Apply Function To Each Cell In Dataframe That Depends On The Column Name In Pandas

How can I apply function to each cell in a DataFrame that depends on the column name? I'm aware of pandas.DataFrame.applymap but it doesn't seem to allow depending on the column na

Solution 1:

I bit improved another answer, axis=0 is by default so can be omit:

a = frame.apply(lambda x: x.apply(format2,args=(x.name)))
print (a)
              b        d        e
Utah     b_1.62  d_-0.61  e_-0.53
Ohio    b_-1.07   d_0.87  e_-2.30
Texas    b_1.74  d_-0.76   e_0.32
Oregon  b_-0.25   d_1.46  e_-2.06

Solution 2:

If you don't want to loop through the columns, you can do something like this:

frame.T.apply(lambda x: x.apply(format2,args=(x.name)), axis=1).TOut[289]: 
              bdeUtahb_0.90d_-0.68e_-0.12Ohiob_-0.94d_-0.27e_0.53Texasb_-0.69d_-0.40e_-0.69Oregonb_-0.85d_-0.67e_-0.01

After transposing the df, the column names become index which can be referenced in apply function by using the .name attribute.

Solution 3:

why not:

>>>frame
               b         d         e
Utah   -0.579869  0.101039 -0.225319
Ohio   -1.791191 -0.026241 -0.531509
Texas   0.785618 -1.422460 -0.740677
Oregon  1.302074  0.241523  0.860346

>>>frame['e'] = ['%.2f' % val for val in frame['e'].values]>>>frame
               b         d      e
Utah   -0.579869  0.101039  -0.23
Ohio   -1.791191 -0.026241  -0.53
Texas   0.785618 -1.422460  -0.74
Oregon  1.302074  0.241523   0.86