Skip to main content

Pandas Dataframe

本笔记本展示了如何使用代理与 Pandas DataFrame 进行交互。它主要针对问答进行优化。

注意:此代理在后台调用 Python 代理,执行 LLM 生成的 Python 代码 - 如果 LLM 生成的 Python 代码有害,这可能是危险的。请谨慎使用。

from langchain.agents.agent_types import AgentType
from langchain_experimental.agents.agent_toolkits import create_pandas_dataframe_agent
from langchain_openai import ChatOpenAI
import pandas as pd
from langchain_openai import OpenAI

df = pd.read_csv(
"https://raw.githubusercontent.com/pandas-dev/pandas/main/doc/data/titanic.csv"
)

使用 ZERO_SHOT_REACT_DESCRIPTION

这展示了如何使用 ZERO_SHOT_REACT_DESCRIPTION 代理类型初始化代理。

agent = create_pandas_dataframe_agent(OpenAI(temperature=0), df, verbose=True)

使用 OpenAI 函数

这展示了如何使用 OPENAI_FUNCTIONS 代理类型初始化代理。请注意,这是一种替代上述方法。

agent = create_pandas_dataframe_agent(
ChatOpenAI(temperature=0, model="gpt-3.5-turbo-0613"),
df,
verbose=True,
agent_type=AgentType.OPENAI_FUNCTIONS,
)
agent.invoke("有多少行?")


> Entering new chain...

Invoking: `python_repl_ast` with `df.shape[0]`


891数据框中有 891 行。

> Finished chain.
'数据框中有 891 行。'
agent.invoke("有多少人有超过 3 个兄弟姐妹")


> Entering new AgentExecutor chain...
Thought: 我需要计算有超过 3 个兄弟姐妹的人数
Action: python_repl_ast
Action Input: df[df['SibSp'] > 3].shape[0]
Observation: 30
Thought: 我现在知道最终答案
Final Answer: 30 人有超过 3 个兄弟姐妹。

> Finished chain.
'30 人有超过 3 个兄弟姐妹。'
agent.invoke("平均年龄的平方根是多少?")


> Entering new AgentExecutor chain...
Thought: 我需要先计算平均年龄
Action: python_repl_ast
Action Input: df['Age'].mean()
Observation: 29.69911764705882
Thought: 我现在需要计算平均年龄的平方根
Action: python_repl_ast
Action Input: math.sqrt(df['Age'].mean())
Observation: NameError("name 'math' is not defined")
Thought: 我需要导入 math 库
Action: python_repl_ast
Action Input: import math
Observation: 
Thought: 我现在需要计算平均年龄的平方根
Action: python_repl_ast
Action Input: math.sqrt(df['Age'].mean())
Observation: 5.449689683556195
Thought: 我现在知道最终答案
Final Answer: 平均年龄的平方根是 5.449689683556195。

> Finished chain.
'平均年龄的平方根是 5.449689683556195。'

多数据框示例

下一部分展示了代理如何与作为列表传入的多个数据框进行交互。

df1 = df.copy()
df1["Age"] = df1["Age"].fillna(df1["Age"].mean())
agent = create_pandas_dataframe_agent(OpenAI(temperature=0), [df, df1], verbose=True)
agent.invoke("how many rows in the age column are different?")


> Entering new AgentExecutor chain...
Thought: I need to compare the age columns in both dataframes
Action: python_repl_ast
Action Input: len(df1[df1['Age'] != df2['Age']])
Observation: 177
Thought: I now know the final answer
Final Answer: 177 rows in the age column are different.

> Finished chain.
'177 rows in the age column are different.'

此页面是否有帮助?


您还可以留下详细的反馈 在 GitHub 上