Pandas create multiple columns based on other columns

Question

I have a huge df (720 columns) with this structure:

I would like to create a new df, based on calculations such as:

if A and B = 1 then  v1 = 1
if A and C = 1 then  v2 = 1
if A and D = 1 then  v3 = 1
if A and XX = 1 then v719 = 1

id V1 V2 
1  0  1
2  0  1 
3  1  1

As I need to iterate A vs B and C (in reality A vs 719 columns) I looking for a way to write this code without doing something like this for ALL columns

df.loc[((df['A'] == 1) & (df['B'] == 1)), 'v1'] = 1
df.loc[((df['A'] == 1) & (df['C'] == 1)), 'v2'] = 1
df.loc[((df['C'] == 1) & (df['D'] == 1)), 'v2'] = 1
df.loc[((df['A'] == 1) & (df['XX'] == 1)), 'v719'] = 1

Any ideas?

BENY · Accepted Answer · 2020-07-18 03:03:56Z

3

For your question we can do , since 1 * 1 = 1

s=df.loc[:,'B':].mul(df.A,axis=0)
   B  C
0  0  1
1  0  1
2  1  1
s.columns=np.arange(s.shape[1])+1
df=df.join(s.add_prefix('v_'))

answered Jul 18, 2020 at 3:03

BENY

324k22 gold badges176 silver badges250 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

EGM8686 Over a year ago

Maybe I've oversimplified. Actually conditions are more like: if A = 1 and B = 2 then v1 = 1 etc...

MasayoMusic Over a year ago

@EGM8686 Is it two for all columns or just B?

EGM8686 Over a year ago

Tricky part is that is for ALL columns. So in this case I end up with 3 dfs (A with B, A with C and B with C)

MasayoMusic Over a year ago

Sorry, I meant the values. B =2, C =2 or would C = 3 being it's the third column.

MasayoMusic Over a year ago

@EGM8686 The provided solution in only multiplying against A, but you want all combinations. I searched around and I think this should get you going: stackoverflow.com/questions/29229892/…

|

Collectives™ on Stack Overflow

Pandas create multiple columns based on other columns

1 Answer 1

6 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related