How do I create an empty NumPy array and append items to it?

General Discussions
1

I want to create an empty array and add items to it one at a time, similar to Python lists:

xs = []
for item in data:
    xs.append(item)

Can I use this approach with NumPy arrays, or is there a better way to create empty NumPy array and append elements efficiently?

2

Hey! I’ve tried this exact pattern before :sweat_smile:. You can start with an empty NumPy array and append items like this:

import numpy as np

arr = np.array([])  # empty array
for item in data:
    arr = np.append(arr, item)

It works fine for small datasets.

Caveat: np.append creates a new array each time, so it’s not very efficient for large loops.

3

If you know the number of items in advance, I usually pre-allocate the array:

import numpy as np

arr = np.empty(len(data))  # creates an empty array of the desired size
for i, item in enumerate(data):
    arr[i] = item

This avoids repeated copying and is much faster on large datasets.

I always use this method when performance matters.

4

A pattern I often use is combining Python lists with NumPy at the end:

import numpy as np

xs = []
for item in data:
    xs.append(item)

arr = np.array(xs)

Works well for unknown sizes.

Keeps the code simple and readable.

I usually use this for quick scripts or when performance is not critical.