NumPy 1.23.0 Release Notes#
The NumPy 1.23.0 release continues the ongoing work to improve the handling and promotion of dtypes, increase the execution speed, clarify the documentation, and expire old deprecations. The highlights are:
loadtxtin C, greatly improving its performance.
Exposing DLPack at the Python level for easy data exchange.
Changes to the promotion and comparisons of structured dtypes.
Improvements to f2py.
See below for the details,
A masked array specialization of
ndenumerateis now available as
numpy.ma.ndenumerate. It provides an alternative to
numpy.ndenumerateand skips masked values by default.
numpy.from_dlpackhas been added to allow easy exchange of data using the DLPack protocol. It accepts Python objects that implement the
__dlpack_device__methods and returns a ndarray object which is generally the view of the data of the input object.
Noneis deprecated. It must now be a method and may wish to call
super().__array_finalize__(obj)after checking for
Noneor if the NumPy version is sufficiently new.
axis=np.MAXDIMS) in many cases had the same meaning as
axis=None. This is deprecated and
axis=Nonemust be used instead.
The hook function
PyDataMem_SetEventHookhas been deprecated and the demonstration of its use in tool/allocation_tracking has been removed. The ability to track allocations is now built-in to python via
numpy.distutilshas been deprecated, as a result of
distutilsitself being deprecated. It will not be present in NumPy for Python >= 3.12, and will be removed completely 2 years after the release of Python 3.12 For more details, see Status of numpy.distutils and migration advice.
numpy.loadtxtwill now give a
DeprecationWarningwhen an integer
dtypeis requested but the value is formatted as a floating point number.
NpzFile.iterkeys()methods have been removed as part of the continued removal of Python 2 compatibility. This concludes the deprecation from 1.15.
asscalarfunctions have been removed.
UPDATEIFCOPYarray flag has been removed together with the enum
NPY_ARRAY_UPDATEIFCOPY. The associated (and deprecated)
PyArray_XDECREF_ERRwas also removed. These were all deprecated in 1.14. They are replaced by
NPY_ARRAY_WRITEBACKIFCOPY, that requires calling
PyArray_ResolveWritebackIfCopybefore the array is deallocated.
Exceptions will be raised during array-like creation. When an object raised an exception during access of the special attributes
__array_interface__, this exception was usually ignored. This behaviour was deprecated in 1.21, and the exception will now be raised.
Multidimensional indexing with non-tuple values is not allowed. Previously, code such as
ind = [[0, 1], [0, 1]]produced a
FutureWarningand was interpreted as a multidimensional index (i.e.,
arr[tuple(ind)]). Now this example is treated like an array index over a single dimension (
arr[array(ind)]). Multidimensional indexing with anything but a tuple was deprecated in NumPy 1.15.
Changing to a dtype of different size in F-contiguous arrays is no longer permitted. Deprecated since Numpy 1.11.0. See below for an extended explanation of the effects of this change.
crackfortran has support for operator and assignment overloading#
crackfortran parser now understands operator and assignment
definitions in a module. They are added in the
body list of the
module which contains a new key
implementedby listing the names
of the subroutines or functions implementing the operator or
f2py supports reading access type attributes from derived type statements#
As a result, one does not need to use
private statements to
specify derived type access properties.
ndmin added to
This parameter behaves the same as
np.loadtxt now supports quote character and single converter function#
numpy.loadtxt now supports an additional
quotechar keyword argument
which is not set by default. Using
quotechar='"' will read quoted fields
as used by the Excel CSV dialect.
Further, it is now possible to pass a single callable rather than a dictionary
Changing to dtype of a different size now requires contiguity of only the last axis#
Previously, viewing an array with a dtype of a different item size required that the entire array be C-contiguous. This limitation would unnecessarily force the user to make contiguous copies of non-contiguous arrays before being able to change the dtype.
This change affects not only
ndarray.view, but other construction
mechanisms, including the discouraged direct assignment to
This change expires the deprecation regarding the viewing of F-contiguous arrays, described elsewhere in the release notes.
Deterministic output files for F2PY#
For F77 inputs,
f2py will generate
unconditionally, though these may be empty. For free-form inputs,
modname-f2pywrappers2.f90 will both be generated
unconditionally, and may be empty. This allows writing generic output rules in
meson and other build systems. Older behavior can be restored
f2py. Using via meson details usage.
keepdims parameter for
keepdims was added to the functions
numpy.ma.average. The parameter has the same meaning as it
does in reduction functions such as
equal_nan added to
np.unique was changed in 1.21 to treat all
NaN values as equal and return
equal_nan=False will restore pre-1.21 behavior
NaNs as unique. Defaults to
np.linalg.norm preserves float input types, even for scalar results#
Previously, this would promote to
float64 when the
ord argument was
not one of the explicitly listed values, e.g.
>>> f32 = np.float32([1, 2]) >>> np.linalg.norm(f32, 2).dtype dtype('float32') >>> np.linalg.norm(f32, 3) dtype('float64') # numpy 1.22 dtype('float32') # numpy 1.23
This change affects only
float16 vectors with
Changes to structured (void) dtype promotion and comparisons#
In general, NumPy now defines correct, but slightly limited, promotion for structured dtypes by promoting the subtypes of each field instead of raising an exception:
>>> np.result_type(np.dtype("i,i"), np.dtype("i,d")) dtype([('f0', '<i4'), ('f1', '<f8')])
For promotion matching field names, order, and titles are enforced, however
padding is ignored.
Promotion involving structured dtypes now always ensures native byte-order for
all fields (which may change the result of
and ensures that the result will be “packed”, i.e. all fields are ordered
contiguously and padding is removed.
See Structure Comparison and Promotion for further details.
repr of aligned structures will now never print the long form including
itemsize unless the structure includes padding not
In alignment with the above changes to the promotion logic, the casting safety has been updated:
"equiv"enforces matching names and titles. The itemsize is allowed to differ due to padding.
"safe"allows mismatching field names and titles
The cast safety is limited by the cast safety of each included field.
The order of fields is used to decide cast safety of each individual field. Previously, the field names were used and only unsafe casts were possible when names mismatched.
The main important change here is that name mismatches are now considered “safe” casts.
NPY_RELAXED_STRIDES_CHECKING has been removed#
NumPy cannot be compiled with
anymore. Relaxed strides have been the default for many years and
the option was initially introduced to allow a smoother transition.
np.loadtxt has recieved several changes#
The row counting of
numpy.loadtxt was fixed.
loadtxt ignores fully
empty lines in the file, but counted them towards
max_rows is used and the file contains empty lines, these will now
not be counted. Previously, it was possible that the result contained fewer
max_rows rows even though more data was available to be read.
If the old behaviour is required,
itertools.islice may be used:
import itertools lines = itertools.islice(open("file"), 0, max_rows) result = np.loadtxt(lines, ...)
While generally much faster and improved,
numpy.loadtxt may now fail to
converter certain strings to numbers that were previously successfully read.
The most important cases for this are:
Parsing floating point values such as
1.0into integers is now deprecated.
Parsing hexadecimal floats such as
_was previously accepted as a thousands delimiter
100_000. This will now result in an error.
If you experience these limitations, they can all be worked around by passing
converters=. NumPy now supports passing a single converter
to be used for all columns to make this more convenient.
converters=float.fromhex can read hexadecimal float numbers
converters=int will be able to read
Further, the error messages have been generally improved. However, this means
that error types may differ. In particularly, a
ValueError is now always
raised when parsing of a single entry fails.
ndarray.__array_finalize__ is now callable#
This means subclasses can now use
without worrying whether
ndarray is their superclass or not.
The actual call remains a no-op.
Add support for VSX4/Power10#
With VSX4/Power10 enablement, the new instructions available in Power ISA 3.1 can be used to accelerate some NumPy operations, e.g., floor_divide, modulo, etc.
np.fromiter now accepts objects and subarrays#
numpy.fromiter function now supports object and
subarray dtypes. Please see he function documentation for
Math C library feature detection now uses correct signatures#
Compiling is preceded by a detection phase to determine whether the
underlying libc supports certain math operations. Previously this code
did not respect the proper signatures. Fixing this enables compilation
wasm-ld backend (compilation for web assembly) and reduces
the number of warnings.
np.kron now maintains subclass information#
np.kron maintains subclass information now such as masked arrays
while computing the Kronecker product of the inputs
>>> x = ma.array([[1, 2], [3, 4]], mask=[[0, 1], [1, 0]]) >>> np.kron(x,x) masked_array( data=[[1, --, --, --], [--, 4, --, --], [--, --, 4, --], [--, --, --, 16]], mask=[[False, True, True, True], [ True, False, True, True], [ True, True, False, True], [ True, True, True, False]], fill_value=999999)
np.kron output now follows
ufunc ordering (
to determine the output class type
>>> class myarr(np.ndarray): >>> __array_priority__ = -1 >>> a = np.ones([2, 2]) >>> ma = myarray(a.shape, a.dtype, a.data) >>> type(np.kron(a, ma)) == np.ndarray False # Before it was True >>> type(np.kron(a, ma)) == myarr True
Performance improvements and changes#
numpy.loadtxt is now generally much faster than previously as most of it
is now implemented in C.
Faster reduction operators#
Reduction operations like
numpy.logical_and.reduce on contiguous integer-based arrays are now
numpy.where is now much faster than previously on unpredictable/random
Faster operations on NumPy scalars#
Many operations on NumPy scalars are now significantly faster, although
rare operations (e.g. with 0-D arrays rather than scalars) may be slower
in some cases.
However, even with these improvements users who want the best performance
for their scalars, may want to convert a known NumPy scalar into a Python
numpy.kron is about 80% faster as the product is now computed