为 pandas dataframe 中的序数创建虚拟变量

Question

I am trying to create dummy variables in python in the pandas dataframe format. I have a variable called "Weight Group" and I want to transform the variables like so:

Before transformation:

    Weight_Group
0      1
1      5
2      4
3      2
4      2
5      3
6      1

After transformation:

    WD_1  WD_2  WD_3  WD_4  WD_5
0      1    0     0     0     0
1      1    1     1     1     1
2      1    1     1     1     0 
3      1    1     0     0     0
4      1    1     0     0     0
5      1    1     1     0     0
6      1    0     0     0     0

I know that pandas has the get_dummies() function that creates dummy variables, but it doesn't give me the functionality that I want, where someone in weight group 3 has ones in the WG_1, WG_2, and WG_3 columns. I have a lot of data points so a fast method would be great. If anyone has any ideas on how I can implement this I would really appreciate it!

Answer 1

You can call pd.get_dummies() and then replace your 0 tallies with NaN and use bfill() (plus a bit of extra cleanup for display):

pd.get_dummies(df['Weight_Group'], prefix='WD').replace(0,np.nan).bfill(axis=1).fillna(0).astype(int)

Yields:

   WD_1  WD_2  WD_3  WD_4  WD_5
0     1     0     0     0     0
1     1     1     1     1     1
2     1     1     1     1     0
3     1     1     0     0     0
4     1     1     0     0     0
5     1     1     1     0     0
6     1     0     0     0     0