This documentation page is for backwards compatibility. For new projects, we recommend using the new semantic dataframes.

​
Migrate SmartDataframe

To migrate from SmartDataframe to the new semantic dataframes:

  1. Replace SmartDataframe imports with semantic dataframe imports
  2. Update configuration to use pai.config.set()
  3. Use the new .chat() method syntax

Example migration:

# Old code
from pandasai import SmartDataframe
smart_df = SmartDataframe(df)

# New code
import pandasai as pai

df = pai.DataFrame(df)

​
SmartDataframe (Legacy)

The SmartDataframe class was the previous way to create dataframes in PandasAI. It provided similar functionality to the current semantic dataframes but with a different API.

​
Basic Usage

from pandasai import SmartDataframe
import pandas as pd

# Create a pandas DataFrame
df = pd.DataFrame({
    'name': ['John', 'Emma', 'Alex', 'Sarah'],
    'age': [28, 24, 32, 27],
    'city': ['New York', 'London', 'Paris', 'Tokyo']
})

# Convert to SmartDataframe
smart_df = SmartDataframe(df)

# Ask questions about your data
response = smart_df.chat("What is the average age?")

​
Configuration

SmartDataframe accepted configuration options through its constructor:

smart_df = SmartDataframe(df, config={
    "llm": llm,                              # LLM instance
    "save_logs": True,                       # Save conversation logs
    "verbose": False                         # Print detailed logs
})