promptflow.integrations.langchain module

class promptflow.integrations.langchain.LangChainEventType(value)

Bases: Enum

An enumeration.

AGENT = ('AGENT', 3)

CHAIN = ('CHAIN', 1)

LLM = ('LLM', 0)

TOOL = ('TOOL', 2)

class promptflow.integrations.langchain.PromptFlowCallbackHandler

Bases: BaseCallbackHandler

PromptFlowCallbackHandler implements the langchain.callbacks.base.BaseCallbackHandler interface, which has a method for each event that can be subscribed to. The appropriate method will be called on the handler when the event is triggered.

property always_verbose: bool

Whether to always be verbose.

on_agent_action(action: AgentAction, **kwargs: Any) → None

Run on agent action.

Parameters:

action (AgentAction) – The action from agent.

on_agent_finish(finish: AgentFinish, **kwargs: Any) → None

Run on agent end.

Parameters:

finish (AgentFinish) – The finish from agent.

on_chain_end(outputs: Dict[str, Any], **kwargs: Any) → None

Run when chain ends running.

Parameters:

outputs (Dict[str, Any]) – The outputs from chain.

on_chain_error(error: Union[Exception, KeyboardInterrupt], **kwargs: Any) → None

Run when chain errors.

Parameters:

error (Union[Exception, KeyboardInterrupt]) – The error from chain.

on_chain_start(serialized: Dict[str, Any], inputs: Dict[str, Any], **kwargs: Any) → None

Run when chain starts running.

Parameters:

• serialized (Dict[str, Any]) – The serialized chain object.

• inputs (Dict[str, Any]) – The inputs used to run chain.

on_llm_end(response: LLMResult, **kwargs: Any) → None

Run when LLM ends running.

Parameters:

response (LLMResult) – The response from LLM.

on_llm_error(error: Union[Exception, KeyboardInterrupt], **kwargs: Any) → None

Run when LLM errors.

Parameters:

error (Union[Exception, KeyboardInterrupt]) – The error from LLM.

on_llm_new_token(token: str, **kwargs: Any) → None

Run on new LLM token. Only available when streaming is enabled.

Parameters:

token (str) – The new token.

on_llm_start(serialized: Dict[str, Any], prompts: List[str], **kwargs: Any) → None

Run when LLM starts running.

Parameters:

• serialized (Dict[str, Any]) – The serialized LLM object.

• prompts (List[str]) – The prompts used to run LLM.

on_text(text: str, **kwargs: Any) → None

Run on arbitrary text.

Parameters:

text (str) – The text.

on_tool_end(output: str, **kwargs: Any) → None

Run when tool ends running.

Parameters:

output (str) – The output from tool.

on_tool_error(error: Union[Exception, KeyboardInterrupt], **kwargs: Any) → None

Run when tool errors.

Parameters:

error (Union[Exception, KeyboardInterrupt]) – The error from tool.

on_tool_start(serialized: Dict[str, Any], input_str: str, **kwargs: Any) → None

Run when tool starts running.

Parameters:

• serialized (Dict[str, Any]) – The serialized tool object.

• input_str (str) – The input string used to run tool.