Support operation on multiple data frames, for example concat, merge, join, append, compare, etc

[shuangquanhuang]

🚀 The feature

Support operation on multiple data frames, for example concat, merge, join, append, compare, etc

Motivation, pitch

Support operation on multiple data frames, for example concat, merge, join, append, compare, etc

Alternatives

No response

Additional context

No response

[gventuri]

@shuangquanhuang super nice addition, will work on it!

[ambujpawar]

Hey, @gventuri I can work on it. However, before starting with the issue. I have a few questions:

• How do you propose to handle this? Because currently the user queries like pandas_ai.run(df, prompt='Which are the 5 happiest countries?')
• Should it be handle datafranes like pandas_ai.run(df1, df2, prompt='Merge the two databases?')

Or do you have something else in mind?

[gventuri]

@ambujpawar maybe we could also accept a list in the first param and then conditionally run the workflow on only 1 df vs multiple dfs.

Something like this:

pandas_ai.run(df1, prompt='Which are the 5 happiest countries?')
pandas_ai.run([df1, df2], prompt='Which are the 5 happiest countries?')

What do you think?

[ambujpawar]

I dont have a strong preference for any. But out of the these two options, I prefer:

pandas_ai.run(df1, df2, prompt='Which are the 5 happiest countries?')

Because to be it seems to be more "easy" for the user as well as more intuitive. So shall I go ahead with that?

[mzy2240]

The current version does not understand groupby-related operation, which might also be good to have. For example, if you add a continent column to the example and ask which countries are happiest in each continent, it will give random answer.

[HenriqueAJNB]

This groupby tasks should be another issue. Can you create that @mzy2240?

[gventuri]

@mzy2240 @HenriqueAJNB I've created one (#127 )

[gventuri]

Feature added with this commit: be4c717

Closing