preproccessing image data from csv file using pandas

Asked Jan 15 '18 at 08:26

Active Jan 15 '18 at 08:26

Viewed 145 times

In my dataframe I have one pixel column which has row vector of images in each row. like this:

              pixels
        0     70 66 34 56 23 45 8 5 34... 
        1     79 63 54 96 83 57 8 5 34...

I want to make column of each element in row vector. like this:

              0  1  2  3  4  5  6  7  8
        0     70 66 34 56 23 45 81 52 34... 
        1     79 63 54 96 83 57 81 53 34...

asked Jan 15 '18 at 08:26

008karan

it seems you need `df.pixels.str.split()` – jezrael Jan 15 '18 at 08:27
it is dupe :( I think – jezrael Jan 15 '18 at 08:28
it giving error:'DataFrame' object has no attribute 'pixels' – 008karan Jan 15 '18 at 08:33
What is `print (df.columns.tolist())` ? – jezrael Jan 15 '18 at 08:34
[0,1,2,3,4,5,.....199] it is index i think – 008karan Jan 15 '18 at 08:35
So there is no `pixels` column? – jezrael Jan 15 '18 at 08:36
giving ..['pixels'] – 008karan Jan 15 '18 at 08:42
So maybe need `df1 = df['pixels'].str.split()` – jezrael Jan 15 '18 at 08:42
gives: KeyError: 'pixels' – 008karan Jan 15 '18 at 08:59
OK, what is `print (df.head().to_dict())` of first DataFrame? – jezrael Jan 15 '18 at 09:00
after some change: df1 = df['pixels'].str.split() like this 0 [70, 80, 82, 72, 58, 58, 60, 63, 54, 58, 60, 4... 1 [151, 150, 147, 155, 148, 133, 111, 140, 170, ... 2 [231, 212, 156, 164, 174, 138, 161, 173, 182, ... – 008karan Jan 15 '18 at 09:14

0 Answers0