Say I have the following dataframe:
a=np.array([[10,20,30,40],[5,20,35,50],[0,5,10,15]])
b=pd.DataFrame(a, columns=["n1", "n2", "n3", "n4"])
I do the following calculation on it (difference between 1st and 2nd row):
c=b.diff(1,0)
c.loc[[1]]
And then I want to insert this row of differences (given by c.loc[[1]]) into my dataframe b. How do I do that? When I try, I get an error ValueError: cannot set a row with mismatched columns"
Use setting with enlargement by Series selected by loc with one []:
b.loc[len(b)] = b.diff().loc[1]
print (b)
n1 n2 n3 n4
0 10.0 20.0 30.0 40.0
1 5.0 20.0 35.0 50.0
2 0.0 5.0 10.0 15.0
3 -5.0 0.0 5.0 10.0
Or use concat with DataFrame with [[]]:
c = pd.concat([b, b.diff().loc[[1]]], ignore_index=True)
print (c)
n1 n2 n3 n4
0 10.0 20.0 30.0 40.0
1 5.0 20.0 35.0 50.0
2 0.0 5.0 10.0 15.0
3 -5.0 0.0 5.0 10.0
Try
c = b.diff(axis=0).iloc[1]
or
c = b.diff(axis=0).loc[1]
The output:
print(c)
n1 -5.0
n2 0.0
n3 5.0
n4 10.0
Name: 1, dtype: float64
to add c to b use append
b = b.append(c)
output of print(b)
n1 n2 n3 n4
0 10.0 20.0 30.0 40.0
1 5.0 20.0 35.0 50.0
2 0.0 5.0 10.0 15.0
1 -5.0 0.0 5.0 10.0
