[FEA] DataFrame diff

[beckernick]

For pandas compatibility, it would be nice to implement diff for DataFrames. This is currently implemented with a Numba kernel for Series and with the use of shift for Groupby.

The DataFrame.diff method calculates the difference of a Dataframe element compared with another element in the Dataframe, treating each column independently.

We should be able to remove the Numba implementation and instead rely on shift now that we have this feature.

import pandas as pd
​
df = pd.DataFrame({'a': [1, 2, 3, 4, 5, 6],
                   'b': [1, 1, 2, 3, 5, 8],
                   'c': [1, 4, 9, 16, 25, 36]})
​
print(df, "\n")
print(df.diff(), "\n")
print(df.diff(periods=2))
   a  b   c
0  1  1   1
1  2  1   4
2  3  2   9
3  4  3  16
4  5  5  25
5  6  8  36 

     a    b     c
0  NaN  NaN   NaN
1  1.0  0.0   3.0
2  1.0  1.0   5.0
3  1.0  1.0   7.0
4  1.0  2.0   9.0
5  1.0  3.0  11.0 

     a    b     c
0  NaN  NaN   NaN
1  NaN  NaN   NaN
2  2.0  1.0   8.0
3  2.0  2.0  12.0
4  2.0  3.0  16.0
5  2.0  5.0  20.0