Python | Numpy Nanmedian () function



numpy.nanmedian() can be used to calculate the median of an array, ignoring the NaN value. If the array is NaN and we can determine the median without being influenced by the NaN value. Let`s see examples of different types about the numpu.nanmedian () method.

Syntax: numpy.nanmedian (a, axis = None, out = None, overwrite_input = False, keepdims =)

Parametrs:
a: [arr_like] input array
axis: we can use axis = 1 means row wise or axis = 0 means column wise.
out: output array
overwrite_input: If True, then allow use of memory of input array a for calculations. The input array will be modified by the call to median.
keepdims: If this is set to True, the axes which are reduced are left in the result as dimensions with size one. With this option, the result will broadcast correctly against the original a.

Returns: It return median in ndarray.

Example # 1:

# Python code for demonstration
# using numpy.nanmedian

import numpy as np

 
# create a 2d array with nan value

arr = np.array ([[ 12 , 10 , 34 ], [ 45 , 23 < code class = "plain">, np.nan]])

 

print ( " Shape of array is " , arr. shape)

 

print ( "Median of array without using nanmedian function:" ,

np.median (arr))

 

print ( "Using nanmedian function:" , np.nanmedian (arr))

Output :

 Shape of array is (2, 3) Median of array without using nanmedian function: nan Using nanmedian function: 23.0  

Example # 2:

 

# Python code for demonstration
# using numpy.nanmedian
# with axis

import numpy as np

 
# create a 2d array with nan value.

arr = np.array ([[ 12 , 10 , 34 ], [ 45 , 23 , np.nan]])

 

print ( "Shape of array is" , arr.shape)

 

print ( " Median of array with axis = 0: " ,

  np.median (arr, axis = 0 ))

 

print ( " Using nanmedian function: " ,

  np.nanmedian (arr, axis = 0 ))

Exit :

 Shape of array is (2, 3) Median of array with axis = 0: [28.5 16.5 nan] Using nanmedian function: [28.5 16.5 34.] 

Example # 3:

# Python code for demonstration
# using numpy.nanmedian
# with axis = 1

import numpy as np

 
# create a 2d matrix with nan value

arr = np.array ([[ 12 , 10 , 34 ],

[ 45 , 23 , np.nan], 

[ 7 , 8 , np.nan]])

 

print ( "Shape of array is" , arr.shape)

 

print ( "Median of array with axis = 0:" ,

np.median (arr, axis = 1 ))

 

print ( "Using nanme dian function: " ,

  np .nanmedian (arr, axis = 1 ))

Exit:

 Shape of array is (3, 3) Median of array with axis = 0: [12. nan nan] Using nanmedian function: [12. 34. 7.5]