CODEWITHSUNDEEP
pythonpandas
python noob
python noob

Reputation: 87

Update column values - Pandas

Sample dataset:

    id  count
0   3   NaN
1   3   cat
2   3   human
3   3   bird
4   3   NaN
5   8   pegion
6   8   NaN
7   8   NaN
8   8   rat
9   8   NaN
10  8   human
11  8   monkey

Loop through each row. In count column, value of NaN and human is 0. All other strings value is 1. Then sum with previous row value until id contains same value. (continue recursively)

The output I want:


    id  count
0   3   0
1   3   1
2   3   1
3   3   2
4   3   2
5   8   1
6   8   1
7   8   1
8   8   2
9   8   2
10  8   2
11  8   3

Upvotes: 0

Views: 56

Answers (1)

Cimbali
Cimbali

Reputation: 11395

This can be done in 3 simple steps:

  • find which rows are non-human and non-NaN
  • group by id
  • compute the cumulative sum in each group
>>> (df['count'].ne('human') & df['count'].notna()).groupby(df['id']).cumsum()
0     0
1     1
2     1
3     2
4     2
5     1
6     1
7     1
8     2
9     2
10    2
11    3

Upvotes: 1

Related Questions