Stacked numpy arrays

Question

I am considerably new to Python.I am willing to stack a few 1-d arrays in python and create a 2-d array using that. The arrays which I want to stack are returned from another function. The minimum reproducible code that I have written for this purpose has been shown below:

def fun1(i): # this function returns the array
  return array # This array is a function of i
h0= np.empty(5,dtype=object)
arrays=[h0]
for i in range(4):
 arr=fun1(i)
 arrays.append(arr)

h=np.vstack(arrays)
print (h)

The desired output is of the form :

[[1 1 1 1 1]
 [2 2 2 2 2]
 [3 3 3 3 3]
 [4 4 4 4 4]]

But I get :

[[None None None None None]
 [1 1 1 1 1]
 [2 2 2 2 2]
 [3 3 3 3 3]
 [4 4 4 4 4]]

I understand that I get the above output because an empty array of dtype=object has all elements None . But I am not being able to solve the problem. Any help regarding this would be highly appreciated.

Why did you use the h0 at all?

hpaulj
– hpaulj

2020-06-01 14:01:10 +00:00
Commented Jun 1, 2020 at 14:01 — hpaulj
– hpaulj, Commented Jun 1, 2020 at 14:01

pakallis · Accepted Answer · 2020-06-01 11:59:26Z

1

You are getting None because you are creating an empty array with dtype object:

h0 = np.empty(5,dtype=object)

This line creates an array with None elements.

You can remove this line so that it works as you expect:

def fun1(i): # this function returns the array
  return array # This array is a function of i
arrays=[]
for i in range(4):
 arr=fun1(i)
 arrays.append(arr)

h = np.vstack(arrays)
print (h)

answered Jun 1, 2020 at 11:59

pakallis

1816 bronze badges

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

Comments

Hasibul Islam Polok · Accepted Answer · 2020-06-01 12:00:38Z

0

if np.where(~a.any(axis=1))[0]:
    #do nothing
else:
    #print as you like

answered Jun 1, 2020 at 12:00

Hasibul Islam Polok

1059 bronze badges

Comments

Hammad Ahmed · Accepted Answer · 2020-06-01 14:01:54Z

0

try this

h = np.vstack([
    fun1(i)
    for i in range(4)
])

print(h)

^^ this is assuming that fun1() is some proprietary function returning arrays of same length

but if you want a 2D array with equal rows entries and incrementing column-wise like

[[1 1 1 1 1]
 [2 2 2 2 2]
 [3 3 3 3 3]
 [4 4 4 4 4]]

you can also try a faster method

x = np.arange(1,5).reshape([-1,1])
h = np.repeat(x, 5, axis= 1)

answered Jun 1, 2020 at 14:01

Hammad Ahmed

8758 silver badges17 bronze badges

Collectives™ on Stack Overflow

Stacked numpy arrays

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related