Fill empty numpy array inside for loop

Question

I have a 2-D numpy array X with shape (100, 4). I want to find the sum of each row of that array and store it inside a new numpy array x_new with shape (100,0). What I've done so far doesn't work. Any suggestions ?. Below is my approach.

x_new = np.empty([100,0])
for i in range(len(X)):
    array = np.append(x_new, sum(X[i]))

You can just use x_new = X.sum(axis=1) to get an array of size 100, and then reshape that to 100x1 if you really need it to be two-dimensional. Note that it is 100x1, not 100x0. — Frank Yellin
– Frank Yellin, Commented Oct 24, 2021 at 20:34
you can't put any values in a (100,0) shape array. See the 0? np.append is hard to use unless you really read its docs. Even then it is slower than list append. — hpaulj
– hpaulj, Commented Oct 24, 2021 at 21:02
As a very general rule of thumb, if you find yourself using a for loop with numpy arrays take a step back and take a different approach, non-looping approach to your problem. There's almost always a way with numpy's vectorization and broadcasting. — ddejohn
– ddejohn, Commented Oct 24, 2021 at 22:14

hpaulj · Accepted Answer · 2021-10-24 22:16:38Z

1

Using the sum method on a 2d array:

In [8]: x = np.arange(12).reshape(3,4)
In [9]: x
Out[9]: 
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])
In [10]: x.sum(axis=1)
Out[10]: array([ 6, 22, 38])
In [12]: x.sum(axis=1, keepdims=True)
Out[12]: 
array([[ 6],
       [22],
       [38]])
In [13]: _.shape
Out[13]: (3, 1)

reference: https://numpy.org/doc/stable/reference/generated/numpy.sum.html

answered Oct 24, 2021 at 22:16

hpaulj

233k14 gold badges260 silver badges392 bronze badges

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

Collectives™ on Stack Overflow

Fill empty numpy array inside for loop

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related