numpy-deprecated-type-alias (NPY001)

Fix is sometimes available.

What it does

Checks for deprecated NumPy type aliases.

Why is this bad?

NumPy's np.int has long been an alias of the builtin int; the same is true of np.float and others. These aliases exist primarily for historic reasons, and have been a cause of frequent confusion for newcomers.

These aliases were deprecated in 1.20, and removed in 1.24. Note, however, that np.bool and np.long were reintroduced in 2.0 with different semantics, and are thus omitted from this rule.

Examples

import numpy as np

np.int

Use instead:

int