docker部署m3e-large-api无法启动

Question

#遇到问题的现象描述
使用docker一键部署的m3e-large-api无法启动。期待解答。
使用命令：docker run -d -p 6100:6008 --gpus all --name=m3e-large-api stawky/m3e-large-api:latest

#问题相关代码片，运行结果，报错内容
2024-01-11 15:09:29 No sentence-transformers model found with name ./moka-ai_m3e-large. Creating a new one with MEAN pooling.
2024-01-11 15:09:32 Traceback (most recent call last):
2024-01-11 15:09:32 File "/usr/local/bin/uvicorn", line 8, in
2024-01-11 15:09:32 sys.exit(main())
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/click/core.py", line 1157, in call
2024-01-11 15:09:32 return self.main(*args, **kwargs)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/click/core.py", line 1078, in main
2024-01-11 15:09:32 rv = self.invoke(ctx)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/click/core.py", line 1434, in invoke
2024-01-11 15:09:32 return ctx.invoke(self.callback, **ctx.params)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/click/core.py", line 783, in invoke
2024-01-11 15:09:32 return __callback(*args, **kwargs)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/main.py", line 416, in main
2024-01-11 15:09:32 run(
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/main.py", line 587, in run
2024-01-11 15:09:32 server.run()
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/server.py", line 61, in run
2024-01-11 15:09:32 return asyncio.run(self.serve(sockets=sockets))
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/asyncio/runners.py", line 44, in run
2024-01-11 15:09:32 return loop.run_until_complete(main)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/asyncio/base_events.py", line 616, in run_until_complete
2024-01-11 15:09:32 return future.result()
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/server.py", line 68, in serve
2024-01-11 15:09:32 config.load()
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/config.py", line 467, in load
2024-01-11 15:09:32 self.loaded_app = import_from_string(self.app)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/site-packages/uvicorn/importer.py", line 21, in import_from_string
2024-01-11 15:09:32 module = importlib.import_module(module_str)
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/importlib/init.py", line 127, in import_module
2024-01-11 15:09:32 return _bootstrap._gcd_import(name[level:], package, level)
2024-01-11 15:09:32 File "", line 1014, in _gcd_import
2024-01-11 15:09:32 File "", line 991, in _find_and_load
2024-01-11 15:09:32 File "", line 975, in _find_and_load_unlocked
2024-01-11 15:09:32 File "", line 671, in _load_unlocked
2024-01-11 15:09:32 File "", line 843, in exec_module
2024-01-11 15:09:32 File "", line 219, in _call_with_frames_removed
2024-01-11 15:09:32 File "/app/localembedding.py", line 42, in
2024-01-11 15:09:32 class ChatMessage(BaseModel):
2024-01-11 15:09:32 File "pydantic/main.py", line 197, in pydantic.main.ModelMetaclass.new
2024-01-11 15:09:32 File "pydantic/fields.py", line 506, in pydantic.fields.ModelField.infer
2024-01-11 15:09:32 File "pydantic/fields.py", line 436, in pydantic.fields.ModelField.init
2024-01-11 15:09:32 File "pydantic/fields.py", line 552, in pydantic.fields.ModelField.prepare
2024-01-11 15:09:32 File "pydantic/fields.py", line 668, in pydantic.fields.ModelField._type_analysis
2024-01-11 15:09:32 File "/usr/local/lib/python3.8/typing.py", line 774, in subclasscheck
2024-01-11 15:09:32 return issubclass(cls, self.origin)
2024-01-11 15:09:32 TypeError: issubclass() arg 1 must be a class

#我的初步解答思路是

#操作环境、软件版本等相关信息
win10 + docker4.26.1 + Ubuntu（不记得版本了，大概是22）

Answer 1

【以下回答由 GPT 生成】

由于报错信息中提到了缺少sentence-transformers模型，因此需要确保在启动容器之前，需要提前下载并加载对应的模型。下面是具体的解决方案：
1. 确保已经下载了正确的sentence-transformers模型，并且该模型的地址为./moka-ai_m3e-large。

2. 创建一个Dockerfile，内容如下：

```Dockerfile
FROM stawky/m3e-large-api:latest
WORKDIR /app
COPY ./moka-ai_m3e-large /app/moka-ai_m3e-large

1. 构建一个自定义的镜像，使用以下命令：

docker build -t my-m3e-image .

1. 使用该自定义镜像来运行容器，确保正确映射端口：

docker run -d -p 6100:6008 --gpus all --name=m3e-large-api my-m3e-image

通过以上步骤，可以确保正确加载了所需的模型，从而解决了启动时缺少模型的问题。 ```

如果你已经解决了该问题, 非常希望你能够分享一下解决方案, 写成博客, 将相关链接放在评论区, 以帮助更多的人 ^-^