ENH: Ability to use "normalize" function (df['col'].value_counts(normalize=True) when you do df.groupby(cols).size()

[bbennett36]

groupby.size() should have the ability to "normalize" the results and return them as a percentage.

df['col'].value_counts(normalize=True) 
  
A = 0.25
B = 0.25  
C = 0.25  
D = 0.25

To accomplish this result with a groupby.size() you have to do the following -

df2 = df.groupby(['subset_product', 'subset_close']).size().reset_index(name='prod_count')
a = df2.groupby('subset_product')['prod_count'].transform('sum')
df2['prod_count'] = df2['prod_count'].div(a)

Is there a way to add a feature to include this as a parameter in .size()

[rhshadrach]

Please include a full reproducible example.

df.groupby(...).size() / df.shape[0] gives you the normalized counts.

[MarcoGorelli]

Hi @bbennett36 - does df.groupby(...).size() / df.shape[0] work for you? If so, I think this can be closed, I don't think it's worth adding an extra parameter in size for this

[bbennett36]

@MarcoGorelli That gives the percentage of the entire dataframe. I was looking for a percentage within the group.

I'll try to get a reproducible example on here today.

[MarcoGorelli]

Hi @bbennett36 - closing for now to reduce the queue, will reopen if you post a reproducible example