熊猫：增量计算列中的出现次数

Question

I have a DataFrame (df) which contains a 'Name' column. In a column labeled 'Occ_Number' I would like to keep a running tally on the number of appearances of each value in 'Name'.

For example:

Name            Occ_Number
 abc                     1
 def                     1
 ghi                     1
 abc                     2
 abc                     3
 def                     2
 jkl                     1
 jkl                     2

I've been trying to come up with a method using

>df['Name'].value_counts()

but can't quite figure out how to tie it all together. I can only get the grand total from value_counts(). My process thus far involves creating a list of the 'Name' column string values which contain counts greater than 1 with the following code:

>things = df['Name'].value_counts()
>things = things[things > 1]
>queries = things.index.values

I was hoping to then somehow cycle through 'Name' and conditionally add to Occ_Number by checking against queries, but this is where I'm getting stuck. Does anybody know of a way to do this? I would appreciate any help. Thank you!

Answer 1

You can use cumcount to avoid a dummy column:

>>> df["Occ_Number"] = df.groupby("Name").cumcount()+1
>>> df
  Name  Occ_Number
0  abc           1
1  def           1
2  ghi           1
3  abc           2
4  abc           3
5  def           2
6  jkl           1
7  jkl           2

Answer 2

You can add a helper column and then use cumsum :

df =pd.DataFrame({'Name':['abc', 'def', 'ghi', 'abc', 'abc', 'def', 'jkl', 'jkl']})

add count:

df['counts'] =1

group by name:

cs =df.groupby('Name')['counts'].cumsum()
# set series name
cs.name = 'Occ_number'

join series back to dataframe:

# remove helper column
del df['counts']
df.join(cs)

returns:

    Name    Occ_number
 0  abc     1
 1  def     1
 2  ghi     1
 3  abc     2
 4  abc     3
 5  def     2
 6  jkl     1
 7  jkl     2