SciPy 1.14.0 Release Notes

SciPy 1.14.0 is the culmination of 3 months of hard work. It contains
many new features, numerous bug-fixes, improved test coverage and better
documentation. There have been a number of deprecations and API changes
in this release, which are documented below. All users are encouraged to
upgrade to this release, as there are a large number of bug-fixes and
optimizations. Before upgrading, we recommend that users check that
their own code does not use deprecated SciPy functionality (to do so,
run your code with python -Wd and check for DeprecationWarning s).
Our development attention will now shift to bug-fix releases on the
1.14.x branch, and on adding new features on the main branch.

This release requires Python 3.10+ and NumPy 1.23.5 or greater.

For running on PyPy, PyPy3 6.0+ is required.

Highlights of this release

• SciPy now supports the new Accelerate library introduced in macOS 13.3, and
  has wheels built against Accelerate for macOS >=14 resulting in significant
  performance improvements for many linear algebra operations.
• A new method, cobyqa, has been added to scipy.optimize.minimize - this
  is an interface for COBYQA (Constrained Optimization BY Quadratic
  Approximations), a derivative-free optimization solver, designed to
  supersede COBYLA, developed by the Department of Applied Mathematics, The
  Hong Kong Polytechnic University.
• scipy.sparse.linalg.spsolve_triangular is now more than an order of
  magnitude faster in many cases.

New features

scipy.fft improvements

• A new function, scipy.fft.prev_fast_len, has been added. This function
  finds the largest composite of FFT radices that is less than the target
  length. It is useful for discarding a minimal number of samples before FFT.

scipy.io improvements

• wavfile now supports reading and writing of wav files in the RF64
  format, allowing files greater than 4 GB in size to be handled.

scipy.constants improvements

• Experimental support for the array API standard has been added.

scipy.interpolate improvements

• scipy.interpolate.Akima1DInterpolator now supports extrapolation via the
  extrapolate argument.

scipy.optimize improvements

• scipy.optimize.HessianUpdateStrategy now also accepts square arrays for
  init_scale.
• A new method, cobyqa, has been added to scipy.optimize.minimize - this
  is an interface for COBYQA (Constrained Optimization BY Quadratic
  Approximations), a derivative-free optimization solver, designed to
  supersede COBYLA, developed by the Department of Applied Mathematics, The
  Hong Kong Polytechnic University.
• There are some performance improvements in
  scipy.optimize.differential_evolution.
• scipy.optimize.approx_fprime now has linear space complexity.

scipy.signal improvements

• scipy.signal.minimum_phase has a new argument half, allowing the
  provision of a filter of the same length as the linear-phase FIR filter
  coefficients and with the same magnitude spectrum.

scipy.sparse improvements

• Sparse arrays now support 1D shapes in COO, DOK and CSR formats.
  These are all the formats we currently intend to support 1D shapes.
  Other sparse array formats raise an exception for 1D input.
• Sparse array methods min/nanmin/argmin and max analogs now return 1D arrays.
  Results are still COO format sparse arrays for min/nanmin and
  dense np.ndarray for argmin.
• Sparse matrix and array objects improve their repr and str output.
• A special case has been added to handle multiplying a dia_array by a
  scalar, which avoids a potentially costly conversion to CSR format.
• scipy.sparse.csgraph.yen has been added, allowing usage of Yen's K-Shortest
  Paths algorithm on a directed on undirected graph.
• Addition between DIA-format sparse arrays and matrices is now faster.
• scipy.sparse.linalg.spsolve_triangular is now more than an order of
  magnitude faster in many cases.

scipy.spatial improvements

• Rotation supports an alternative "scalar-first" convention of quaternion
  component ordering. It is available via the keyword argument scalar_first
  of from_quat and as_quat methods.
• Some minor performance improvements for inverting of Rotation objects.

scipy.special improvements

• Added scipy.special.log_wright_bessel, for calculation of the logarithm of
  Wright's Bessel function.
• The relative error in scipy.special.hyp2f1 calculations has improved
  substantially.
• Improved behavior of boxcox, inv_boxcox, boxcox1p, and
  inv_boxcox1p by preventing premature overflow.

scipy.stats improvements

• A new function scipy.stats.power can be used for simulating the power
  of a hypothesis test with respect to a specified alternative.
• The Irwin-Hall (AKA Uniform Sum) distribution has been added as
  scipy.stats.irwinhall.
• Exact p-value calculations of scipy.stats.mannwhitneyu are much faster
  and use less memory.
• scipy.stats.pearsonr now accepts n-D arrays and computes the statistic
  along a specified axis.
• scipy.stats.kstat, scipy.stats.kstatvar, and scipy.stats.bartlett
  are faster at performing calculations along an axis of a large n-D array.

Array API Standard Support

Experimental support for array libraries other than NumPy has been added to
existing sub-packages in recent versions of SciPy. Please consider testing
these features by setting an environment variable SCIPY_ARRAY_API=1 and
providing PyTorch, JAX, or CuPy arrays as array arguments.

As of 1.14.0, there is support for

• scipy.cluster

• scipy.fft

• scipy.constants

• scipy.special: (select functions)

  • scipy.special.log_ndtr
  • scipy.special.ndtr
  • scipy.special.ndtri
  • scipy.special.erf
  • scipy.special.erfc
  • scipy.special.i0
  • scipy.special.i0e
  • scipy.special.i1
  • scipy.special.i1e
  • scipy.special.gammaln
  • scipy.special.gammainc
  • scipy.special.gammaincc
  • scipy.special.logit
  • scipy.special.expit
  • scipy.special.entr
  • scipy.special.rel_entr
  • scipy.special.xlogy
  • scipy.special.chdtrc

• scipy.stats: (select functions)

  • scipy.stats.describe
  • scipy.stats.moment
  • scipy.stats.skew
  • scipy.stats.kurtosis
  • scipy.stats.kstat
  • scipy.stats.kstatvar
  • scipy.stats.circmean
  • scipy.stats.circvar
  • scipy.stats.circstd
  • scipy.stats.entropy
  • scipy.stats.variation
  • scipy.stats.sem
  • scipy.stats.ttest_1samp
  • scipy.stats.pearsonr
  • scipy.stats.chisquare
  • scipy.stats.skewtest
  • scipy.stats.kurtosistest
  • scipy.stats.normaltest
  • scipy.stats.jarque_bera
  • scipy.stats.bartlett
  • scipy.stats.power_divergence
  • scipy.stats.monte_carlo_test

Deprecated features

• scipy.stats.gstd, scipy.stats.chisquare, and
  scipy.stats.power_divergence have deprecated support for masked array
  input.
• scipy.stats.linregress has deprecated support for specifying both samples
  in one argument; x and y are to be provided as separate arguments.
• The conjtransp method for scipy.sparse.dok_array and
  scipy.sparse.dok_matrix has been deprecated and will be removed in SciPy
  1.16.0.
• The option quadrature="trapz" in scipy.integrate.quad_vec has been
  deprecated in favour of quadrature="trapezoid" and will be removed in
  SciPy 1.16.0.
• scipy.special.{comb,perm} have deprecated support for use of exact=True in
  conjunction with non-integral N and/or k.

Backwards incompatible changes

• Many scipy.stats functions now produce a standardized warning message when
  an input sample is too small (e.g. zero size). Previously, these functions
  may have raised an error, emitted one or more less informative warnings, or
  emitted no warnings. In most cases, returned results are unchanged; in almost
  all cases the correct result is NaN.

Expired deprecations

There is an ongoing effort to follow through on long-standing deprecations.
The following previously deprecated features are affected:

• Several previously deprecated methods for sparse arrays were removed:
  asfptype, getrow, getcol, get_shape, getmaxprint,
  set_shape, getnnz, and getformat. Additionally, the .A and
  .H attributes were removed.

• scipy.integrate.{simps,trapz,cumtrapz} have been removed in favour of
  simpson, trapezoid, and cumulative_trapezoid.

• The tol argument of scipy.sparse.linalg.{bcg,bicstab,cg,cgs,gcrotmk, mres,lgmres,minres,qmr,tfqmr} has been removed in favour of rtol.
  Furthermore, the default value of atol for these functions has changed
  to 0.0.

• The restrt argument of scipy.sparse.linalg.gmres has been removed in
  favour of restart.

• The initial_lexsort argument of scipy.stats.kendalltau has been
  removed.

• The cond and rcond arguments of scipy.linalg.pinv have been
  removed.

• The even argument of scipy.integrate.simpson has been removed.

• The turbo and eigvals arguments from scipy.linalg.{eigh,eigvalsh}
  have been removed.

• The legacy argument of scipy.special.comb has been removed.

• The hz/nyq argument of signal.{firls, firwin, firwin2, remez} has
  been removed.

• Objects that weren't part of the public interface but were accessible through
  deprecated submodules have been removed.

• float128, float96, and object arrays now raise an error in
  scipy.signal.medfilt and scipy.signal.order_filter.

• scipy.interpolate.interp2d has been replaced by an empty stub (to be
  removed completely in the future).

• Coinciding with changes to function signatures (e.g. removal of a deprecated
  keyword), we had deprecated positional use of keyword arguments for the
  affected functions, which will now raise an error. Affected functions are:

  • sparse.linalg.{bicg, bicgstab, cg, cgs, gcrotmk, gmres, lgmres, minres, qmr, tfqmr}
  • stats.kendalltau
  • linalg.pinv
  • integrate.simpson
  • linalg.{eigh,eigvalsh}
  • special.comb
  • signal.{firls, firwin, firwin2, remez}

Other changes

• SciPy now uses C17 as the C standard to build with, instead of C99. The C++
  standard remains C++17.
• macOS Accelerate, which got a major upgrade in macOS 13.3, is now supported.
  This results in significant performance improvements for linear algebra
  operations, as well as smaller binary wheels.
• Cross-compilation should be smoother and QEMU or similar is no longer needed
  to run the cross interpreter.
• Experimental array API support for the JAX backend has been added to several
  parts of SciPy.

Authors

• Name (commits)
• h-vetinari (34)
• Steven Adams (1) +
• Max Aehle (1) +
• Ataf Fazledin Ahamed (2) +
• Luiz Eduardo Amaral (1) +
• Trinh Quoc Anh (1) +
• Miguel A. Batalla (7) +
• Tim Beyer (1) +
• Andrea Blengino (1) +
• boatwrong (1)
• Jake Bowhay (51)
• Dietrich Brunn (2)
• Evgeni Burovski (177)
• Tim Butters (7) +
• CJ Carey (5)
• Sean Cheah (46)
• Lucas Colley (73)
• Giuseppe "Peppe" Dilillo (1) +
• DWesl (2)
• Pieter Eendebak (5)
• Kenji S Emerson (1) +
• Jonas Eschle (1)
• fancidev (2)
• Anthony Frazier (1) +
• Ilan Gold (1) +
• Ralf Gommers (125)
• Rohit Goswami (28)
• Ben Greiner (1) +
• Lorenzo Gualniera (1) +
• Matt Haberland (260)
• Shawn Hsu (1) +
• Budjen Jovan (3) +
• Jozsef Kutas (1)
• Eric Larson (3)
• Gregory R. Lee (4)
• Philip Loche (1) +
• Christian Lorentzen (5)
• Sijo Valayakkad Manikandan (2) +
• marinelay (2) +
• Nikolay Mayorov (1)
• Nicholas McKibben (2)
• Melissa Weber Mendonça (7)
• João Mendes (1) +
• Samuel Le Meur-Diebolt (1) +
• Tomiță Militaru (2) +
• Andrew Nelson (35)
• Lysandros Nikolaou (1)
• Nick ODell (5) +
• Jacob Ogle (1) +
• Pearu Peterson (1)
• Matti Picus (5)
• Ilhan Polat (9)
• pwcnorthrop (3) +
• Bharat Raghunathan (1)
• Tom M. Ragonneau (2) +
• Tyler Reddy (101)
• Pamphile Roy (18)
• Atsushi Sakai (9)
• Daniel Schmitz (5)
• Julien Schueller (2) +
• Dan Schult (13)
• Tomer Sery (7)
• Scott Shambaugh (4)
• Tuhin Sharma (1) +
• Sheila-nk (4)
• Skylake (1) +
• Albert Steppi (215)
• Kai Striega (6)
• Zhibing Sun (2) +
• Nimish Telang (1) +
• toofooboo (1) +
• tpl2go (1) +
• Edgar Andrés Margffoy Tuay (44)
• Andrew Valentine (1)
• Valerix (1) +
• Christian Veenhuis (1)
• void (2) +
• Warren Weckesser (3)
• Xuefeng Xu (1)
• Rory Yorke (1)
• Xiao Yuan (1)
• Irwin Zaid (35)
• Elmar Zander (1) +
• Zaikun ZHANG (1)
• ਗਗਨਦੀਪ ਸਿੰਘ (Gagandeep Singh) (4) +

A total of 85 people contributed to this release.
People with a "+" by their names contributed a patch for the first time.
This list of names is automatically generated, and may not be fully complete.