基于groupby一个列值和pandas中另一列的计数创建一个新列？

Question

I have a pandas dataframe

df = pd.DataFrame({'Birds': ['Falcon','Falcon','Parrot','Peacock','Peacock'],
                   'Name': ['A', 'D', 'B', 'C', 'C']})

I need to create a new column,

df = pd.DataFrame({'Birds': ['Falcon','Falcon','Parrot','Peacock','Peacock'],
                   'Name': ['A', 'D', 'B', 'C', 'C']
                   'Count':['1','1','0','0','0'] })

Falcon has two names, so each records given 1, parrot and peacock has only one name which is B for parrot and C for peacock, so new column has 0.

I tried using groupby

df.groupby(['Birds','Name']).size()

this returns

Birds    Name
Falcon   A        1
         D        1
Parrot   B        1
Peacock  C        2
dtype: int64

Not sure how to convert this

Another way, subset and drop duplicates

df2 = df.drop_duplicates(subset=['Birds', 'Name'], keep='first')
df2['Birds'].value_counts()

this returns

Falcon     2
Peacock    1
Parrot     1
Name: Birds, dtype: int64

Not sure how to use this to create new column in original as 1 and 0

Answer 1

You can use transform combined with nunique :

df["count"] =  df.groupby("Birds")["Name"].transform(lambda x: x.nunique() - 1)

Without lambda - Option 1

df["count"] =  df.groupby("Birds")["Name"].transform("nunique") - pd.Series([1] * df.shape[0])

Without lambda - Option 2

df["count"] =  df.groupby("Birds")["Name"].transform("nunique")
df["count"] =  df["count"] -1