Ошибка агента LangChain ⇐ Python

[Anonymous]

Я часто получаю эту ошибку в агенте langchain. Я выполняю операцию RAG и использую инструменты и агенты langchain. Для приложения я использую сервисы AWS, такие как Bedrock и Lambda.
Вот как я определил агентов:

Код: Выделить всё

        agent = initialize_agent(
tools=tools, llm=llm, agent="zero-shot-react-description", verbose=True
)

но я получаю ошибку типа:

Код: Выделить всё

handle_parsing_errors=True` to the AgentExecutor. This is the error: Could not parse LLM output: ``
Traceback (most recent call last):
File "/var/task/src/amaHandler.py", line 365, in lambdaHandler
finalAnswer = agent.run(query)
File "/var/lang/lib/python3.10/site-packages/langchain_core/_api/deprecation.py", line 180, in warning_emitting_wrapper
return wrapped(*args, **kwargs)
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 600, in run
return self(args[0], callbacks=callbacks, tags=tags, metadata=metadata)[
File "/var/lang/lib/python3.10/site-packages/langchain_core/_api/deprecation.py", line 180, in warning_emitting_wrapper
return wrapped(*args, **kwargs)
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 383, in __call__
return self.invoke(
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 166, in invoke
raise e
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 156, in invoke
self._call(inputs, run_manager=run_manager)
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1433, in _call
next_step_output = self._take_next_step(
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1139, in _take_next_step
[
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1139, in 
[
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1178, in _iter_next_step
raise ValueError(

Также отметим, что эта ошибка возникает не каждый раз, а примерно 1/5 раза.
Мой LangSmith показывает:

Код: Выделить всё

ValueError: An output parsing error occurred. In order to pass this error back to the agent and have it try again, pass `handle_parsing_errors=True` to the AgentExecutor.  This is the error: Could not parse LLM output: ` I now know the final answer to the original input.question
`

Traceback (most recent call last):
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1167, in _iter_next_step
output = self.agent.plan(
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 732, in plan
return self.output_parser.parse(full_output)
File "/var/lang/lib/python3.10/site-packages/langchain/agents/mrkl/output_parser.py", line 67, in parse
raise OutputParserException(
langchain_core.exceptions.OutputParserException: Could not parse LLM output: ` I now know the final answer to the original input.question
`

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/var/task/src/amaHandler.py", line 365, in lambdaHandler
finalAnswer = agent.run(query)
File "/var/lang/lib/python3.10/site-packages/langchain_core/_api/deprecation.py", line 180, in warning_emitting_wrapper
return wrapped(*args, **kwargs)
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 600, in run
return self(args[0], callbacks=callbacks, tags=tags, metadata=metadata)[
File "/var/lang/lib/python3.10/site-packages/langchain_core/_api/deprecation.py", line 180, in warning_emitting_wrapper
return wrapped(*args, **kwargs)
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 383, in __call__
return self.invoke(
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 166, in invoke
raise e
File "/var/lang/lib/python3.10/site-packages/langchain/chains/base.py", line 156, in invoke
self._call(inputs, run_manager=run_manager)
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1433, in _call
next_step_output = self._take_next_step(
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1139, in _take_next_step
[
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1139, in 
[
File "/var/lang/lib/python3.10/site-packages/langchain/agents/agent.py", line 1178, in _iter_next_step
raise ValueError(
ValueError: An output parsing error occurred. In order to pass this error back to the agent and have it try again, pass `handle_parsing_errors=True` to the AgentExecutor. This is the error: Could not parse LLM output: ` I now know the final answer to the original input.question
`

STRoutputparser сгенерировал ответ, но агент не сработал.
Попробовал то, что сказал агент: handle_parsing_error = true
, но ничего не вышло.
Пытался найти решения в Интернете, но ничего не нашел.

Подробнее здесь: https://stackoverflow.com/questions/790 ... nt-failing