mirror of
https://github.com/hwchase17/langchain
synced 2024-11-10 01:10:59 +00:00
23 lines
629 B
Python
23 lines
629 B
Python
"""**Agent** is a class that uses an LLM to choose
|
|
a sequence of actions to take.
|
|
|
|
In Chains, a sequence of actions is hardcoded. In Agents,
|
|
a language model is used as a reasoning engine to determine which actions
|
|
to take and in which order.
|
|
|
|
Agents select and use **Tools** and **Toolkits** for actions.
|
|
"""
|
|
from langchain_experimental.agents.agent_toolkits import (
|
|
create_csv_agent,
|
|
create_pandas_dataframe_agent,
|
|
create_spark_dataframe_agent,
|
|
create_xorbits_agent,
|
|
)
|
|
|
|
__all__ = [
|
|
"create_csv_agent",
|
|
"create_pandas_dataframe_agent",
|
|
"create_spark_dataframe_agent",
|
|
"create_xorbits_agent",
|
|
]
|