How to avoid the backticks ( ` ) around the Though in create_spark_dataframe_agent procedure.

[MayurPatil27-11]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

agent = create_spark_dataframe_agent(llm=CxxxChxModel(model_name='@xxx',temperature=0.001), df=df, additional_tools=[PythonREPLTool()], agent_type=AgentType.OPENAI_MULTI_FUNCTIONS, allow_dangerous_code = True,verbose = True,handle_parsing_errors = True,prefix=prompt,agent_executor_kwargs=dict(handle_parsing_errors='''Remove backticks "`" and "*" from above output, or use the remove_backticks_tool tool, Check your output and make sure it confirms, use the Action/Action Input syntax, give previous observation as output if observation/action is repeating'''),kwargs={"remove_backticks":remove_backticks_tool})

Description

Had created the agent as above mentioned code, but while running the query using python repl tool it is sometime giving the Observation as expected. But while thinking about that particular observation it adds backticks around the Thought; which then creates the problem while Parsing the LLM.

System Info

NA