pandas dataframe: compare value in one column with previous value

Question

I have a pandas dataframe in which I want to add a column (col_new), which values depend on a comparison of values in a existing column (col_exist).

Existing column (type=objects) contains As and Bs. New column should count, starting with 1. If an A follows an A, the count should rise by one. If an A follows a B, the count should rise by one. If a B follows an A, the count should not rise. If a B follows a B, the count should not rise.

col_exist  col_new
    A       1
    A       2
    A       3
    B       3
    A       4
    B       4
    B       4
    A       5
    B       5

I am completely new to programming, so thank you in advance for your adequade answer.

Answer 1

Use eq and cumsum:

df['col_new'] = df['col_exist'].eq('A').cumsum()

output:

  col_exist  col_new
0         A        1
1         A        2
2         A        3
3         B        3
4         A        4
5         B        4
6         B        4
7         A        5
8         B        5