Release v3 is currently in beta. This documentation reflects the features and functionality in progress and may change before the final release.

​
Chat

The .chat() method is PandasAI’s core feature that enables natural language interaction with your data. It allows you to:
  • Query your data using plain English
  • Generate visualizations and statistical analyses
  • Work with multiple DataFrames simultaneously
For a more UI-based data analysis experience, check out our Data Platform.

​
Basic Usage

import pandasai as pai

df_customers = pai.load("company/customers")

response = df_customers.chat("Which are our top 5 customers?")

​
Chat with multiple DataFrames

import pandasai as pai

df_customers = pai.load("company/customers")
df_orders = pai.load("company/orders")
df_products = pai.load("company/products")

response = pai.chat('Who are our top 5 customers and what products do they buy most frequently?', df_customers, df_orders, df_products)

​
Available Output Formats

PandasAI supports multiple output formats for responses, each designed to handle different types of data and analysis results effectively. This document outlines the available output formats and their use cases.

​
DataFrame Response

Used when the result is a pandas DataFrame. This format preserves the tabular structure of your data and allows for further data manipulation.

​
Chart Response

Handles visualization outputs, supporting various types of charts and plots generated during data analysis.

​
String Response

Returns textual responses, explanations, and insights about your data in a readable format.

​
Number Response

Specialized format for numerical outputs, typically used for calculations, statistics, and metrics.

​
Error Response

Provides structured error information when something goes wrong during the analysis process.

​
Usage

The response format is automatically determined based on the type of analysis performed and the nature of the output. You don’t need to explicitly specify the format - PandasAI will choose the most appropriate one for your results. Example: 
import pandasai as pai

df = pai.load("my-org/users")

response = df.chat("Who is the user with the highest age?") # Returns a String response
response = df.chat("How many users in total?") # Returns a Number response
response = df.chat("Show me the data") # Returns a DataFrame response
response = df.chat("Plot the distribution") # Returns a Chart response

​
Response Types Details

Each response type is designed to handle specific use cases:
  • String Response: Provides textual analysis and explanations
  • Number Response: Returns numerical results from calculations
  • DataFrame Response: Preserves the structure and functionality of pandas DataFrames
  • Chart Response: Handles various visualization formats and plotting libraries
  • Error Response: Structured error handling with informative messages
The response system is extensible and type-safe, ensuring that outputs are properly formatted and handled according to their specific requirements.

​
Response Object Methods

The response object provides several useful methods and properties to interact with the results:

​
Value Property

By default, when you print a response object, it automatically returns its .value property: 
response = df.chat("What is the average age?")
print(response)  # Automatically calls response.value
# Output: The average age is 34.5 years

# For charts, printing will display the visualization
chart_response = df.chat("Plot age distribution")
print(chart_response)  # Displays the chart

​
Generated Code

You can inspect the code that was generated to produce the result: 
response = df.chat("Calculate the correlation between age and salary")
print(response.last_code_executed)
# Output: df['age'].corr(df['salary'])

​
Saving Charts

For chart responses, you can save the visualization to a file: 
chart_response = df.chat("Create a scatter plot of age vs salary")
chart_response.save("scatter_plot.png")  # Saves the chart as PNG