构建LangChain应用出现的TypeError错误

在阅读LangChain官网给出的一些案列时，实际运行却报错，案列代码如下：

from langchain.prompts import ChatPromptTemplate
from langchain_community.chat_models import ChatOpenAI
from langchain_core.output_parsers import StrOutputParser

prompt = ChatPromptTemplate.from_template("tell me a short joke about {topic}")
model = ChatOpenAI()
output_parser = StrOutputParser()

chain = prompt | model | output_parser

chain.invoke({"topic": "ice cream"})

错误1：TypeError: Expected a Runnable, callable or dict.Instead got an unsupported type:

完整报错信息如下：

Traceback (most recent call last):
  File "~/PycharmProjects/LangChain/main.py", line 14, in <module>
    chain = prompt | model | output_parser
  File "~/opt/anaconda3/envs/langchain/lib/python3.8/site-packages/langchain/schema/runnable/base.py", line 1165, in __or__
    last=coerce_to_runnable(other),
  File "~/opt/anaconda3/envs/langchain/lib/python3.8/site-packages/langchain/schema/runnable/base.py", line 2774, in coerce_to_runnable
    raise TypeError(
TypeError: Expected a Runnable, callable or dict.Instead got an unsupported type: <class 'langchain_core.output_parsers.string.StrOutputParser'>

出现这个错误的原因是因为输出解析器不正确，不支持StrOutputParser而是要使用BaseOutputParser，因此我们可以自己来实现一个BaseOutputParser：
如下：

class CommaSeparatedListOutputParser(BaseOutputParser):
    """Parse the output of an LLM call to a comma-separated list."""

    def parse(self, text: str):
        """Parse the output of an LLM call."""

        return text.strip()

即更新代码如下：

import os
from langchain.prompts import ChatPromptTemplate
from langchain_community.chat_models import ChatOpenAI
from langchain.schema import BaseOutputParser

os.environ["OPENAI_API_KEY"] = "xxx"

class CommaSeparatedListOutputParser(BaseOutputParser):
    """Parse the output of an LLM call to a comma-separated list."""

    def parse(self, text: str):
        """Parse the output of an LLM call."""

        return text.strip()


prompt = ChatPromptTemplate.from_template("tell me a short joke about {topic}")
model = ChatOpenAI()
output_parser = CommaSeparatedListOutputParser()

chain = prompt | model | output_parser

chain.invoke({"topic": "ice cream"})

仍然报错2

错误2:TypeError: Got unknown type (‘messages’, [HumanMessage(content=‘tell me a short joke about ice cream’)])

错误原因是引入ChatOpenAI的包不对，原始的引入是from langchain_community.chat_models import ChatOpenAI改为from langchain.chat_models import ChatOpenAI即可，修改上面2处问题后，即可正确运行代码

完整正确的代码如下

import os
from langchain.prompts import ChatPromptTemplate
from langchain.chat_models import ChatOpenAI
from langchain.schema import BaseOutputParser

os.environ["OPENAI_API_KEY"] = "xxx"


class CommaSeparatedListOutputParser(BaseOutputParser):
    """Parse the output of an LLM call to a comma-separated list."""

    def parse(self, text: str):
        """Parse the output of an LLM call."""

        return text.strip()


prompt = ChatPromptTemplate.from_template("tell me a short joke about {topic}")
model = ChatOpenAI()
output_parser = CommaSeparatedListOutputParser()

chain = prompt | model | output_parser

res = chain.invoke({"topic": "ice cream"})
print(res)

输出：

Why did the ice cream go to therapy?
Because it had too many toppings and couldn't keep its sprinkles together!