I have a DataFrame by pandas, and it contains a lots of NaN values.

the following figure is about data what I have,

and next one is what I expected.

I have tried to fill NaN values with average(mean) and most frequents, but it is not what i want to.

Is there any package or method to fill the values with scaled for this?

one thing I would like to comment is, I do NOT want to drop this values.

if any solution, please let me know. thanks.

EDITED:

I found

After interpolate values with SalePrice

I uploaded code and sample data.

you can see sample data and its result from: https://gist.github.com/joonas-yoon/f5d01db4470ff87e442dc01c99f04c47#file-sample-txt

Thanks for all of comments and replies.