Add column to pandas Dataframe (ascending block of numbers)

Jailbone :

I have a data looking like this:

Col1

aaa1    
bbb1 
ccc1
1
2    
3
aaa2
bbb2
ccc2
4
5
6

I want to add a new column to make it look like this:

Col1     Col2

aaa1    1
bbb1    1
ccc1    1
1       1
2       1   
3       1
aaa2    2
bbb2    2
ccc2    2
4       2
5       2
6       2

So every time a row with aaa1, aaa2... is reached the number in the row of the new column should increase by one. Furthermore I want then to delete all rows of the form aaa, bbb and ccc so the result looks like this:

Col1     Col2

1       1
2       1   
3       1
4       2
5       2
6       2

I appreciate your help!

anky_91 :

IIUC, you can check if str.contains aaa and then cumsum for Col2 , then check using pd.to_numeric if value is numeric in Col1 and remove which is not:

df['Col2'] = df['Col1'].str.contains('aaa').cumsum()
out = df[pd.to_numeric(df['Col1'],errors='coerce').notna()]

---

print(out)

   Col1  Col2
3     1     1
4     2     1
5     3     1
9     4     2
10    5     2
11    6     2