sciPy stats.nanmean() function | Python

# Arithmetic Mean 
import scipy
import numpy as np

arr1 = [1, 3, np.nan, 27] 
 
print("Arithmetic Mean using nanmean :", scipy.nanmean(arr1))

print("Arithmetic Mean without handling nan value :", scipy.mean(arr1)) 

# Arithmetic Mean 
from scipy import mean
from scipy import nanmean
import numpy as np

arr1 = [[1, 3, 27], 
        [3, np.nan, 6], 
        [np.nan, 6, 3], 
        [3, 6, np.nan]] 
 
print("Arithmetic Mean is :", mean(arr1)) 
print("Arithmetic Mean handling nan :", nanmean(arr1)) 

# using axis = 0
print("\nArithmetic Mean is with default axis = 0 : \n", 
      mean(arr1, axis = 0))
print("\nArithmetic Mean handling nan with default axis = 0 : \n", 
      nanmean(arr1, axis = 0))

# using axis = 1
print("\nArithmetic Mean is with default axis = 1 : \n", 
      mean(arr1, axis = 1))  
print("\nArithmetic Mean handling nan with default axis = 1 : \n", 
      nanmean(arr1, axis = 1))