How to rank only duplicated rows and without Nan?

I have a table with data:

   Col1
0   1.0
1   1.0
2   1.0
3   2.0
4   3.0
5   4.0
6   NaN

How can I rank only duplicated values (without taking into account NaN as well)?
My current output is where unfortunately unique values are ranked as well:

   Col1   Rn
0   1.0  1.0
1   1.0  2.0
2   1.0  3.0
3   2.0  1.0
4   3.0  1.0
5   4.0  1.0
6   NaN  NaN

The output I need is:

   Col1   Rn
0   1.0  1.0
1   1.0  2.0
2   1.0  3.0
3   2.0  NaN
4   3.0  NaN
5   4.0  NaN
6   NaN  NaN

Example of the code:

import numpy as np
import pandas as pd

df = pd.DataFrame([[1],
                   [1],
                   [1],
                   [2],
                   [3],
                   [4],
                   [np.NaN]], columns=['Col1'])
print(df)


# Adding row_number for each pair:
df['Rn'] = df[df['Col1'].notnull()].groupby('Col1')['Col1'].rank(method="first", ascending=True)
print(df)

# I managed to select only necessary rows for mask, but how can I apply it along with groupby?:
m = df.dropna().loc[df['Col1'].duplicated(keep=False)]
print(m)

Thank you!

>Solution :

Try:

m = df['Col1'].duplicated(keep=False)
df['Rn'] = df[m].groupby('Col1')['Col1'].rank(method="first", ascending=True)
print(df)

Prints:

   Col1   Rn
0   1.0  1.0
1   1.0  2.0
2   1.0  3.0
3   2.0  NaN
4   3.0  NaN
5   4.0  NaN
6   NaN  NaN