ImportError: cannot import name 'VisibleDeprecationWarning' from 'numpy'

TL;DR: VisibleDeprecationWarning was removed in numpy 2.0, so from numpy import VisibleDeprecationWarning cannot work anymore. Replace it with Python's builtin warnings module. If you cannot change the code, pin numpy below 2.0 as a stopgap.

The error

ImportError: cannot import name 'VisibleDeprecationWarning' from 'numpy'

Fix it

1. Find every place the removed name is imported:

   grep -rn "VisibleDeprecationWarning" [your project dir]

Expected: a list of files doing from numpy import VisibleDeprecationWarning.

1. Replace the import with the stdlib warnings module:

   import warnings

and where the code raised or filtered it, use warnings.warn(msg, DeprecationWarning) or the builtin DeprecationWarning class. Expected: no remaining references to the numpy import.

1. Re-run the import that failed. Expected: it imports cleanly.
2. If you cannot edit the code (a pinned third-party package), downgrade numpy instead:

   pip install "numpy<2.0"

Expected: the old import works again.

When this applies

Use this when the traceback points at from numpy import VisibleDeprecationWarning and numpy 2.x is installed.

When it does not apply

If the error names a different removed symbol (np.float, np.unicode_, np.asfarray), each needs its own replacement. If numpy is still 1.x, something else is wrong.

Compatibility

numpy 2.0 and later. VisibleDeprecationWarning existed through numpy 1.26.

Root cause

NumPy 2.0 did a long-planned cleanup of deprecated names. VisibleDeprecationWarning was a thin subclass of UserWarning, so the stdlib warnings module plus DeprecationWarning is the intended replacement.

Edge cases

Some libraries vendored the name; upgrading the library to a numpy-2-compatible release is better than patching. If tests filter on the warning class, update the filter to DeprecationWarning too.