demo/demo_retrieval_agent.ipynb报错

[XduSyL]

Initial Checks

• I have searched GitHub for a duplicate issue and I'm sure this is something new
• I have read and followed the docs & demos and still think this is a bug
• I am confident that the issue is with modelscope-agent (not my code, or another library in the ecosystem)

What happened + What you expected to happen

作者您好，非常感谢您的工作
我在研究您的工作时，出现了一些问题，还请麻烦您解答一下
在运行demo/demo_retrieval_agent.ipynb的例子进行向量库检索时，在实例化memory类时报错了

memory = MemoryWithRetrievalKnowledge(storage_path=storage_path,
                                         name = memory_agent_name,
                                         memory_path=memory_history_path)

跳到了这一行

self.embedding = embedding or ModelScopeEmbeddings(
            model_id='damo/nlp_gte_sentence-embedding_chinese-base')

报了如下错误
Could not import some python packages.Please install it with pip install modelscope.
ImportError: cannot import name 'models' from partially initialized module 'oss2' (most likely due to a circular import) (d:\Anaconda\envs\agent\lib\site-packages\oss2_init_.py)

The above exception was the direct cause of the following exception:

File "D:\Code\modelscope-agent\modelscope_agent\storage\vector_storage.py", line 30, in init
self.embedding = embedding or ModelScopeEmbeddings(
File "D:\Code\modelscope-agent\modelscope_agent\memory\memory_with_retrieval_knowledge.py", line 31, in init
self.store_knowledge = KnowledgeVector(
File "D:\Code\modelscope-agent\main.py", line 20, in
memory = MemoryWithRetrievalKnowledge(storage_path=storage_path,
ImportError: Could not import some python packages.Please install it with pip install modelscope.

Versions / Dependencies

无

Reproduction script

无

Issue Severity

None

[zzhangpurdue]

是否本地pip install modelscope

[XduSyL]

是否本地pip install modelscope
在本地安装过modelscope了

[XduSyL]

是否本地pip install modelscope

执行其他demo是正常的

[zzhangpurdue]

目前没法复现，可能的是，里面有些requirements出现了版本问题。你是完全跟据demo一行行执行下来的么？那个requirements.txt也是完整安装的么？

[XduSyL]

请问一下在demo/demo_qwen_agent.ipynb的demo中，如果我要与大模型进行多轮会话的话要怎么操作呀，因为，我看RolePlay这个类实例化之后，最大执行次数默认为10，但是好像是单次就break了。

[zzhangpurdue]

这里的执行测试默认10，相当于他在做self-reflection，每一步他的结果他会去分析，还有什么不对的或者需要完善的，如果没有了，这一轮结束。也就是该单次聊天结束，需要多轮的话，参考这个链接