Problem in NumPy array while appending values

Question

Code

import numpy as np

arr3 = np.array([], dtype=float)
val=1
for i in range(7):
    if i == 0 or i == 3:
        np.append(arr3,np.nan)
    else:       
        np.append(arr3,val)
        val+=1
arr3

Output

array([], dtype=float64)

When I run this code it does not provide any output. I don't know the reason.

Why didn't you use list append? np.append is a poor imitation of something works better. — hpaulj
– hpaulj, Commented Apr 27, 2019 at 15:55
i created the array using numpy package so i thought that i should use one of those functions provided in numpy for managing the array. — jarvis
– jarvis, Commented Apr 28, 2019 at 15:51

Daweo · Accepted Answer · 2019-04-27 12:51:42Z

1

Note that as numpy.append docs says append does not occur in-place, that is numpy.append returns new array. Consider following example:

import numpy as np
arr = np.array([0],dtype='uint8')
arr2 = np.append(arr,1)
print(arr) #[0]
print(arr2) #[0 1]

Your code should work after altering following lines:

np.append(arr3,np.nan)

to

arr3=np.append(arr3,np.nan)

and

np.append(arr3,val)

to

arr3=np.append(arr3,val)

answered Apr 27, 2019 at 12:51

Daweo

38.2k3 gold badges17 silver badges32 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

hpaulj · Accepted Answer · 2019-04-28 16:09:49Z

List append is faster than np.append (or any of the other variations on np.concatenate). List append works in-place, adding a 'pointer' to the existing list. np.concatenate (even with the np.append cover) creates a new array each call, requiring a full copy. If you must build an array iteratively, use lists:

In [85]: alist = []                                                                  
In [86]: val=1                                                                       
In [87]: for i in range(7): 
    ...:     if i==0 or i==3: 
    ...:         alist.append(np.nan) 
    ...:     else: 
    ...:         alist.append(val) 
    ...:         val +=1 
    ...:                                                                             
In [88]: alist                                                                       
Out[88]: [nan, 1, 2, nan, 3, 4, 5]
In [89]: arr = np.array(alist)                                                       
In [90]: arr                                                                         
Out[90]: array([nan,  1.,  2., nan,  3.,  4.,  5.])

Collectives™ on Stack Overflow

Problem in NumPy array while appending values

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related