stack it first and then use value_counts:
import pandas as pd
from StringIO import StringIO
data = \
"""A,B,D,E,None
B,C,E,None,None
A,B,D,E,None
A,B,C,E,None
A,B,C,D,E"""
# Creation of the dataframe
df = pd.read_csv(StringIO(data),sep = ',',header=None)
# Perform the magic operation (stack and value_counts) on the dataframe
res = df.stack().value_counts()
The result looks like:
E 5
B 5
None 5
A 4
C 3
D 3
dtype: int64
This question has already been asked on StackOverflow