Pandas Dataframe
This notebook shows how to use agents to interact with aPandas DataFrame. It is mostly optimized for question answering.
NOTE: this agent calls thePython agent under the hood, which executes LLM generated Python code - this can be bad if the LLM generated Python code is harmful. Use cautiously.
NOTE: Since langchain migrated to v0.3 you should upgrade langchain_openai and langchain. This would avoid import errors.
pip install --upgrade langchain_openaipip install --upgrade langchain
from langchain.agents.agent_typesimport AgentType
from langchain_experimental.agents.agent_toolkitsimport create_pandas_dataframe_agent
from langchain_openaiimport ChatOpenAI
import pandasas pd
from langchain_openaiimport OpenAI
df= pd.read_csv(
"https://raw.githubusercontent.com/pandas-dev/pandas/main/doc/data/titanic.csv"
)
UsingZERO_SHOT_REACT_DESCRIPTION
This shows how to initialize the agent using theZERO_SHOT_REACT_DESCRIPTION agent type.
agent= create_pandas_dataframe_agent(OpenAI(temperature=0), df, verbose=True)
Using OpenAI Functions
This shows how to initialize the agent using the OPENAI_FUNCTIONS agent type. Note that this is an alternative to the above.
agent= create_pandas_dataframe_agent(
ChatOpenAI(temperature=0, model="gpt-3.5-turbo-0613"),
df,
verbose=True,
agent_type=AgentType.OPENAI_FUNCTIONS,
)
agent.invoke("how many rows are there?")
[1m> Entering new chain...[0m
[32;1m[1;3m
Invoking: `python_repl_ast` with `df.shape[0]`
[0m[36;1m[1;3m891[0m[32;1m[1;3mThere are 891 rows in the dataframe.[0m
[1m> Finished chain.[0m
'There are 891 rows in the dataframe.'
agent.invoke("how many people have more than 3 siblings")
[1m> Entering new AgentExecutor chain...[0m
[32;1m[1;3mThought: I need to count the number of people with more than 3 siblings
Action: python_repl_ast
Action Input: df[df['SibSp'] > 3].shape[0][0m
Observation: [36;1m[1;3m30[0m
Thought:[32;1m[1;3m I now know the final answer
Final Answer: 30 people have more than 3 siblings.[0m
[1m> Finished chain.[0m
'30 people have more than 3 siblings.'
agent.invoke("whats the square root of the average age?")
[1m> Entering new AgentExecutor chain...[0m
[32;1m[1;3mThought: I need to calculate the average age first
Action: python_repl_ast
Action Input: df['Age'].mean()[0m
Observation: [36;1m[1;3m29.69911764705882[0m
Thought:[32;1m[1;3m I now need to calculate the square root of the average age
Action: python_repl_ast
Action Input: math.sqrt(df['Age'].mean())[0m
Observation: [36;1m[1;3mNameError("name 'math' is not defined")[0m
Thought:[32;1m[1;3m I need to import the math library
Action: python_repl_ast
Action Input: import math[0m
Observation: [36;1m[1;3m[0m
Thought:[32;1m[1;3m I now need to calculate the square root of the average age
Action: python_repl_ast
Action Input: math.sqrt(df['Age'].mean())[0m
Observation: [36;1m[1;3m5.449689683556195[0m
Thought:[32;1m[1;3m I now know the final answer
Final Answer: The square root of the average age is 5.449689683556195.[0m
[1m> Finished chain.[0m
'The square root of the average age is 5.449689683556195.'
Multi DataFrame Example
This next part shows how the agent can interact with multiple dataframes passed in as a list.
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?")
[1m> Entering new AgentExecutor chain...[0m
[32;1m[1;3mThought: I need to compare the age columns in both dataframes
Action: python_repl_ast
Action Input: len(df1[df1['Age'] != df2['Age']])[0m
Observation: [36;1m[1;3m177[0m
Thought:[32;1m[1;3m I now know the final answer
Final Answer: 177 rows in the age column are different.[0m
[1m> Finished chain.[0m
'177 rows in the age column are different.'
Related
- Toolconceptual guide
- Toolhow-to guides