Unable to analyze the DataFrame when it contains data in list format.

[ReeveWu]

Unable to analyze the DataFrame when it contains data in list format.

[dosubot]

Hey there, @ReeveWu! I'm here to help you with any bugs, questions, or contributions you have. Let's squash those bugs together! 🤖

The issue you're encountering is due to the Agent class in pandasai not handling unhashable types like lists in DataFrame columns. The error message "unhashable type: 'list'" indicates that the Agent class is trying to process the list data type in a way that requires hashing, which is not possible for lists.

To work around this issue, you can preprocess the DataFrame to convert the list columns into a format that can be handled by the Agent class. One common approach is to convert the lists into strings or another hashable type.

Here's an example of how you can modify your DataFrame before passing it to the Agent:

import pandas as pd
import os
from pandasai import Agent, Config
from pandasai.llm import OpenAI

data = {
    'Employee_ID': [101, 102, 103, 104],
    'Employee_Name': ['Alice', 'Bob', 'Charlie', 'Diana'],
    'Projects': [['Project A', 'Project B'], ['Project C'], ['Project D', 'Project E', 'Project F'], ['Project G']],
    'Salary': [70000, 80000, 75000, 90000]
}

df = pd.DataFrame(data)

# Convert list columns to strings
df['Projects'] = df['Projects'].apply(lambda x: ', '.join(x))

agent = Agent(
    dfs=df, 
    config=Config(llm=OpenAI(api_token=os.getenv("OAI_API_KEY"), model="gpt-4o"))
)

print(agent.chat('Tell me the average salary of the employees'))

This code converts the Projects column from lists to comma-separated strings, which should be hashable and thus avoid the "unhashable type: 'list'" error.

If you need to perform more complex analysis on the list data, you might need to implement additional preprocessing steps or custom handling within your analysis logic.

To continue talking to Dosu, mention @dosu.