How to do value_counts in all values in Pandas Dataframe

Question

For examples, I get a dataframe as follows:

df

     col1 col2 col3
row1 1    2    3    
row2 1    0    0    
row3 1    0    0    

do_value_counts(df)
   value_count
0  4
1  3
2  1
3  1

I try to reshape dataframe into (n,1) but it is not easy for Pandas, so I try to initialize a numpy array with it and reshape it into (n, 1).
Then I use np.bincount() to do the things like pd.value_counts() .
BUT I get (31280, 1) shape numpy array and object too deep for desired array when I try to use np.bincount on it.
So I am wondering if there is a better way to implement it?
If not mind could anyone help me?
Thanks in advances.

Answer 1

try to melt the dataframe before:

df.melt().value.value_counts()
# 0     4
# 1     3
# 2     1
# 3     1

a numpy way would be :

unique, counts = np.unique(df.values.ravel(), return_counts=True)
counts_df = pd.DataFrame({"counts":counts}, index=unique) 
#    counts
# 0     4
# 1     3
# 2     1
# 3     1

Answer 2

User this, maybe it will solve what you need.

item, count = np.unique(np.array(df).flatten(), return_counts=True)
pd.DataFrame(count, index=item, columns=['value_counts'])

value_counts
0   4
1   3
2   1
3   1