Pydantic ValidationError报错：ConversationBufferMemory2缺少chat字段

当前错误 #

ValidationError: 

File "/code/apps/llm_module/llm.py", line 183, in get_response
    memory = ConversationBufferMemory2()
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/langchain/load/serializable.py", line 74, in __init__
    super().__init__(**kwargs)
  File "pydantic/main.py", line 341, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for ConversationBufferMemory2
chat
  field required (type=value_error.missing)

相关代码片段 #

llm.py #

chat = Chat.objects.get(id=chat_id)
memory = ConversationBufferMemory2()
memory.set_chat(chat)

models2.py #

from langchain.memory.buffer import ConversationBufferMemory
class ConversationBufferMemory2(ConversationBufferMemory):
    """Buffer for storing conversation memory."""

    chat:Chat

    #def __init__(self,*args, **kwargs):
    #    super().__init__(*args, **kwargs)
        
    def set_chat(self,chat:Chat):
        self.chat=chat
        
    @property
    def buffer(self) -> Any:
        """String buffer of memory. Each message is separated by a newline."""
        if self.return_messages:
            return self.chat.messages
        else:
            return get_buffer_string(
                self.chat.messages,
                human_prefix=self.human_prefix,
                ai_prefix=self.ai_prefix,
            )

解决方案 #

方法一：实例化时直接传入chat参数 #

修改llm.py中的代码，直接在实例化阶段传入必填的chat字段，无需后续调用set_chat：

chat = Chat.objects.get(id=chat_id)
memory = ConversationBufferMemory2(chat=chat)

方法二：将chat字段设为可选字段 #

修改models2.py中chat字段的定义，标记为可选并设置默认值None，这样实例化时可以暂时不传值，之后再通过set_chat方法赋值：

from typing import Optional
from langchain.memory.buffer import ConversationBufferMemory

class ConversationBufferMemory2(ConversationBufferMemory):
    """Buffer for storing conversation memory."""

    chat: Optional[Chat] = None
    
    def set_chat(self, chat: Chat):
        self.chat = chat
        
    @property
    def buffer(self) -> Any:
        """String buffer of memory. Each message is separated by a newline."""
        if self.return_messages:
            return self.chat.messages
        else:
            return get_buffer_string(
                self.chat.messages,
                human_prefix=self.human_prefix,
                ai_prefix=self.ai_prefix,
            )

注意：使用self.chat.messages前要确保chat已经通过set_chat正确赋值，避免触发AttributeError。

方法三：重写__init__方法延迟验证（不推荐） #

如果必须保留chat为必填字段但需要延迟设置，可以重写__init__方法，暂时跳过Pydantic的初始化验证：

from langchain.memory.buffer import ConversationBufferMemory

class ConversationBufferMemory2(ConversationBufferMemory):
    """Buffer for storing conversation memory."""

    chat: Chat
    
    def __init__(self, *args, **kwargs):
        # 先移除chat参数，避免初始化时触发验证
        chat = kwargs.pop('chat', None)
        super().__init__(*args, **kwargs)
        if chat is not None:
            self.chat = chat
        
    def set_chat(self, chat: Chat):
        self.chat = chat
        
    @property
    def buffer(self) -> Any:
        """String buffer of memory. Each message is separated by a newline."""
        if self.return_messages:
            return self.chat.messages
        else:
            return get_buffer_string(
                self.chat.messages,
                human_prefix=self.human_prefix,
                ai_prefix=self.ai_prefix,
            )

这种方法需要额外的参数处理逻辑，仅在特殊场景下使用。

内容的提问来源于stack exchange，提问作者Reddspark