Inserting a row into a pandas dataframe based on row value?

Question

I have a DataFrame:

df = pd.DataFrame({'B':[2,1,2],'C':['a','b','a']})
  B C
0 2 'a'
1 1 'b'
2 2 'a'

I want to insert a row above any occurrence of 'b', that is a duplicate of that row but with 'b' changed to 'c', so I end up with this:

  B C
0 2 'a'
1 1 'b'
1 1 'c'
2 2 'a'

For the life of me, I can't figure out how to do this.

you said above but in output its below, your first df produces a instead of c in third row. — shivsn
– shivsn, Commented Aug 31, 2016 at 17:09
@Divakar, that's unlikely to happen, but if it did, then I would just insert a 'c' row above each one of the 'b' rows — BobbyJohnsonOG
– BobbyJohnsonOG, Commented Aug 31, 2016 at 17:18

Ciarán Tobin · Accepted Answer · 2016-08-31 17:43:06Z

5

Here's one way of doing it:

duplicates = df[df['C'] == 'b'].copy()
duplicates['C'] = 'c'
df.append(duplicates).sort_index()

answered Aug 31, 2016 at 17:43

Ciarán Tobin

7,5661 gold badge31 silver badges46 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

BobbyJohnsonOG Over a year ago

Beautiful, didn't realize there was a copy function!

Divakar · Accepted Answer · 2016-08-31 17:50:01Z

1

Working at NumPy level, here's a vectorized approach -

arr = df.values
idx = np.flatnonzero(df.C=='b')
newvals = arr[idx]
newvals[:,df.columns.get_loc("C")] = 'c'
out = np.insert(arr,idx+1,newvals,axis=0)
df_index = np.insert(np.arange(arr.shape[0]),idx+1,idx,axis=0)
df_out = pd.DataFrame(out,index=df_index)

Sample run -

In [149]: df
Out[149]: 
   B  C
0  2  a
1  1  b
2  2  d
3  4  d
4  3  b
5  8  a
6  4  a
7  2  b

In [150]: df_out
Out[150]: 
   0  1
0  2  a
1  1  b
1  1  c
2  2  d
3  4  d
4  3  b
4  3  c
5  8  a
6  4  a
7  2  b
7  2  c

edited Aug 31, 2016 at 17:50

answered Aug 31, 2016 at 17:42

Divakar

222k19 gold badges273 silver badges374 bronze badges

Collectives™ on Stack Overflow

Inserting a row into a pandas dataframe based on row value?

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related