+
XjA ^ RI t ^ RIt^ RIt^ RIt^ RIt^ RIHt ^ RIH t . ROt
] ! R4 R 4 t ] ! R4 R 4 t R t
R tRsRsRR ltRR
lt] ! R4 RR l4 tR tR
tR tR# ) N)ndarray)
set_modulenumpyc D ^ RI H p ^ RIHpHpHp R\
P R\ P R\ P ! 4 /.p. . reV F5 pW7, ' d VP V4 K$ VP V4 K7 VP RRVRVR V//4 VP R
\
P P P R 4 ' * /4 ^ R IHp VP# V! 4 4 V ! V4 R # \$ d \' R
4 L"i ; i)a
Print information about various resources in the system
including available intrinsic support and BLAS/LAPACK library
in use
.. versionadded:: 1.24.0
See Also
--------
show_config : Show libraries in the system on which NumPy was built.
Notes
-----
1. Information is derived with the help of `threadpoolctl `_
library if available.
2. SIMD related information is derived from ``__cpu_features__``,
``__cpu_baseline__`` and ``__cpu_dispatch__``
)pprint__cpu_baseline____cpu_dispatch____cpu_features__
numpy_versionpythonunamesimd_extensionsbaselinefound not_found&ignore_floating_point_errors_in_matmulN)threadpool_infozWARNING: `threadpoolctl` not found in system! Install it by `pip install threadpoolctl`. Once installed, try `np.show_runtime` again for more detailed build information)r numpy._core._multiarray_umathr r
r np __version__sysversionplatformr append_core_multiarray_umath_blas_supports_fpe
threadpoolctlr extend ImportErrorprint) r r r
r config_foundfeatures_foundfeatures_not_foundfeaturer s V/home/luca/screen-checking/.venv/lib/python3.14/site-packages/numpy/lib/_utils_impl.py show_runtimer( s * #++ !
L
*,R&# $ $ ! !' * % %g . $
( ^ +
0**==dC C
61O -. < 6
5 66s D D D c X ^ RI p V P fK \ P P \ P P V P 4 RR4 pV# ^ RIHp \ P P \ P P VP 4 R4 pV# )ac
Return the directory that contains the NumPy \*.h header files.
Extension modules that need to compile against NumPy may need to use this
function to locate the appropriate include directory.
Notes
-----
When using ``setuptools``, for example in ``setup.py``::
import numpy as np
...
Extension('extension_name', ...
include_dirs=[np.get_include()])
...
Note that a CLI tool ``numpy-config`` was introduced in NumPy 2.0, using
that is likely preferred for build systems other than ``setuptools``::
$ numpy-config --cflags
-I/path/to/site-packages/numpy/_core/include
# Or rely on pkg-config:
$ export PKG_CONFIG_PATH=$(numpy-config --pkgconfigdir)
$ pkg-config --cflags
-I/path/to/site-packages/numpy/_core/include
Examples
--------
>>> np.get_include()
'.../site-packages/numpy/core/include' # may vary
Nr include) r show_configospathjoindirname__file__ numpy._corer )r dr s r' get_includer3 M sq F GGLL 8'9 M
H $ GGLL 8) D H c \ P pV FB p\ VP 4 4 pV' g K& \ V\ V4 V,
4 pKD V\ P 8X d ^ pV# )zM
Determines the leading whitespace that could be removed from all the lines.
)r maxsizelenlstripmin)linesindentlinecontents & r' _get_indentr> { sY [[Fdkkm$ 7TW!45F Mr4 c ` \ V 4 pTpT pR pVP V4 pV F pWC8X d Rp MTp V\ V4 , \ V 4 , pWB8 dA V^, \ V4 , pVR, RV^, , , V, pKv WY, V, pK V# )z, z,
)r7 split)
name argumentswidth
firstwidthknewstrsepstrarglistargumentaddstrs
&&& r' _split_linerM s TJA
F
Foof%G ?FF
H
F + 9QX .Ae^cZ!^&<>!$ A(+u'7'788"+..* &k 2 2G OOG, %%g.'.G$ !r4 c " RpRpR p\ V R\ V 4 4 p\ VRV4 pV P pV P P pVf \
P p\ RWaR7 \ RV P VR7 \ R WqR7 \ R
V P VR7 \ R V! V P P 4 VR7 \ R V! V P P 4 VR7 \ R
V P P VR7 \ R\ V P P P" 4 V 2VR7 \ RRVR7 VR9 d% \ V \
P V 2VR7 Rp MRVR8X d' \ V RV 2VR7 \
P R8g p M%\ V RV 2VR7 \
P R8g p \ RV! V 4 VR7 \ RV P 2VR7 R# )a Provide information about ndarray obj.
Parameters
----------
obj : ndarray
Must be ndarray, not checked.
output
Where printed output goes.
Notes
-----
Copied over from the numarray module prior to its removal.
Adapted somewhat as only numpy is an option now.
Called by info.
r@ c V # N )r^ s &r' _info.. s 1r4 __class__rR Nzclass: filezshape: z strides: z
itemsize: z aligned: z contiguous: z fortran: zdata pointer: z byteorder: rA )endrj F>biglittlez
byteswap: ztype: )|=)getattrtypestridesdtype byteorderr stdoutr" shapeitemsizeflagsaligned
contiguousfortranhexctypes_as_parameter_value)
objoutputextraticbpclsnmrs endianbyteswaps
&& r' _infor s $
E
C B
#{DI
.C j# &BkkG
YY
F
~ )R% )SYYV, +w, , 62 +r#))++,6: ."SYY 1 12@ +syy((v6
SZZ 6 6 < <=>ugF
-Sv.
cmm_SE*8 3
S V,==E)
VC5!/==H, ,8 62 F399+
V,r4 c ^ RI p^ RIp\ V R4 ' g \ V R4 ' d V P p M \ V R4 ' d
V P p Vf \
P pV f \ \ 4 R# \ V \ 4 ' d \ WR7 R# \ V \ 4 ' d \ f \ V4 w s s^ p. p\ F p \ V, V , p \ V 4 V9 d \! RV R2VR7 MJVP# \ V 4 4 \! R V R
2VR7 \ V 4 \! R V, VR7 V^,
pK V^ 8X d \! R V R
2VR7 R# \! RV R2VR7 R# VP' V 4 ' g VP) V 4 ' d V P* p
\ VP- V 4 4 p \1 W, 4 V8 d \3 WV4 p MW, p \! RV , R, VR7 \! VP5 V 4 VR7 R# VP7 V 4 ' Ed{ V P* p
\ VP- V 4 4 p \1 W, 4 V8 d \3 WV4 p MW, p \! RV , R, VR7 VP5 V 4 p
V
f: \ V R4 ' d' \! VP5 V P8 4 VR7 M \! VP5 V 4 VR7 VP; V 4 pV Uu. uF q^ , R8w g K VNK ppV' di \! RVR7 V FS p\= WR4 pVe. TP? VP5 V4 ;' g R4 w pp\! RV RX 2VR7 KU R# R# \ V R4 ' d \! VP5 V 4 VR7 R# R# \$ d EKJ i ; i \. d Rp ELKi ; i \. d Rp ELi ; iu upi )a
Get help information for an array, function, class, or module.
Parameters
----------
object : object or str, optional
Input object or name to get information about. If `object` is
an `ndarray` instance, information about the array is printed.
If `object` is a numpy object, its docstring is given. If it is
a string, available modules are searched for matching objects.
If None, information about `info` itself is returned.
maxwidth : int, optional
Printing width.
output : file like object, optional
File like object that the output is written to, default is
``None``, in which case ``sys.stdout`` will be used.
The object has to be opened in 'w' or 'a' mode.
toplevel : str, optional
Start search at this level.
Notes
-----
When used interactively with an object, ``np.info(obj)`` is equivalent
to ``help(obj)`` on the Python prompt or ``obj?`` on the IPython
prompt.
Examples
--------
>>> np.info(np.polyval) # doctest: +SKIP
polyval(p, x)
Evaluate the polynomial p at x.
...
When using a string for `object` it is possible to get multiple results.
>>> np.info('fft') # doctest: +SKIP
*** Found in numpy ***
Core FFT routines
...
*** Found in numpy.fft ***
fft(a, n=None, axis=-1)
...
*** Repeat reference found in numpy.fft.fftpack ***
*** Total of 3 references found. ***
When the argument is an array, information about the array is printed.
>>> a = np.array([[1 + 2j, 3, -4], [-5j, 6, 0]], dtype=np.complex64)
>>> np.info(a)
class: ndarray
shape: (2, 3)
strides: (24, 8)
itemsize: 8
aligned: True
contiguous: True
fortran: False
data pointer: 0x562b6e0d2860 # may vary
byteorder: little
byteswap: False
type: complex64
N_ppimport_importer_ppimport_module_ppimport_attr)r z$
*** Repeat reference found in z *** ri z *** Found in z ***-z Help for z not found.z
*** Total of z references found. ***z()rA
__init___z
Methods:
Nonez z -- __doc__) inspectpydochasattrr r r rv inforV r r str _namedictra _dictlistidr" r KeyError
isfunctionismethodrR signature Exceptionr7 rM getdocisclassr
allmethodsrq splitdoc)objectmaxwidthr toplevelr r numfoundobjlistnamestrr rC rD argstrdoc1methodsmethpublic_methodsthisobjmethstrothers &&&& r' r r s D ,-- 6- . .(( ) * *&&
~
~ T
FG $ $
f$ FC #0#: Iy G
(0c7g% A'%P % NN2c7+ .wit<6JI#.v6A
! q= IfX[1 ? ""*+AC
F # #w'7'7'?'? G - -f 56I t 8 + (;F%F
cFlT!/
gnnV$62 G - -f 56I t 8 + (;F%F
cFlT!/~~f% <vz**gnnV__5FC '..(v 6""6*+2E741gn$$7E " 0&!&5&%*^^ #NN7 3 = =v& NGU 4&wi0v>
'
# #
gnnV$62
$u
I I $ FsC B PP) 7P= Q*Q P&%P&)
P:9P:=
Q
Qc V P ^ 8X d V# V P RVR7 p\ P ! V4 p\ P P V4 ' d VP
R4 pVP 4 ' g V# \ V\ P 4 ' d V# \ P ! WVR7 V# )a7
Utility function to check median result from data for NaN values at the end
and return NaN in that case. Input result can also be a MaskedArray.
Parameters
----------
data : array
Sorted input data to median function
result : Array or MaskedArray
Result of median function.
axis : int
Axis along which the median was computed.
Returns
-------
result : scalar or ndarray
Median or NaN in axes which contained NaN in the input. If the input
was an array, NaN will be inserted in-place. If a scalar, either the
input itself or a scalar NaN.
)axisF)where) sizetaker isnanma
isMaskedArrayfilledanyrV genericcopyto)dataresultr potential_nansns &&& r' _median_nancheckr s * yyA~
YYrY-N
A
uu1
HHUO 5577
&"**%% IIfA. Mr4 c ^ RI Hp HpHp \ V 4 ^ 8X d \ V4 ^ 8X d R# RP V 4 pV F- pW$, ' d VRV R2,
pK VRV R2,
pK/ V# )a
Returns a string containing the CPU features supported
by the current build.
The format of the string can be explained as follows:
- Dispatched features supported by the running machine end with `*`.
- Dispatched features not supported by the running machine
end with `?`.
- Remaining features represent the baseline.
Returns:
str: A formatted string indicating the supported CPU features.
r r@ rA *?)r r r
r r7 r. )r r
r enabled_featuresr& s r' _opt_infor s} !c*:&;q&@xx 01# $ $ !G9A . !G9A . $ r4 c
V P e V P RJp. p. p. p. pV P P 4 F w rg\ V^ , 4 pW^ , Jd RpVP V4 VP V4 VP V^, 4 TP \ V4 ^8 d RMV^, 4 K V' g V # RVRVRVRVRV P /p \ P ! WP R7 # V P eF V P w r\ V
4 p V P f WJ d V # \ P ! W34 # V P f V # \ P ! V P 4 # ) a
Returns the dtype unchanged if it contained no metadata or a copy of the
dtype if it (or any of its structure dtypes) contained metadata.
This utility is used by `np.save` and `np.savez` to drop metadata before
saving.
.. note::
Due to its limitation this function may move to a more appropriate
home or change in the future and is considered semi-public API only.
.. warning::
This function does not preserve more strange things like record dtypes
and user dtypes may simply return the wrong thing. If you need to be
sure about the latter, check the result with:
``np.can_cast(new_dtype, dtype, casting="no")``.
NTnamesformatsoffsetstitlesrx )align) fieldsmetadataitems
drop_metadatar r7 rx r rt isalignedstructsubdtyper )
rt found_metadatar r r r rC fieldfield_dt structurer rw new_subdtypes
" r' r r sW *
|| t3 <<--/KD$U1X.HQx'!% LL NN8 $ NN58 $ MM#e*q.$eAh ? 0 L
UIw 7Hf (
xx )>)>?? # ..$X. >> !l&>Lxx -.. >> !Lxx ""r4 )r3 r r( )r rd )NL Nr )r, r r rW r r r1 r numpy._utilsr __all__r( r3 r> rM r r ra r r r r r re r4 r' r s
#
G: :z G*
*
Z .(
(3-l G^3 ^3B(T @<#r4