2D Arrays

Let’s now create a higher dimensional array, namely a 2D array:

Basically, creating a higher-dimensional NumPy array involves passing a higher-dimensional list as the argument of the array() function.

Note

Any NumPy array object is called an ndarray.

Here is a visualization of our 2D array:

We can think of it as a 2x3 matrix.

3D Array (Optional)

Creating 3D arrays is nearly identical to creating 2D arrays. The only difference is that we now need to pass a 3D list as an argument:

However, visualizing a 3D array is a bit more complex, but it can still be done:

The array is 3x3x3, which is why we have a cube with each side equal to 3. The innermost 1D arrays lie along axis 2 (e.g., [1, 2, 3]), where each small cube with a side length of 1 is a particular element (number).

All the elements of a 3D array are stored inside these innermost 1D arrays. The cube is just a visual representation to make things clear. The total number of elements (small cubes) is 27 (the volume of the cube).

However, in most cases, you will only deal with 1D and 2D arrays.