Besides basic array creation by explicitly specifying the elements, numpy also allows automatic array creation using special functions. Here are two of the most common functions for creating exclusively 1D arrays:

• arange();
• linspace().

arange()

The numpy.arange() function is similar to Python's built-in range() function; however, it returns an ndarray. Essentially, it creates an array with evenly spaced elements within a specified interval.

Here are its three most important parameters and their roles:

1. start:
   • Default value: 0;
   • Represents the first element of the array.
2. stop:
   • No default value;
   • Defines the endpoint, which is not included in the array.
3. step:
   • Default value: 1;
   • Specifies the increment added to each subsequent element.

Explanation:

• The first element of the array is determined by start;
• Each subsequent element is calculated by adding step to the previous element;
• This process continues until the value reaches or exceeds stop (which is not included in the array).

Let’s see this function in action:

For array_1, we only set the stop parameter to 11. For array_2, we set both start to 1 and stop to 11. For array_3, we specified all three parameters with step=2.

linspace()

While arange() can work with real numbers, it is often better to use numpy.linspace() for this purpose. With linspace(), instead of the step parameter, there is a num parameter used to specify the number of samples (numbers) within a given interval (default is 50).

Let’s see how we can use this function:

As you can see, everything is quite simple here.

Endpoint

Let’s focus on the endpoint boolean parameter. Its default value is True, meaning the stop value is inclusive. Setting it to False excludes the stop value, thus making the step smaller and shifting the interval. Let's take a look at array_inclusive and array_exclusive for comparison:

When endpoint=True, the interval [0, 1] is divided into 4 equal segments and includes the endpoint itself (1), resulting in a step size of (1 - 0) / 4 = 0.25. This generates the values: [0, 0.25, 0.5, 0.75, 1].

When endpoint=False, the interval [0, 1) is divided into 5 equal segments since the endpoint is excluded, resulting in a step size of (1 - 0) / 5 = 0.2. This generates the values: [0, 0.2, 0.4, 0.6, 0.8].

Note

You can always learn more about these functions in their documentation: arange, linspace.