Multi-Index Sorting in Pandas

Question

I have a multi-index DataFrame created via a groupby operation. I\'m trying to do a compound sort using several levels of the index, but I can\'t seem to find a sort functi

Answer 1

If you want try to avoid multiple swaps within a very deep MultiIndex, you also could try with this:

1. Slicing by level X (by list comprehension + .loc + IndexSlice)
2. Sort the desired level (sortlevel(2))
3. Concatenate every group of level X indexes

Here you have the code:

import pandas as pd
idx = pd.IndexSlice
g = pd.concat([grouped.loc[idx[i,:,:],:].sortlevel(2) for i in grouped.index.levels[0]])
g