Detailed SciPy Roadmap¶
Most of this roadmap is intended to provide a high-level view on what is most needed per SciPy submodule in terms of new functionality, bug fixes, etc. Besides important “business as usual” changes, it contains ideas for major new features - those are marked as such, and are expected to take significant dedicated effort. Things not mentioned in this roadmap are not necessarily unimportant or out of scope, however we (the SciPy developers) want to provide to our users and contributors a clear picture of where SciPy is going and where help is needed most.
This is the detailed roadmap. A very high-level overview with only the most important ideas is SciPy Roadmap.
This roadmap will be evolving together with SciPy. Updates can be submitted as pull requests. For large or disruptive changes you may want to discuss those first on the scipy-dev mailing list.
In general, we want to evolve the API to remove known warts as much as possible, however as much as possible without breaking backwards compatibility.
Also, it should be made (even) more clear what is public and what is private in SciPy. Everything private should be named starting with an underscore as much as possible.
Test coverage of code added in the last few years is quite good, and we aim for a high coverage for all new code that is added. However, there is still a significant amount of old code for which coverage is poor. Bringing that up to the current standard is probably not realistic, but we should plug the biggest holes.
Besides coverage there is also the issue of correctness - older code may have a
few tests that provide decent statement coverage, but that doesn’t necessarily
say much about whether the code does what it says on the box. Therefore code
review of some parts of the code (
particular) is necessary.
The documentation is in good shape. Expanding of current docstrings and putting them in the standard NumPy format should continue, so the number of reST errors and glitches in the html docs decreases. Most modules also have a tutorial in the reference guide that is a good introduction, however there are a few missing or incomplete tutorials - this should be fixed.
asv-based benchmark system is in reasonable shape. It is quite easy to
add new benchmarks, however running the benchmarks is not very intuitive.
Making this easier is a priority. In addition, we should run them in our CI
(gh-8779 is an ongoing attempt at this).
Use of Cython¶
Regarding Cython code:
It’s not clear how much functionality can be Cythonized without making the .so files too large. This needs measuring.
Cython’s old syntax for using NumPy arrays should be removed and replaced with Cython memoryviews.
Windows build issues¶
SciPy critically relies on Fortran code. This is still problematic on Windows. There are currently only two options: using Intel Fortran, or using MSVC + gfortran. The former is expensive, while the latter works (it’s what we use for releases) but is quite hard to do correctly. For allowing contributors and end users to reliably build SciPy on Windows, using the Flang compiler looks like the best way forward long-term.
Continuous integration is in good shape, it currently covers the Windows, macOS and Linux, ARM64 and ppc64le platforms, as well as a range of versions of our dependencies and building release quality wheels.
Size of binaries¶
SciPy binaries are quite large (e.g. an unzipped manylinux wheel for 1.4.1 is
91 MB), and this can be problematic - for example for use in AWS Lambda, which
has a 250 MB size limit. We aim to keep binary size as low as possible; when
adding new compiled extensions, this needs checking. Stripping of debug symbols
multibuild can likely be improved (see this issue).
This module is in good shape.
This module is basically done, low-maintenance and without open issues.
This module is in good shape.
Needed for ODE solvers:
Documentation is pretty bad, needs fixing
A new ODE solver interface (
solve_ivp) was added in SciPy 1.0.0. In the future we can consider (soft-)deprecating the older API.
The numerical integration functions are in good shape. Support for integrating complex-valued functions and integrating multiple intervals (see gh-3325) could be added.
Ideas for new features:
Spline fitting routines with better user control.
Transparent tensor-product splines.
Mesh refinement and coarsening of B-splines and corresponding tensor products.
PCM float will be supported, for anything else use
audiolabor other specialized libraries.
Raise errors instead of warnings if data not understood.
Other sub-modules (matlab, netcdf, idl, harwell-boeing, arff, matrix market) are in good shape.
scipy.linalg is in good shape.
Reduce duplication of functions with
numpy.linalg, make APIs consistent.
get_lapack_funcsshould always use
Wrap more LAPACK functions
One too many funcs for LU decomposition, remove one
Ideas for new features:
Add type-generic wrappers in the Cython BLAS and LAPACK
Make many of the linear algebra routines into gufuncs
BLAS and LAPACK
The Python and Cython interfaces to BLAS and LAPACK in
scipy.linalg are one
of the most important things that SciPy provides. In general
is in good shape, however we can make a number of improvements:
Library support. Our released wheels now ship with OpenBLAS, which is currently the only feasible performant option (ATLAS is too slow, MKL cannot be the default due to licensing issues, Accelerate support is dropped because Apple doesn’t update Accelerate anymore). OpenBLAS isn’t very stable though, sometimes its releases break things and it has issues with threading (currently the only issue for using SciPy with PyPy3). We need at the very least better support for debugging OpenBLAS issues, and better documentation on how to build SciPy with it. An option is to use BLIS for a BLAS interface (see numpy gh-7372).
Support for newer LAPACK features. In SciPy 1.2.0 we increased the minimum supported version of LAPACK to 3.4.0. Now that we dropped Python 2.7, we can increase that version further (MKL + Python 2.7 was the blocker for >3.4.0 previously) and start adding support for new features in LAPACK.
scipy.misc will be removed as a public module. Most functions in it have
been moved to another submodule or deprecated. The few that are left:
who: these are NumPy functions
central_diff_weight: remove, possibly replacing them with more extensive functionality for numerical differentiation.
ndimage is a powerful interpolation engine. Users come
with an expectation of one of two models: a pixel model with
1) elements having centers
(0.5, 0.5), or a data point model,
where values are defined at points on a grid. Over time, we’ve become
convinced that the data point model is better defined and easier to
implement, but this should be clearly communicated in the documentation.
More importantly, still, SciPy implements one variant of this data
point model, where datapoints at any two extremes of an axis share a
spatial location under periodic wrapping mode. E.g., in a 1D array,
you would have
x[-1] co-located. A very common
use-case, however, is for signals to be periodic, with equal spacing
between the first and last element along an axis (instead of zero
spacing). Wrapping modes for this use-case were added in
gh-8537, next the
interpolation routines should be updated to use those modes.
This should address several issues, including gh-1323, gh-1903, gh-2045
The morphology interface needs to be standardized:
binary dilation/erosion/opening/closing take a “structure” argument, whereas their grey equivalent take size (has to be a tuple, not a scalar), footprint, or structure.
a scalar should be acceptable for size, equivalent to providing that same value for each axis.
for binary dilation/erosion/opening/closing, the structuring element is optional, whereas it’s mandatory for grey. Grey morphology operations should get the same default.
other filters should also take that default value where possible.
This module is in reasonable shape, although it could use a bit more maintenance. No major plans or wishes here.
Overall this module is in good shape. Two good global optimizers were added in 1.2.0; large-scale optimizers is still a gap that could be filled. Other things that are needed:
Many ideas for additional functionality (e.g. integer constraints, sparse matrix support, performance improvements) in
linprog, see gh-9269.
Add functionality to the benchmark suite to compare results more easily (e.g. with summary plots).
fmin_*functions in the documentation,
scipy.optimizehas an extensive set of benchmarks for accuracy and speed of the global optimizers. That has allowed adding new optimizers (
dual_annealing) with significantly better performance than the existing ones. The
optimizebenchmark system itself is slow and hard to use however; we need to make it faster and make it easier to compare performance of optimizers via plotting performance profiles.
Convolution and correlation: (Relevant functions are convolve, correlate,
fftconvolve, convolve2d, correlate2d, and sepfir2d.) Eliminate the overlap with
ndimage (and elsewhere). From
(and anywhere else we find them), pick the “best of class” for 1-D, 2-D and n-d
convolution and correlation, put the implementation somewhere, and use that
consistently throughout SciPy.
B-splines: (Relevant functions are bspline, cubic, quadratic, gauss_spline, cspline1d, qspline1d, cspline2d, qspline2d, cspline1d_eval, and spline_filter.) Move the good stuff to interpolate (with appropriate API changes to match how things are done in interpolate), and eliminate any duplication.
Filter design: merge firwin and firwin2 so firwin2 can be removed.
Continuous-Time Linear Systems: remove lsim2, impulse2, step2. The
lsim, impulse and step functions now “just work” for any input system.
Further improve the performance of
ltisys (fewer internal transformations
between different representations). Fill gaps in lti system conversion functions.
Second Order Sections: Make SOS filtering equally capable as existing methods. This includes ltisys objects, an lfiltic equivalent, and numerically stable conversions to and from other filter representations. SOS filters could be considered as the default filtering method for ltisys objects, for their numerical stability.
Wavelets: what’s there now doesn’t make much sense. Continuous wavelets only at the moment - decide whether to completely rewrite or remove them. Discrete wavelet transforms are out of scope (PyWavelets does a good job for those).
The sparse matrix formats are mostly feature-complete, however the main issue
is that they act like
numpy.matrix (which will be deprecated in NumPy at
some point). What we want is sparse arrays, that act like
This is being worked on in https://github.com/pydata/sparse, which is quite far
along. The tentative plan is:
Start depending on
pydata/sparseonce it’s feature-complete enough (it still needs a CSC/CSR equivalent) and okay performance-wise.
Add support for
scipy.sparse.linalg(and perhaps to
Indicate in the documentation that for new code users should prefer
pydata/sparseover sparse matrices.
When NumPy deprecates
numpy.matrix, vendor that or maintain it as a stand-alone package.
Regarding the different sparse matrix formats: there are a lot of them. These should be kept, but improvements/optimizations should go into CSR/CSC, which are the preferred formats. LIL may be the exception, it’s inherently inefficient. It could be dropped if DOK is extended to support all the operations LIL currently provides.
This module is in good shape.
Arpack is in good shape.
callback keyword is inconsistent
tol keyword is broken, should be relative tol
Fortran code not re-entrant (but we don’t solve, maybe re-use from PyKrilov)
add sparse Cholesky or incomplete Cholesky
look at CHOLMOD
Ideas for new features:
Wrappers for PROPACK for faster sparse SVD computation.
QHull wrappers are in good shape, as is
KDTreewill be removed, and
cKDTreewill be renamed to
KDTreein a backwards-compatible way.
distance_wrap.cneeds to be cleaned up (maybe rewrite in Cython).
Though there are still a lot of functions that need improvements in precision, probably the only show-stoppers are hypergeometric functions, parabolic cylinder functions, and spheroidal wave functions. Three possible ways to handle this:
Get good double-precision implementations. This is doable for parabolic cylinder functions (in progress). I think it’s possible for hypergeometric functions, though maybe not in time. For spheroidal wavefunctions this is not possible with current theory.
Port Boost’s arbitrary precision library and use it under the hood to get double precision accuracy. This might be necessary as a stopgap measure for hypergeometric functions; the idea of using arbitrary precision has been suggested before by @nmayorov and in gh-5349. Likely necessary for spheroidal wave functions, this could be reused: https://github.com/radelman/scattering.
Add clear warnings to the documentation about the limits of the existing implementations.
scipy.stats subpackage aims to provide fundamental statistical
methods as might be covered in standard statistics texts such as Johnson’s
“Miller & Freund’s Probability and Statistics for Engineers”, Sokal & Rohlf’s
“Biometry”, or Zar’s “Biostatistical Analysis”. It does not seek to duplicate
the advanced functionality of downstream packages (e.g. StatsModels,
LinearModels, PyMC3); instead, it can provide a solid foundation on which
they can build. (Note that these are rough guidelines, not strict rules.
“Advanced” is an ill-defined and subjective term, and “advanced” methods
may also be included in SciPy, especially if no other widely used and
well-supported package covers the topic. Also note that some duplication
with downstream projects is inevitable and not necessarily a bad thing.)
The following improvements will help SciPy better serve this role.
Add fundamental and widely used hypothesis tests:
the various types of analysis of variance (ANOVA):
two-way ANOVA (single replicate, uniform number of replicates, variable number of replicates)
multiway ANOVA (i.e. generalize two-way ANOVA)
analysis of covariance (ANCOVA)
Where appropriate, include confidence intervals for the statistic in the results of any statistical test.
Add additional tools for meta-analysis; currently we have just combine_pvalues.
Enhance the fit method of the continuous probability distributions:
Expand the options for fitting to include:
method of moments
maximal product spacings
method of L-moments / probability weighted moments
Include measures of goodness-of-fit in the results
Handle censored data
Implement additional widely used continuous and discrete probability distributions:
noncentral hypergeometric distribution (both Fisher’s and Wallenius’)
negative hypergeometric distribution
multivariate hypergeometric distribution
multivariate t distribution
Improve the core calculations provided by SciPy’s probability distributions so they can robustly handle wide ranges of parameter values. Specifically, replace many of the PDF and CDF methods from the Fortran library CDFLIB used in scipy.special with better code, perhaps ported from the Boost C++ library.
In addition, we should:
Continue work on making the function signatures of
stats.mstatsmore consistent, and add tests to ensure that that remains the case.
Bunchobjects from functions that now return many values, and for functions for which extra return values are desired (see gh-3665).
Address the various issues regarding
stats.mannwhitneyu, and pick up the stalled PR in gh-4933.