Numpy Array Data Type Objects (dtype)

Programming Diary
1

Here’s another fun little example regarding numpy indices.

import numpy as np

array = np.array([1, 2])
print(array)

[1 2]

Now, I’m trying to assign a new value to the first item of the array:

array[0] = 1.5
print(array)

[1 2]

This is rather unexpected.
I would have expected the following:

[1.5 2]

The issue here is that numpy automatically detects the data type that describes how the bytes in the fixed-size block of memory corresponding to an array item should be interpreted.
For more information, refer to: Data type objects (dtype) — NumPy v1.25 Manual

To fix this issue, explicitly set the dtype when first creating the array:

array = np.array([1, 2], dtype=np.float64)

now,

array[0] = 1.5

yields the expected result.

print(array)

[1.5 2]
3 Likes