VwPolicy#

class langchain_experimental.rl_chain.base.VwPolicy(model_repo: ModelRepository, vw_cmd: List[str], feature_embedder: Embedder, vw_logger: VwLogger, *args: Any, **kwargs: Any)[来源]#

Vowpal Wabbit 策略。

方法

__init__(model_repo, vw_cmd, ...)

learn(event)

log(event)

predict(event)

save()

Parameters:
__init__(model_repo: ModelRepository, vw_cmd: List[str], feature_embedder: Embedder, vw_logger: VwLogger, *args: Any, **kwargs: Any)[来源]#
Parameters:
learn(event: TEvent) → None[来源]#
Parameters:

事件 (TEvent)

Return type:

无

log(event: TEvent) → None[来源]#
Parameters:

事件 (TEvent)

Return type:

无

predict(event: TEvent) → Any[source]#
Parameters:

事件 (TEvent)

Return type:

任何

save() → None[来源]#
Return type:

无