slicing pandas dataframe with an array of indices

Question

I have a pandas dataframe like this..

df = pd.DataFrame({'A' : [5,6,3,4,4,5,6,7,12,13], 'B' : 
     [1,2,3,5,5,6,7,8,9,10,]})

df

    A   B
0   5   1
1   6   2
2   3   3
3   4   5
4   4   5
5   5   6
6   6   7  
7   7   8
8  12   9
9  13  10

and I have an array of indices

array = np.array([0,1,2,4,7,8])

Now I can subset the dataframe with the array indices like this

df.iloc[array]

Which gives me a dataframe with indices present in the array.

    A  B
0   5  1
1   6  2
2   3  3
4   4  5
7   7  8
8  12  9

Now I want all the rows which are not present in the array index, row index which i want is [3,5,6,9] I am trying to do something like this but it gives me an error.

df.iloc[~loc]

Answer 1

You can use isin with inverting a boolean Series by ~:

import pandas as pd
import numpy as np

df = pd.DataFrame({'A' : [5,6,3,4,4,5,6,7,12,13], 'B' : 
     [1,2,3,5,5,6,7,8,9,10,]})

print df
    A   B
0   5   1
1   6   2
2   3   3
3   4   5
4   4   5
5   5   6
6   6   7
7   7   8
8  12   9
9  13  10

array = np.array([0,1,2,4,7,8])
print array
[0 1 2 4 7 8]

print df.index.isin(array)
[ True  True  True False  True False False  True  True False]

print ~df.index.isin(array)
[False False False  True False  True  True False False  True]

print df[ ~df.index.isin(array)]
    A   B
3   4   5
5   5   6
6   6   7
9  13  10

Answer 2

Use sets:

In [7]: wanted = [0, 1, 2, 4, 7, 8]

In [8]: not_wanted = set(df.index) - set(wanted)

In [9]: not_wanted
Out[9]: {3, 5, 6, 9}

In [11]: not_wanted = list(not_wanted)

In [12]: not_wanted
Out[12]: [9, 3, 5, 6]

In [13]: df.iloc[not_wanted]
Out[13]: 
    A   B
9  13  10
3   4   5
5   5   6
6   6   7