Closed
Description
Describe the issue:
It seems that the dtype of np.vectorize results differs between NumPy 2.2 and earlier versions and NumPy 2.3. Is this an intended change?
Reproduce the code example:
>>> np.vectorize(lambda x: x + x)(np.array([1, 2, 3, 4], dtype=np.int32)).dtype
dtype('int64') # dtype('int32') if numpy<=2.2Python and NumPy Versions:
Python: 3.13.1
NumPy: 2.3.0
Runtime Environment:
>>> np.show_runtime()
[{'numpy_version': '2.3.0',
'python': '3.13.1 (main, Jan 7 2025, 10:32:19) [GCC 9.4.0]',
'uname': uname_result(system='Linux', node='pg00', release='5.15.0-135-generic', version='#146-Ubuntu SMP Sat Feb 15 17:06:22 UTC 2025', machine='x86_64')},
{'simd_extensions': {'baseline': ['SSE', 'SSE2', 'SSE3'],
'found': ['SSSE3',
'SSE41',
'POPCNT',
'SSE42',
'AVX',
'F16C',
'FMA3',
'AVX2',
'AVX512F',
'AVX512CD',
'AVX512_SKX'],
'not_found': ['AVX512_KNL',
'AVX512_KNM',
'AVX512_CLX',
'AVX512_CNL',
'AVX512_ICL',
'AVX512_SPR']}},
{'architecture': 'SkylakeX',
'filepath': '/home/imanishi/.pyenv/versions/3.13.1/lib/python3.13/site-packages/numpy.libs/libscipy_openblas64_-56d6093b.so',
'internal_api': 'openblas',
'num_threads': 64,
'prefix': 'libscipy_openblas',
'threading_layer': 'pthreads',
'user_api': 'blas',
'version': '0.3.29'}]
Context for the issue:
If this behavior is intentional, we will update new CuPy versions to conform to ensure compatibility with this NumPy specification.