How to delete specific number of random rows in Pandas dataframe based on condition?

Question

I want to delete specific 'n' number of rows from a dataframe, where the rows to be deleted are chosen randomly. Also, it must select the rows based on a condition on particular column values.

For example, my dataframe is as below:

C1    C2    C3
1     0     a
2     1     b
3     0     c
4     0     d
5     0     e
6     1     f
7     1     g
8     1     h
9     0     i

Now, I want to remove n=2 rows randomly, that has a condition where C2==1.

The resultant frame can be as below:

C1    C2    C3
1     0     a
3     0     c
4     0     d
5     0     e
6     1     f
8     1     h
9     0     i

or

C1    C2    C3
1     0     a
2     1     b
3     0     c
4     0     d
5     0     e
7     1     g
9     0     i

or maybe other possibles too. The question here dows shows to remove 'n' sentences randomly, but it doesn't include providding the condition.

Answer 1

Filter rows by boolean indexing with DataFrame.sample for random rows, last use drop:

N = 2
df1 = df.drop(df[df['C2'].eq(1)].sample(N).index)
print (df1)
   C1  C2 C3
0   1   0  a
1   2   1  b
2   3   0  c
3   4   0  d
4   5   0  e
6   7   1  g
8   9   0  i

Or use np.random.choice for random index values:

df = df.drop(np.random.choice(df.index[df['C2'].eq(1)], N))