How to connect Connectors with Agents

[vsneelesh]

I would like to know if it is possible to connect pandasAI Agents with connectors. If yes how to do it.
I would like to add the field description of dataset using connectors and also use the properties of agents. The docs only show the integration with smartdataframe/lake

[dosubot[bot]]

To connect pandasAI Agents with connectors and add the field description of a dataset while using the properties of agents, you can follow the example below. The Agent class constructor allows you to pass a description and connectors. Here is how you can do it:

import pandas as pd
from pandasai import Agent
from pandasai.connectors import PandasConnector

# Example DataFrame
data = {
    'Name': ['Alice', 'Bob', 'Charlie'],
    'Age': [25, 30, 35],
    'City': ['New York', 'Los Angeles', 'Chicago']
}
df = pd.DataFrame(data)

# Create a PandasConnector
connector = PandasConnector({"original_df": df})

# Description of the dataset
description = "This dataset contains information about individuals including their name, age, and city."

# Initialize the Agent with the connector and description
agent = Agent(dfs=connector, description=description)

# Now you can use the agent's properties and methods
print(agent.description)  # Output: This dataset contains information about individuals including their name, age, and city.

In this example:

• A PandasConnector is created with the DataFrame.
• The Agent is initialized with the connector and a description.
• The description property of the Agent instance holds the description of the dataset.

Additionally, if you are using a different connector like DatabricksConnector, you can follow a similar approach:

import os
from pandasai import Agent
from pandasai.ee.connectors import DatabricksConnector

databricks_connector = DatabricksConnector(
    config={
        "host": "adb-*****.azuredatabricks.net",
        "database": "default",
        "token": "dapidfd412321",
        "port": 443,
        "table": "loan_payments_data",
        "httpPath": "/sql/1.0/warehouses/213421312",
        "where": [
            ["loan_status", "=", "PAIDOFF"],
        ],
    }
)

os.environ["PANDASAI_API_KEY"] = "your-api-key"

# Description of the dataset
description = "This dataset contains loan payment information including loan status and payment details."

agent = Agent(databricks_connector, description=description)

response = agent.chat("How many people from the United States?")
print(response)

In this example:

• A DatabricksConnector is created with the configuration details.
• The Agent is initialized with the connector and a description.
• The description property of the Agent instance holds the description of the dataset ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[vsneelesh]

Thanks. Got resolved