Update dependency numpy to v2

[renovate]

This PR contains the following updates:

Package	Change	Age	Adoption	Passing	Confidence
numpy (source, changelog)	^1.14.0 -> ^2.0.0

---

Release Notes

numpy/numpy (numpy)

v2.0.0

Compare Source

NumPy 2.0.0 Release Notes

NumPy 2.0.0 is the first major release since 2006. It is the result of
11 months of development since the last feature release and is the work
of 212 contributors spread over 1078 pull requests. It contains a large
number of exciting new features as well as changes to both the Python
and C APIs.

This major release includes breaking changes that could not happen in a
regular minor (feature) release - including an ABI break, changes to
type promotion rules, and API changes which may not have been emitting
deprecation warnings in 1.26.x. Key documents related to how to adapt to
changes in NumPy 2.0, in addition to these release notes, include:

• The numpy-2-migration-guide
• The Numpy 2.0-specific advice in for downstream package authors

Highlights

Highlights of this release include:

• New features:
  • A new variable-length string dtype, numpy.dtypes.StringDType and a new
    numpy.strings namespace with performant ufuncs for string operations,
  • Support for float32 and longdouble in all
    numpy.fft functions,
  • Support for the array API standard in the main numpy
    namespace.
• Performance improvements:
  • Sorting functions sort, argsort,
    partition, argpartition have been
    accelerated through the use of the Intel x86-simd-sort and
    Google Highway libraries, and may see large (hardware-specific)
    speedups,
  • macOS Accelerate support and binary wheels for macOS >=14, with
    significant performance improvements for linear algebra
    operations on macOS, and wheels that are about 3 times smaller,
  • numpy.char fixed-length string operations have
    been accelerated by implementing ufuncs that also support
    numpy.dtypes.StringDType in addition to the
    fixed-length string dtypes,
  • A new tracing and introspection API,
    numpy.lib.introspect.opt_func_info, to determine
    which hardware-specific kernels are available and will be
    dispatched to.
  • numpy.save now uses pickle protocol version 4 for saving
    arrays with object dtype, which allows for pickle objects larger
    than 4GB and improves saving speed by about 5% for large arrays.
• Python API improvements:
  • A clear split between public and private API, with a new module
    structure and each public function now available in a single place.
  • Many removals of non-recommended functions and aliases. This
    should make it easier to learn and use NumPy. The number of
    objects in the main namespace decreased by ~10% and in
    numpy.lib by ~80%.
  • Canonical dtype names and a new numpy.isdtype` introspection
    function,
• C API improvements:
  • A new public C API for creating custom dtypes,
  • Many outdated functions and macros removed, and private
    internals hidden to ease future extensibility,
  • New, easier to use, initialization functions: PyArray_ImportNumPyAPI
    and PyUFunc_ImportUFuncAPI.
• Improved behavior:
  • Improvements to type promotion behavior was changed by adopting NEP 50.
    This fixes many user surprises about promotions which previously often
    depended on data values of input arrays rather than only their dtypes.
    Please see the NEP and the numpy-2-migration-guide for details as this
    change can lead to changes in output dtypes and lower precision results
    for mixed-dtype operations.
  • The default integer type on Windows is now int64 rather than
    int32, matching the behavior on other platforms,
  • The maximum number of array dimensions is changed from 32 to 64
• Documentation:
  • The reference guide navigation was significantly improved, and
    there is now documentation on NumPy's
    module structure,
  • The building from source documentation was completely rewritten,

Furthermore there are many changes to NumPy internals, including
continuing to migrate code from C to C++, that will make it easier to
improve and maintain NumPy in the future.

The "no free lunch" theorem dictates that there is a price to pay for
all these API and behavior improvements and better future extensibility.
This price is:

1. Backwards compatibility. There are a significant number of breaking
   changes to both the Python and C APIs. In the majority of cases,
   there are clear error messages that will inform the user how to
   adapt their code. However, there are also changes in behavior for
   which it was not possible to give such an error message - these
   cases are all covered in the Deprecation and Compatibility sections
   below, and in the numpy-2-migration-guide.

   Note that there is a ruff mode to auto-fix many things in Python
   code.

2. Breaking changes to the NumPy ABI. As a result, binaries of packages
   that use the NumPy C API and were built against a NumPy 1.xx release
   will not work with NumPy 2.0. On import, such packages will see an
   ImportError with a message about binary incompatibility.

   It is possible to build binaries against NumPy 2.0 that will work at
   runtime with both NumPy 2.0 and 1.x. See numpy-2-abi-handling for more
   details.

   All downstream packages that depend on the NumPy ABI are advised
   to do a new release built against NumPy 2.0 and verify that that
   release works with both 2.0 and 1.26 - ideally in the period between
   2.0.0rc1 (which will be ABI-stable) and the final 2.0.0 release to
   avoid problems for their users.

The Python versions supported by this release are 3.9-3.12.

NumPy 2.0 Python API removals

• np.geterrobj, np.seterrobj and the related ufunc keyword
  argument extobj= have been removed. The preferred replacement for
  all of these is using the context manager with np.errstate():.

  (gh-23922)

• np.cast has been removed. The literal replacement for
  np.cast[dtype](arg) is np.asarray(arg, dtype=dtype).

• np.source has been removed. The preferred replacement is
  inspect.getsource.

• np.lookfor has been removed.

  (gh-24144)

• numpy.who has been removed. As an alternative for the removed
  functionality, one can use a variable explorer that is available in
  IDEs such as Spyder or Jupyter Notebook.

  (gh-24321)

• Warnings and exceptions present in numpy.exceptions,
  e.g, numpy.exceptions.ComplexWarning,
  numpy.exceptions.VisibleDeprecationWarning, are no
  longer exposed in the main namespace.

• Multiple niche enums, expired members and functions have been
  removed from the main namespace, such as: ERR_*, SHIFT_*,
  np.fastCopyAndTranspose, np.kernel_version, np.numarray,
  np.oldnumeric and np.set_numeric_ops.

  (gh-24316)

• Replaced from ... import * in the numpy/__init__.py with
  explicit imports. As a result, these main namespace members got
  removed: np.FLOATING_POINT_SUPPORT, np.FPE_*, np.NINF,
  np.PINF, np.NZERO, np.PZERO, np.CLIP, np.WRAP, np.WRAP,
  np.RAISE, np.BUFSIZE, np.UFUNC_BUFSIZE_DEFAULT,
  np.UFUNC_PYVALS_NAME, np.ALLOW_THREADS, np.MAXDIMS,
  np.MAY_SHARE_EXACT, np.MAY_SHARE_BOUNDS, add_newdoc,
  np.add_docstring and np.add_newdoc_ufunc.

  (gh-24357)

• Alias np.float_ has been removed. Use np.float64 instead.

• Alias np.complex_ has been removed. Use np.complex128 instead.

• Alias np.longfloat has been removed. Use np.longdouble instead.

• Alias np.singlecomplex has been removed. Use np.complex64
  instead.

• Alias np.cfloat has been removed. Use np.complex128 instead.

• Alias np.longcomplex has been removed. Use np.clongdouble
  instead.

• Alias np.clongfloat has been removed. Use np.clongdouble
  instead.

• Alias np.string_ has been removed. Use np.bytes_ instead.

• Alias np.unicode_ has been removed. Use np.str_ instead.

• Alias np.Inf has been removed. Use np.inf instead.

• Alias np.Infinity has been removed. Use np.inf instead.

• Alias np.NaN has been removed. Use np.nan instead.

• Alias np.infty has been removed. Use np.inf instead.

• Alias np.mat has been removed. Use np.asmatrix instead.

• np.issubclass_ has been removed. Use the issubclass builtin
  instead.

• np.asfarray has been removed. Use np.asarray with a proper dtype
  instead.

• np.set_string_function has been removed. Use np.set_printoptions
  instead with a formatter for custom printing of NumPy objects.

• np.tracemalloc_domain is now only available from np.lib.

• np.recfromcsv and recfromtxt are now only available from
  np.lib.npyio.

• np.issctype, np.maximum_sctype, np.obj2sctype,
  np.sctype2char, np.sctypes, np.issubsctype were all removed
  from the main namespace without replacement, as they where niche
  members.

• Deprecated np.deprecate and np.deprecate_with_doc has been
  removed from the main namespace. Use DeprecationWarning instead.

• Deprecated np.safe_eval has been removed from the main namespace.
  Use ast.literal_eval instead.

  (gh-24376)

• np.find_common_type has been removed. Use numpy.promote_types or
  numpy.result_type instead. To achieve semantics for the
  scalar_types argument, use numpy.result_type and pass 0,
  0.0, or 0j as a Python scalar instead.

• np.round_ has been removed. Use np.round instead.

• np.nbytes has been removed. Use np.dtype(<dtype>).itemsize
  instead.

  (gh-24477)

• np.compare_chararrays has been removed from the main namespace.
  Use np.char.compare_chararrays instead.

• The charrarray in the main namespace has been deprecated. It can
  be imported without a deprecation warning from np.char.chararray
  for now, but we are planning to fully deprecate and remove
  chararray in the future.

• np.format_parser has been removed from the main namespace. Use
  np.rec.format_parser instead.

  (gh-24587)

• Support for seven data type string aliases has been removed from
  np.dtype: int0, uint0, void0, object0, str0, bytes0
  and bool8.

  (gh-24807)

• The experimental numpy.array_api submodule has been removed. Use
  the main numpy namespace for regular usage instead, or the
  separate array-api-strict package for the compliance testing use
  case for which numpy.array_api was mostly used.

  (gh-25911)

__array_prepare__ is removed

UFuncs called __array_prepare__ before running computations for normal
ufunc calls (not generalized ufuncs, reductions, etc.). The function was
also called instead of __array_wrap__ on the results of some linear
algebra functions.

It is now removed. If you use it, migrate to __array_ufunc__ or rely
on __array_wrap__ which is called with a context in all cases,
although only after the result array is filled. In those code paths,
__array_wrap__ will now be passed a base class, rather than a subclass
array.

(gh-25105)

Deprecations

• np.compat has been deprecated, as Python 2 is no longer supported.

• numpy.int8 and similar classes will no longer support conversion
  of out of bounds python integers to integer arrays. For example,
  conversion of 255 to int8 will not return -1. numpy.iinfo(dtype)
  can be used to check the machine limits for data types. For example,
  np.iinfo(np.uint16) returns min = 0 and max = 65535.

  np.array(value).astype(dtype) will give the desired result.

• np.safe_eval has been deprecated. ast.literal_eval should be
  used instead.

  (gh-23830)

• np.recfromcsv, np.recfromtxt, np.disp, np.get_array_wrap,
  np.maximum_sctype, np.deprecate and np.deprecate_with_doc have
  been deprecated.

  (gh-24154)

• np.trapz has been deprecated. Use np.trapezoid or a
  scipy.integrate function instead.

• np.in1d has been deprecated. Use np.isin instead.

• Alias np.row_stack has been deprecated. Use np.vstack directly.

  (gh-24445)

• __array_wrap__ is now passed arr, context, return_scalar and
  support for implementations not accepting all three are deprecated.
  Its signature should be
  __array_wrap__(self, arr, context=None, return_scalar=False)

  (gh-25409)

• Arrays of 2-dimensional vectors for np.cross have been deprecated.
  Use arrays of 3-dimensional vectors instead.

  (gh-24818)

• np.dtype("a") alias for np.dtype(np.bytes_) was deprecated. Use
  np.dtype("S") alias instead.

  (gh-24854)

• Use of keyword arguments x and y with functions
  assert_array_equal and assert_array_almost_equal has been
  deprecated. Pass the first two arguments as positional arguments
  instead.

  (gh-24978)

numpy.fft deprecations for n-D transforms with None values in arguments

Using fftn, ifftn, rfftn, irfftn, fft2, ifft2, rfft2 or
irfft2 with the s parameter set to a value that is not None and
the axes parameter set to None has been deprecated, in line with the
array API standard. To retain current behaviour, pass a sequence [0,
..., k-1] to axes for an array of dimension k.

Furthermore, passing an array to s which contains None values is
deprecated as the parameter is documented to accept a sequence of
integers in both the NumPy docs and the array API specification. To use
the default behaviour of the corresponding 1-D transform, pass the value
matching the default for its n parameter. To use the default behaviour
for every axis, the s argument can be omitted.

(gh-25495)

np.linalg.lstsq now defaults to a new rcond value

numpy.linalg.lstsq now uses the new rcond value of the
machine precision times max(M, N). Previously, the machine precision
was used but a FutureWarning was given to notify that this change will
happen eventually. That old behavior can still be achieved by passing
rcond=-1.

(gh-25721)

Expired deprecations

• The np.core.umath_tests submodule has been removed from the public
  API. (Deprecated in NumPy 1.15)

  (gh-23809)

• The PyDataMem_SetEventHook deprecation has expired and it is
  removed. Use tracemalloc and the np.lib.tracemalloc_domain
  domain. (Deprecated in NumPy 1.23)

  (gh-23921)

• The deprecation of set_numeric_ops and the C functions
  PyArray_SetNumericOps and PyArray_GetNumericOps has been expired
  and the functions removed. (Deprecated in NumPy 1.16)

  (gh-23998)

• The fasttake, fastclip, and fastputmask ArrFuncs deprecation
  is now finalized.

• The deprecated function fastCopyAndTranspose and its C counterpart
  are now removed.

• The deprecation of PyArray_ScalarFromObject is now finalized.

  (gh-24312)

• np.msort has been removed. For a replacement, np.sort(a, axis=0)
  should be used instead.

  (gh-24494)

• np.dtype(("f8", 1) will now return a shape 1 subarray dtype rather
  than a non-subarray one.

  (gh-25761)

• Assigning to the .data attribute of an ndarray is disallowed and
  will raise.

• np.binary_repr(a, width) will raise if width is too small.

• Using NPY_CHAR in PyArray_DescrFromType() will raise, use
  NPY_STRING NPY_UNICODE, or NPY_VSTRING instead.

  (gh-25794)

Compatibility notes

loadtxt and genfromtxt default encoding changed

loadtxt and genfromtxt now both default to encoding=None which may
mainly modify how converters work. These will now be passed str
rather than bytes. Pass the encoding explicitly to always get the new
or old behavior. For genfromtxt the change also means that returned
values will now be unicode strings rather than bytes.

(gh-25158)

f2py compatibility notes

• f2py will no longer accept ambiguous -m and .pyf CLI
  combinations. When more than one .pyf file is passed, an error is
  raised. When both -m and a .pyf is passed, a warning is emitted
  and the -m provided name is ignored.

  (gh-25181)

• The f2py.compile() helper has been removed because it leaked
  memory, has been marked as experimental for several years now, and
  was implemented as a thin subprocess.run wrapper. It was also one
  of the test bottlenecks. See
  gh-25122 for the full
  rationale. It also used several np.distutils features which are
  too fragile to be ported to work with meson.

• Users are urged to replace calls to f2py.compile with calls to
  subprocess.run("python", "-m", "numpy.f2py",... instead, and to
  use environment variables to interact with meson. Native
  files are also an
  option.

  (gh-25193)

Minor changes in behavior of sorting functions

Due to algorithmic changes and use of SIMD code, sorting functions with
methods that aren't stable may return slightly different results in
2.0.0 compared to 1.26.x. This includes the default method of
numpy.argsort and numpy.argpartition.

Removed ambiguity when broadcasting in np.solve

The broadcasting rules for np.solve(a, b) were ambiguous when b had
1 fewer dimensions than a. This has been resolved in a
backward-incompatible way and is now compliant with the Array API. The
old behaviour can be reconstructed by using
np.solve(a, b[..., None])[..., 0].

(gh-25914)

Modified representation for Polynomial

The representation method for
numpy.polynomial.polynomial.Polynomial was updated to
include the domain in the representation. The plain text and latex
representations are now consistent. For example the output of
str(np.polynomial.Polynomial([1, 1], domain=[.1, .2])) used to be
1.0 + 1.0 x, but now is 1.0 + 1.0 (-3.0000000000000004 + 20.0 x).

(gh-21760)

C API changes

• The PyArray_CGT, PyArray_CLT, PyArray_CGE, PyArray_CLE,
  PyArray_CEQ, PyArray_CNE macros have been removed.

• PyArray_MIN and PyArray_MAX have been moved from
  ndarraytypes.h to npy_math.h.

  (gh-24258)

• A C API for working with numpy.dtypes.StringDType
  arrays has been exposed. This includes functions for acquiring and
  releasing mutexes which lock access to the string data, as well as
  packing and unpacking UTF-8 bytestreams from array entries.

• NPY_NTYPES has been renamed to NPY_NTYPES_LEGACY as it does not
  include new NumPy built-in DTypes. In particular the new string
  DType will likely not work correctly with code that handles legacy
  DTypes.

  (gh-25347)

• The C-API now only exports the static inline function versions of
  the array accessors (previously this depended on using "deprecated
  API"). While we discourage it, the struct fields can still be used
  directly.

  (gh-25789)

• NumPy now defines PyArray_Pack to set an individual memory address.
  Unlike PyArray_SETITEM this function is equivalent to setting an
  individual array item and does not require a NumPy array input.

  (gh-25954)

• The ->f slot has been removed from PyArray_Descr. If you use this slot,
  replace accessing it with PyDataType_GetArrFuncs (see its documentation
  and the numpy-2-migration-guide). In some cases using other functions
  like PyArray_GETITEM may be an alternatives.

• PyArray_GETITEM and PyArray_SETITEM now require the import of
  the NumPy API table to be used and are no longer defined in
  ndarraytypes.h.

  (gh-25812)

• Due to runtime dependencies, the definition for functionality
  accessing the dtype flags was moved from numpy/ndarraytypes.h and
  is only available after including numpy/ndarrayobject.h as it
  requires import_array(). This includes PyDataType_FLAGCHK,
  PyDataType_REFCHK and NPY_BEGIN_THREADS_DESCR.

• The dtype flags on PyArray_Descr must now be accessed through the
  PyDataType_FLAGS inline function to be compatible with both 1.x
  and 2.x. This function is defined in npy_2_compat.h to allow
  backporting. Most or all users should use PyDataType_FLAGCHK which
  is available on 1.x and does not require backporting. Cython users
  should use Cython 3. Otherwise access will go through Python unless
  they use PyDataType_FLAGCHK instead.

  (gh-25816)

Datetime functionality exposed in the C API and Cython bindings

The functions NpyDatetime_ConvertDatetime64ToDatetimeStruct,
NpyDatetime_ConvertDatetimeStructToDatetime64,
NpyDatetime_ConvertPyDateTimeToDatetimeStruct,
NpyDatetime_GetDatetimeISO8601StrLen,
NpyDatetime_MakeISO8601Datetime, and
NpyDatetime_ParseISO8601Datetime have been added to the C API to
facilitate converting between strings, Python datetimes, and NumPy
datetimes in external libraries.

(gh-21199)

Const correctness for the generalized ufunc C API

The NumPy C API's functions for constructing generalized ufuncs
(PyUFunc_FromFuncAndData, PyUFunc_FromFuncAndDataAndSignature,
PyUFunc_FromFuncAndDataAndSignatureAndIdentity) take types and
data arguments that are not modified by NumPy's internals. Like the
name and doc arguments, third-party Python extension modules are
likely to supply these arguments from static constants. The types and
data arguments are now const-correct: they are declared as
const char *types and void *const *data, respectively. C code should
not be affected, but C++ code may be.

(gh-23847)

Larger NPY_MAXDIMS and NPY_MAXARGS, NPY_RAVEL_AXIS introduced

NPY_MAXDIMS is now 64, you may want to review its use. This is usually
used in a stack allocation, where the increase should be safe. However,
we do encourage generally to remove any use of NPY_MAXDIMS and
NPY_MAXARGS to eventually allow removing the constraint completely.
For the conversion helper and C-API functions mirroring Python ones such as
take, NPY_MAXDIMS was used to mean axis=None. Such usage must be replaced
with NPY_RAVEL_AXIS. See also migration_maxdims.

(gh-25149)

NPY_MAXARGS not constant and PyArrayMultiIterObject size change

Since NPY_MAXARGS was increased, it is now a runtime constant and not
compile-time constant anymore. We expect almost no users to notice this.
But if used for stack allocations it now must be replaced with a custom
constant using NPY_MAXARGS as an additional runtime check.

The sizeof(PyArrayMultiIterObject) no longer includes the full size of
the object. We expect nobody to notice this change. It was necessary to
avoid issues with Cython.

(gh-25271)

Required changes for custom legacy user dtypes

In order to improve our DTypes it is unfortunately necessary to break
the ABI, which requires some changes for dtypes registered with
PyArray_RegisterDataType. Please see the documentation of
PyArray_RegisterDataType for how to adapt your code and achieve
compatibility with both 1.x and 2.x.

(gh-25792)

New Public DType API

The C implementation of the NEP 42 DType API is now public. While the
DType API has shipped in NumPy for a few versions, it was only usable in
sessions with a special environment variable set. It is now possible to
write custom DTypes outside of NumPy using the new DType API and the
normal import_array() mechanism for importing the numpy C API.

See dtype-api for more details about the API. As always with a new feature,
please report any bugs you run into implementing or using a new DType. It is
likely that downstream C code that works with dtypes will need to be updated to
work correctly with new DTypes.

(gh-25754)

New C-API import functions

We have now added PyArray_ImportNumPyAPI and PyUFunc_ImportUFuncAPI
as static inline functions to import the NumPy C-API tables. The new
functions have two advantages over import_array and import_ufunc:

• They check whether the import was already performed and are
  light-weight if not, allowing to add them judiciously (although this
  is not preferable in most cases).
• The old mechanisms were macros rather than functions which included
  a return statement.

The PyArray_ImportNumPyAPI() function is included in npy_2_compat.h
for simpler backporting.

(gh-25866)

Structured dtype information access through functions

The dtype structures fields c_metadata, names, fields, and
subarray must now be accessed through new functions following the same
names, such as PyDataType_NAMES. Direct access of the fields is not
valid as they do not exist for all PyArray_Descr instances. The
metadata field is kept, but the macro version should also be
preferred.

(gh-25802)

Descriptor elsize and alignment access

Unless compiling only with NumPy 2 support, the elsize and aligment
fields must now be accessed via PyDataType_ELSIZE,
PyDataType_SET_ELSIZE, and PyDataType_ALIGNMENT. In cases where the
descriptor is attached to an array, we advise using PyArray_ITEMSIZE
as it exists on all NumPy versions. Please see
migration_c_descr for more information.

(gh-25943)

NumPy 2.0 C API removals

• npy_interrupt.h and the corresponding macros like NPY_SIGINT_ON
  have been removed. We recommend querying PyErr_CheckSignals() or
  PyOS_InterruptOccurred() periodically (these do currently require
  holding the GIL though).

• The noprefix.h header has been removed. Replace missing symbols
  with their prefixed counterparts (usually an added NPY_ or
  npy_).

  (gh-23919)

• PyUFunc_GetPyVals, PyUFunc_handlefperr, and PyUFunc_checkfperr
  have been removed. If needed, a new backwards compatible function to
  raise floating point errors could be restored. Reason for removal:
  there are no known users and the functions would have made
  with np.errstate() fixes much more difficult).

  (gh-23922)

• The numpy/old_defines.h which was part of the API deprecated since
  NumPy 1.7 has been removed. This removes macros of the form
  PyArray_CONSTANT. The
  replace_old_macros.sed
  script may be useful to convert them to the NPY_CONSTANT version.

  (gh-24011)

• The legacy_inner_loop_selector member of the ufunc struct is
  removed to simplify improvements to the dispatching system. There
  are no known users overriding or directly accessing this member.

  (gh-24271)

• NPY_INTPLTR has been removed to avoid confusion (see intp
  redefinition).

  (gh-24888)

• The advanced indexing MapIter and related API has been removed.
  The (truly) public part of it was not well tested and had only one
  known user (Theano). Making it private will simplify improvements to
  speed up ufunc.at, make advanced indexing more maintainable, and
  was important for increasing the maximum number of dimensions of
  arrays to 64. Please let us know if this API is important to you so
  we can find a solution together.

  (gh-25138)

• The NPY_MAX_ELSIZE macro has been removed, as it only ever
  reflected builtin numeric types and served no internal purpose.

  (gh-25149)

• PyArray_REFCNT and NPY_REFCOUNT are removed. Use Py_REFCNT
  instead.

  (gh-25156)

• PyArrayFlags_Type and PyArray_NewFlagsObject as well as
  PyArrayFlagsObject are private now. There is no known use-case;
  use the Python API if needed.

• PyArray_MoveInto, PyArray_CastTo, PyArray_CastAnyTo are
  removed use PyArray_CopyInto and if absolutely needed
  PyArray_CopyAnyInto (the latter does a flat copy).

• PyArray_FillObjectArray is removed, its only true use was for
  implementing np.empty. Create a new empty array or use
  PyArray_FillWithScalar() (decrefs existing objects).

• PyArray_CompareUCS4 and PyArray_CompareString are removed. Use
  the standard C string comparison functions.

• PyArray_ISPYTHON is removed as it is misleading, has no known
  use-cases, and is easy to replace.

• PyArray_FieldNames is removed, as it is unclear what it would be
  useful for. It also has incorrect semantics in some possible
  use-cases.

• PyArray_TypestrConvert is removed, since it seems a misnomer and
  unlikely to be used by anyone. If you know the size or are limited
  to few types, just use it explicitly, otherwise go via Python
  strings.

  (gh-25292)

• PyDataType_GetDatetimeMetaData is removed, it did not actually do
  anything since at least NumPy 1.7.

  (gh-25802)

• PyArray_GetCastFunc is removed. Note that custom legacy user
  dtypes can still provide a castfunc as their implementation, but any
  access to them is now removed. The reason for this is that NumPy
  never used these internally for many years. If you use simple
  numeric types, please just use C casts directly. In case you require
  an alternative, please let us know so we can create new API such as
  PyArray_CastBuffer() which could use old or new cast functions
  depending on the NumPy version.

  (gh-25161)

New Features

np.add was extended to work with unicode and bytes dtypes.

(gh-24858)

A new bitwise_count function

This new function counts the number of 1-bits in a number.
numpy.bitwise_count works on all the numpy integer types
and integer-like objects.

>>> a = np.array([2**i - 1 for i in range(16)])
>>> np.bitwise_count(a)
array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15],
      dtype=uint8)

(gh-19355)

macOS Accelerate support, including the ILP64

Support for the updated Accelerate BLAS/LAPACK library, including ILP64
(64-bit integer) support, in macOS 13.3 has been added. This brings
arm64 support, and significant performance improvements of up to 10x for
commonly used linear algebra operations. When Accelerate is selected at
build time, or if no explicit BLAS library selection is done, the 13.3+
version will automatically be used if available.

(gh-24053)

Binary wheels are also available. On macOS >=14.0, users who install
NumPy from PyPI will get wheels built against Accelerate rather than
OpenBLAS.

(gh-25255)

Option to use weights for quantile and percentile functions

A weights keyword is now available for numpy.quantile, numpy.percentile,
numpy.nanquantile and numpy.nanpercentile. Only method="inverted_cdf"
supports weights.

(gh-24254)

Improved CPU optimization tracking

A new tracer mechanism is available which enables tracking of the
enabled targets for each optimized function (i.e., that uses
hardware-specific SIMD instructions) in the NumPy library. With this
enhancement, it becomes possible to precisely monitor the enabled CPU
dispatch targets for the dispatched functions.

A new function named opt_func_info has been added to the new namespace
numpy.lib.introspect, offering this tracing capability. This function allows
you to retrieve information about the enabled targets based on function names
and data type signatures.

(gh-24420)

A new Meson backend for f2py

f2py in compile mode (i.e. f2py -c) now accepts the
--backend meson option. This is the default option for Python >=3.12.
For older Python versions, f2py will still default to
--backend distutils.

To support this in realistic use-cases, in compile mode f2py takes a
--dep flag one or many times which maps to dependency() calls in the
meson backend, and does nothing in the distutils backend.

There are no changes for users of f2py only as a code generator, i.e.
without -c.

(gh-24532)

bind(c) support for f2py

Both functions and subroutines can be annotated with bind(c). f2py
will handle both the correct type mapping, and preserve the unique label
for other C interfaces.

Note: bind(c, name = 'routine_name_other_than_fortran_routine') is
not honored by the f2py bindings by design, since bind(c) with the
name is meant to guarantee only the same name in C and Fortran, not in
Python and Fortran.

(gh-24555)

A new strict option for several testing functions

The strict keyword is now available for numpy.testing.assert_allclose,
numpy.testing.assert_equal, and numpy.testing.assert_array_less. Setting
strict=True will disable the broadcasting behaviour for scalars and ensure
that input arrays have the same data type.

(gh-24680,
gh-24770,
gh-24775)

Add np.core.umath.find and np.core.umath.rfind UFuncs

Add two find and rfind UFuncs that operate on unicode or byte
strings and are used in np.char. They operate similar to str.find
and str.rfind.

(gh-24868)

diagonal and trace for numpy.linalg

numpy.linalg.diagonal and numpy.linalg.trace have been added, which are
array API standard-compatible variants of numpy.diagonal and numpy.trace.
They differ in the default axis selection which define 2-D sub-arrays.

(gh-24887)

New long and ulong dtypes

numpy.long and numpy.ulong have been added as NumPy integers mapping to
C's long and unsigned long. Prior to NumPy 1.24, numpy.long was an alias
to Python's int.

(gh-24922)

svdvals for numpy.linalg

numpy.linalg.svdvals has been added. It computes singular values for (a stack
of) matrices. Executing np.svdvals(x) is the same as calling np.svd(x, compute_uv=False, hermitian=False). This function is compatible with the array
API standard.

(gh-24940)

A new isdtype function

numpy.isdtype was added to provide a canonical way to classify NumPy's
dtypes in compliance with the array API standard.

(gh-25054)

A new astype function

numpy.astype was added to provide an array API standard-compatible
alternative to the numpy.ndarray.astype method.

(gh-25079)

Array API compatible functions' aliases

13 aliases for existing functions were added to improve compatibility
with the array API standard:

• Trigonometry: acos, acosh, asin, asinh, atan, atanh,
  atan2.
• Bitwise: bitwise_left_shift, bitwise_invert,
  bitwise_right_shift.
• Misc: concat, permute_dims, pow.
• In numpy.linalg: tensordot, matmul.

(gh-25086)

New unique_* functions

The numpy.unique_all, numpy.unique_counts, numpy.unique_inverse, and
numpy.unique_values functions have been added. They provide functionality of
numpy.unique with different sets of flags. They are array API
standard-compatible, and because the number of arrays they return does not
depend on the values of input arguments, they are easier to target for JIT
compilation.

(gh-25088)

Matrix transpose support for ndarrays

NumPy now offers support for calculating the matrix transpose of an
array (or stack of arrays). The matrix transpose is equivalent to
swapping the last two axes of an array. Both np.ndarray and
np.ma.MaskedArray now expose a .mT attribute, and there is a
matching new numpy.matrix_transpose function.

(gh-23762)

Array API compatible functions for numpy.linalg

Six new functions and two aliases were added to improve compatibility
with the Array API standard for `numpy.linalg`:

• numpy.linalg.matrix_norm - Computes the matrix norm of
  a matrix (or a stack of matrices).

• numpy.linalg.vector_norm - Computes the vector norm of
  a vector (or batch of vectors).

• numpy.vecdot - Computes the (vector) dot product of
  two arrays.

• numpy.linalg.vecdot - An alias for
  numpy.vecdot.

• numpy.linalg.matrix_transpose - An alias for
  numpy.matrix_transpose.

  (gh-25155)

• numpy.linalg.outer has been added. It computes the
  outer product of two vectors. It differs from
  numpy.outer by accepting one-dimensional arrays only.
  This function is compatible with the array API standard.

  (gh-25101)

• numpy.linalg.cross has been added. It computes the
  cross product of two (arrays of) 3-dimensional vectors. It differs
  from numpy.cross by accepting three-dimensional
  vectors only. This function is compatible with the array API
  standard.

  (gh-25145)

A correction argument for var and std

A correction argument was added to numpy.var and numpy.std, which is an
array API standard compatible alternative to ddof. As both arguments serve a
similar purpose, only one of them can be provided at the same time.

(gh-25169)

ndarray.device and ndarray.to_device

An ndarray.device attribute and ndarray.to_device method were added
to numpy.ndarray for array API standard compatibility.

Additionally, device keyword-only arguments were added to:
numpy.asarray, numpy.arange, numpy.empty, numpy.empty_like,
numpy.eye, numpy.full, numpy.full_like, numpy.linspace, numpy.ones,
numpy.ones_like, numpy.zeros, and numpy.zeros_like.

For all these new arguments, only device="cpu" is supported.

(gh-25233)

StringDType has been added to NumPy

We have added a new variable-width UTF-8 encoded string data type, implementing
a "NumPy array of Python strings", including support for a user-provided
missing data sentinel. It is intended as a drop-in replacement for arrays of
Python strings and missing data sentinels using the object dtype. See
NEP 55 and the documentation
of stringdtype for more details.

(gh-25347)

New keywords for cholesky and pinv

The upper and rtol keywords were added to
numpy.linalg.cholesky and numpy.linalg.pinv,
respectively, to improve array API standard compatibility.

For numpy.linalg.pinv, if neither rcond nor rtol is
specified, the rcond's default is used. We plan to deprecate and
remove rcond in the future.

(gh-25388)

New keywords for sort, argsort and linalg.matrix_rank

New keyword parameters were added to improve array API standard
compatibility:

• rtol was added to numpy.linalg.matrix_rank.
• stable was added to numpy.sort and
  numpy.argsort.

(gh-25437)

New numpy.strings namespace for string ufuncs

NumPy now implements some string operations as ufuncs. The old np.char
namespace is still available, and where possible the string manipulation
functions in that namespace have been updated to use the new ufuncs,
substantially improving their performance.

Where possible, we suggest updating code to use functions in
np.strings instead of np.char. In the future we may deprecate
np.char in favor of np.strings.

(gh-25463)

numpy.fft support for different precisions and in-place calculations

The various FFT routines in numpy.fft now do their
calculations natively in float, double, or long double precision,
depending on the input precision, instead of always calculating in
double precision. Hence, the calculation will now be less precise for
single and more precise for long double precision. The data type of the
output array will now be adjusted accordingly.

Furthermore, all FFT routines have gained an out argument that can be
used for in-place calculations.

(gh-25536)

configtool and pkg-config support

A new numpy-config CLI script is available that can be queried for the
NumPy version and for compile flags needed to use the NumPy C API. This
will allow build systems to better support the use of NumPy as a
dependency. Also, a numpy.pc pkg-config file is now included with
Numpy. In order to find its location for use with PKG_CONFIG_PATH, use
numpy-config --pkgconfigdir.

(gh-25730)

Array API standard support in the main namespace

The main numpy namespace now supports the array API standard. See
array-api-standard-compatibility for
details.

(gh-25911)

Improvements

Strings are now supported by any, all, and the logical ufuncs.

(gh-25651)

Integer sequences as the shape argument for memmap

numpy.memmap can now be created with any integer sequence
as the shape argument, such as a list or numpy array of integers.
Previously, only the types of tuple and int could be used without
raising an error.

(gh-23729)

errstate is now faster and context safe

The numpy.errstate context manager/decorator is now faster
and safer. Previously, it was not context safe and had (rare) issues
with thread-safety.

(gh-23936)

AArch64 quicksort speed improved by using Highway's VQSort

The first introduction of the Google Highway library, using VQSort on
AArch64. Execution time is improved by up to 16x in some cases, see the
PR for benchmark results. Extensions to other platforms will be done in
the future.

(gh-24018)

Complex types - underlying C type changes

• The underlying C types for all of NumPy's complex types have been
  changed to use C99 complex types.

• While this change does not affect the memory layout of complex
  types, it changes the API to be used to directly retrieve or write
  the real or complex part of the complex number, since direct field
  access (as in c.real or c.imag) is no longer an option. You can
  now use utilities provided in numpy/npy_math.h to do these
  operations, like this:

  npy_cdouble c;
  npy_csetreal(&c, 1.0);
  npy_csetimag(&c, 0.0);
  printf("%d + %di\n", npy_creal(c), npy_cimag(c));

• To ease cross-version compatibility, equivalent macros and a
  compatibility layer have been added which can be used by downstream
  packages to continue to support both NumPy 1.x and 2.x. See
  complex-numbers for more info.

• numpy/npy_common.h now includes complex.h, which means that
  complex is now a reserved keyword.

(gh-24085)

iso_c_binding support and improved common blocks for f2py

Previously, users would have to define their own custom f2cmap file to
use type mappings defined by the Fortran2003 iso_c_binding intrinsic
module. These type maps are now natively supported by f2py

(gh-24555)

f2py now handles common blocks which have kind specifications from
modules. This further expands the usability of intrinsics like
iso_fortran_env and iso_c_binding.

(gh-25186)

Call str automatically on third argument to functions like assert_equal

The third argument to functions like
numpy.testing.assert_equal now has str called on it
automatically. This way it mimics the built-in assert statement, where
assert_equal(a, b, obj) works like assert a == b, obj.

(gh-24877)

Support for array-like atol/rtol in isclose, allclose

The keywords atol and rtol in numpy.isclose and
numpy.allclose now accept both scalars and arrays. An
array, if given, must broadcast to the shapes of the first two array
arguments.

(gh-24878)

Consistent failure messages in test functions

Previously, some numpy.testing assertions printed messages
that referred to the actual and desired results as x and y. Now,
these values are consistently referred to as ACTUAL and DESIRED.

(gh-24931)

n-D FFT transforms allow s[i] == -1

The numpy.fft.fftn, numpy.fft.ifftn,
numpy.fft.rfftn, numpy.fft.irfftn,
numpy.fft.fft2, numpy.fft.ifft2,
numpy.fft.rfft2 and numpy.fft.irfft2
functions now use the whole input array along the axis i if
s[i] == -1, in line with the array API standard.

(gh-25495)

Guard PyArrayScalar_VAL and PyUnicodeScalarObject for the limited API

PyUnicodeScalarObject holds a PyUnicodeObject, which is not
available when using Py_LIMITED_API. Add guards to hide it and
consequently also make the PyArrayScalar_VAL macro hidden.

(gh-25531)

Changes

• np.gradient() now returns a tuple rather than a list making the
  return value immutable.

  (gh-23861)

• Being fully context and thread-safe, np.errstate can only be
  entered once now.

• np.setbufsize is now tied to np.errstate(): leaving an
  np.errstate context will also reset the bufsize.

  (gh-23936)

• A new public np.lib.array_utils submodule has been introduced and
  it currently contains three functions: byte_bounds (moved from
  np.lib.utils), normalize_axis_tuple and normalize_axis_index.

  (gh-24540)

• Introduce numpy.bool as the new canonical name for
  NumPy's boolean dtype, and make numpy.bool\_ an alias
  to it. Note that until NumPy 1.24, np.bool was an alias to
  Python's builtin bool. The new name helps with array API standard
  compatibility and is a more intuitive name.

  (gh-25080)

• The dtype.flags value was previously stored as a signed integer.
  This means that the aligned dtype struct flag lead to negative flags
  being set (-128 rather than 128). This flag is now stored unsigned
  (positive). Code which checks flags manually may need to adapt. This
  may include code compiled with Cython 0.29.x.

  (gh-25816)

Representation of NumPy scalars changed

As per NEP 51, the scalar representation has been updated to include the type
information to avoid confusion with Python scalars.

Scalars are now printed as np.float64(3.0) rather than just 3.0.
This may disrupt workflows that store representations of numbers (e.g.,
to files) making it harder to read them. They should be stored as
explicit strings, for example by using str() or f"{scalar!s}". For
the time being, affected users can use
np.set_printoptions(legacy="1.25") to get the old behavior (with
possibly a few exceptions). Documentation of downstream projects may
require larger updates, if code snippets are tested. We are working on
tooling for
doctest-plus
to facilitate updates.

(gh-22449)

Truthiness of NumPy strings changed

NumPy strings previously were inconsistent about how they defined if the
string is True or False and the definition did not match the one
used by Python. Strings are now considered True when they are
non-empty and False when they are empty. This changes the following
distinct cases:

• Casts from string to boolean were previously roughly equivalent to
  string_array.astype(np.int64).astype(bool), meaning that only
  valid integers could be cast. Now a string of "0" will be
  considered True since it is not empty. If you need the old
  behavior, you may use the above step (casting to integer first) or
  string_array == "0" (if the input is only ever 0 or 1). To get
  the new result on old NumPy versions use string_array != "".
• np.nonzero(string_array) previously ignored whitespace so that a
  string only containing whitespace was considered False. Whitespace
  is now considered True.

This change does not affect np.loadtxt, np.fromstring, or
np.genfromtxt. The first two still use the integer definition, while
genfromtxt continues to match for "true" (ignoring case). However,
if np.bool_ is used as a converter the result will change.

The change does affect np.fromregex as it uses direct assignments.

(gh-23871)

A mean keyword was added to var and std function

Often when the standard deviation is needed the mean is also needed. The
same holds for the variance and the mean. Until now the mean is then
calculated twice, the change introduced here for the numpy.var and
numpy.std functions allows for passing in a precalculated mean as an keyword
argument. See the docstrings for details and an example illustrating the
speed-up.

(gh-24126)

Remove datetime64 deprecation warning when constructing with timezone

The numpy.datetime64 method now issues a UserWarning rather than a
DeprecationWarning whenever a timezone is included in the datetime string that
is provided.

(gh-24193)

Default integer dtype is now 64-bit on 64-bit Windows

The default NumPy integer is now 64-bit on all 64-bit systems as the
historic 32-bit default on Windows was a common source of issues. Most
users should not notice this. The main issues may occur with code
interfacing with libraries written in a compiled language like C. For
more information see migration_windows_int64.

(gh-24224)

Renamed numpy.core to numpy._core

Accessing numpy.core now emits a DeprecationWarning. In practice we
have found that most downstream usage of numpy.core was to access
functionality that is available in the main numpy namespace. If for
some reason you are using functionality in numpy.core that is not
available in the main numpy namespace, this means you are likely using
private NumPy internals. You can still access these internals via
numpy._core without a deprecation warning but we do not provide any
backward compatibility guarantees for NumPy internals. Please open an
issue if you think a mistake was made and something needs to be made
public.

(gh-24634)

The "relaxed strides" debug build option, which was previously enabled
through the NPY_RELAXED_STRIDES_DEBUG environment variable or the
-Drelaxed-strides-debug config-settings flag has been removed.

(gh-24717)

Redefinition of np.intp/np.uintp (almost never a change)

Due to the actual use of these types almost always matching the use of
size_t/Py_ssize_t this is now the definition in C. Previously, it
matched intptr_t and uintptr_t which would often have been subtly
incorrect. This has no effect on the vast majority of machines since the
size of these types only differ on extremely niche platforms.

However, it means that:

• Pointers may not necessarily fit into an intp typed array anymore.
  The p and P character codes can still be used, however.
• Creating intptr_t or uintptr_t typed arrays in C remains
  possible in a cross-platform way via PyArray_DescrFromType('p').
• The new character codes nN were introduced.
• It is now correct to use the Python C-API functions when parsing to
  npy_intp typed arguments.

(gh-24888)

numpy.fft.helper made private

numpy.fft.helper was renamed to numpy.fft._helper to indicate that
it is a private submodule. All public functions exported by it should be
accessed from numpy.fft.

(gh-24945)

numpy.linalg.linalg made private

numpy.linalg.linalg was renamed to numpy.linalg._linalg to indicate
that it is a private submodule. All public functions exported by it
should be accessed from numpy.linalg.

(gh-24946)

Out-of-bound axis not the same as axis=None

In some cases axis=32 or for concatenate any large value was the same
as axis=None. Except for concatenate this was deprecate. Any out of
bound axis value will now error, make sure to use axis=None.

(gh-25149)

New copy keyword meaning for array and asarray constructors

Now numpy.array and numpy.asarray support
three values for copy parameter:

• None - A copy will only be made if it is necessary.
• True - Always make a copy.
• False - Never make a copy. If a copy is required a ValueError is
  raised.

The meaning of False changed as it now raises an exception if a copy
is needed.

(gh-25168)

The __array__ special method now takes a copy keyword argument.

NumPy will pass copy to the __array__ special method in situations
where it would be set to a non-default value (e.g. in a call to
np.asarray(some_object, copy=False)). Currently, if an unexpected
keyword argument error is raised after this, NumPy will print a warning
and re-try without the copy keyword argument. Implementations of
objects implementing the __array__ protocol should accept a copy
keyword argument with the same meaning as when passed to
numpy.array or numpy.asarray.

(gh-25168)

Cleanup of initialization of numpy.dtype with strings with commas

The interpretation of strings with commas is changed slightly, in that a
trailing comma will now always create a structured dtype. E.g., where
previously `np.d

---

Configuration

📅 Schedule: Branch creation - "on Saturday" in timezone Europe/Berlin, Automerge - At any time (no schedule defined).

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR has been generated by Mend Renovate. View repository job log here.

[gygAlexWeiss]

LGTM