Numpy Concatenate Replacing Previous Data in Array

Question

I am trying to write code that produces a deck of cards. The deck is a 2D array that contains each card as an array. Each card array contains its card value as well as its suit, represented by the values 0 to 3. However, the code outputs this:

[[ 1.  1.]
...
 [13.  1.]
 [ 1.  1.]
...
 [13.  1.]
 [ 1.  2.]
...
 [13.  2.]
 [ 1.  3.]
...
 [13.  3.]]

The first 13 indecies are my issue here, as the code I have written I believe should output [1. 0.] up until [13. 0.]

My code is designed to have 13 of each suit, increasing from 1 to 13 (inclusive). Index 0 of each card represents the card value, Ace to King. Index 1 represents its suit (0=S,1=H,2=C,3=D).

suits = 4
suitsize = np.empty(shape=(13,2))
suitsize[:,0] = np.arange(1,suitsize.shape[0]+1)
a = suitsize
print(suitsize)
for i in range(1,suits):
    a[:,1] = i
    suitsize = np.concatenate([suitsize,a])
print(suitsize)

Whether I use np.empty, or np.zeros, the first 13 indecies all still have their index 1 value (suit) replaced with 1. This means that I end up producing a deck of cards with 0 spades, 26 hearts, 13 clubs, and 13 diamonds.

If anyone could explain to me what is happening here or a fix, please let me know. Thank you!

a = suitsize --> a = suitsize.copy() but I don't see the point of concatenation here at all. — BigBen
– BigBen, Commented Apr 30 at 16:00
This question is similar to: How to modify/update an array in numpy without changing original. If you believe it’s different, please edit the question, make it clear how it’s different and/or how the answers on that question are not helpful for your problem. — BigBen
– BigBen, Commented Apr 30 at 16:04
One simpler approach: cards = np.zeros((52, 2)), cards[:, 0] = np.tile(np.arange(13) + 1, 4), cards[:, 1] = np.repeat(np.arange(4), 13). — BigBen
– BigBen, Commented Apr 30 at 16:11

BigBen · Accepted Answer · 2025-04-30 16:56:13Z

2

You have a classic NumPy issue — mutable arrays. The problem is that when you write a = suitsize, you're not creating a new copy of the array. You're just making a new reference to the same array in the computer's memory. So when you do a[:,1] = i, you're also modifying suitsize at the same time.

You need to create an actual copy of the array, not just a reference. Just change this line a = suitsize to a = suitsize.copy() .

edited Apr 30 at 16:56

BigBen

50.4k7 gold badges29 silver badges45 bronze badges

answered Apr 30 at 16:10

Viktor Sbruev

1,0181 gold badge6 silver badges28 bronze badges

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

Collectives™ on Stack Overflow

Numpy Concatenate Replacing Previous Data in Array

1 Answer 1

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Linked

Related