How can a function be structured to make adjustments to a pandas dataframe?

Question

Anonymous · Accepted Answer

# Pandas challenge
## Statement:
How would you author a function to perform operations on a pandas dataframe?
## Requirements
* How would I know which ways I am to modify the dataframe?
It will be passed as parameter, in natural language.
## what is it to return?
The modified pandas
## Pseudo-code
```python
def modypandas(df: DataFrame, change: string) -> DataFrame
# perform the change
return modified_df
```
## Design Options:
### Agentic
Two agents:
- one to convert the code
- one to execute
Pros:
- Handles well many cases
- Can work with several models
Cons:
- Depends on agentic libraries
- Harder to validate
### Non agentic
Two steps
- One call to convert the request to code
- One to perform the change
Pros:
- Handles well many cases
- Uses standard libraries
- Can work with several models
Cons:
- Hard to validate
### Traditional
Two steps:
- Extract the meaning via intent classifier,
- One to perform the identified task
Pros:
- no dependency on LLMs
- Deterministic
Cons:
- Harder to code
- several edge cases, limited number of options

How can a function be structured to make adjustments to a pandas dataframe?

Practice this question with AI

Go Premium

Practice More Questions

Community Answers

Unlock Community Insights