Pandas apply style on groupby hierarchy

Question

Pandas style keyword only works on rows or columns. Is it possible to split it according to sublevels.

For example

np.random.seed(24)
df = pd.DataFrame({'Types': np.linspace(1, 10, 10)})
df = pd.concat([df, pd.DataFrame(np.random.randn(10, 4), columns=['Names','V1','V2','V3'])],
               axis=1)

df['Types'][0:7] ="Dang"

df['Types'][7:] ="Fang"

df['Names'][0:3] ="Andy"

df['Names'][3:8] ="Flower"

df['Names'][8:] ="Avril"

df2 = pd.groupby(df,['Types','Names']).mean()
df2

Now I want to highlight the max value according to sublevel

def highlight_max(x):
     return ['background-color: yellow' if v == x.max() else ''
                for v in x]

df2.style.apply(highlight_max,axis=0,subset=['V1'])

It will highlight in this case max value in column 'V1'. I want basically two max values as per the group levels. So I want to highlighted values. Is there any simple way to do that? None of the intro material covers this basic feature.

jezrael · Accepted Answer · 2018-04-07 19:11:30Z

4

You need return DataFrame of colors for set styles. So need create new df with same index and columns with default values - here empty string and then change values by condition created by GroupBy.transform and compare by eq (==):

def highlight_max(x): 
   c1 = 'background-color: yellow'
   c2 = '' 
   m = x.groupby(level=0)['V1'].transform('max').eq(x['V1'])

   df1 = pd.DataFrame(c2, index=x.index, columns=x.columns)
   df1.loc[m, 'V1'] = c1
   return df1

df2.style.apply(highlight_max,axis=None)

edited Apr 7, 2018 at 19:11

answered Apr 7, 2018 at 19:05

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

Collectives™ on Stack Overflow

Pandas apply style on groupby hierarchy

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related