")
end_time = time.time()
print(f"Execution time: {end_time - start_time} seconds")
query_embedding = model.encode("How big is London")
passage_embeddings = model.encode([
    "London is known for its financial district",
    "London has 9,787,426 inhabitants at the 2011 census",
    "The United Kingdom is the fourth largest exporter of goods in the world",
])

similarity = model.similarity(query_embedding, passage_embeddings)
# => tensor([[0.4659, 0.6142, 0.2697]])

print(similarity)
end_time = time.time()
print(f"Execution time: {end_time - start_time} seconds")

程序的输出结果是：

Execution time: 117.6684799194336 seconds
Execution time: 127.67643761634827 seconds
tensor([[0.4656, 0.6142, 0.2697]])
Execution time: 128.92580246925354 seconds

看得出来，光是from sentence_transformers import SentenceTransformer这一行就花了快两分钟。

想问问大家：是我哪里操作错了吗？还是这个库本身就有这样的特性？（安装的时候我是按照说明，激活虚拟环境后运行pip install -U sentence-transformers完成的）

我的设备是Dell Latitude 5540笔记本。

备注：内容来源于stack exchange，提问作者yam