Get the median value of n columns on a Data Frame

Question

I have the following data frame:

From which a want to get the average row-wise of non zero columns.

E.G. for row 0: (1303 + 1316 + 1322 + 1315)/4
row 2: (1632 + 1628 + 1609)/3

Answer 1

Use sum twice - all values by sum of Trues processes like 1:

df = df.sum(axis=1).div(df.ne(0).sum(1))

Timings:

np.random.seed(1997)
df = pd.DataFrame(np.random.randint(3, size=(1000,1000)))
#print (df)


In [60]: %timeit (df.replace(0,np.nan).mean(1))
1 loop, best of 3: 188 ms per loop

In [61]: %timeit (df.sum(axis=1).div(df.ne(0).sum(1)))
10 loops, best of 3: 21.8 ms per loop

Answer 2

Using replace , from 0 to np.nan

df.replace(0,np.nan).mean(1)