Updated 2020-05-26 19:11:07 UTC
Updated PRs (new commits but old needs-work label) [24]
- gh-12221: WIP, DOC: update 1.5.0 release notes
- gh-11899: Added sparse nonzero functionality to min, max, argmin, argmax
- gh-11862: ENH: Add a Quadratic Assignment Problem Solver in scipy.optimize
- gh-11788: MAINT: fix domain edge error for betainc a or b ==0 case is 1.
- gh-11695: ENH: stats: add method of moments to `rv_continuous.fit`
- gh-11691: ENH: add a stack of reversal functions to linprog
- gh-11682: DOC: drastic rewrite of constants documentation
- gh-11658: WIP, TST, CI: use OpenBLAS for Travis CI Linux
- gh-11313: [WIP] ENH add Wright's generalized Bessel function
- gh-11249: ENH: optimize: interpolative decomposition redundancy removal for linprog
- gh-10639: Matrix identification methods and test
- gh-10598: ENH: added centre and contains feature
- gh-10561: ENH: add poisson two sample test (E-test) to scipy.stats
- gh-10237: [ENH] Coloured signal generation + calculation
- gh-10215: WIP, TST: add compiled cov to Azure CI reports
- gh-9963: ENH: Add the Irwinhall distribution
- gh-9958: Remerge gh-9619: "FIX: Sparse matrix addition/subtraction eliminates explicit zeros"
- gh-9411: ENH: Faster implementation of scipy.sparse.block_diag
- gh-8902: EHN: add nanmin/nanmax in scipy.sparse
- gh-8548: BUG: Fix ellipj when m is near 1.0 (#8480)
- gh-8157: ENH: added association measurements Cramer's V, Tschuprow's T and Phi to scipy.contingency
- gh-7157: ENH: Add polynomial features and extra basis functions to scipy.interpolate.rbf
- gh-5411: WIP: sparse: Sparse bitwise operations (was #3941)
- gh-4747: Implemented Kaiser-Bessel derived window
Needs review [130]
- gh-12227: BLD: Set macos min version when building rectangular_lsap
- gh-12225: BUG: boxcox data constancy check #5112 #10978
- gh-12223: BUG: stats: wilcoxon returned p > 1 for certain inputs.
- gh-12217: BUG: stats: Fix handling of edge cases in median_abs_deviation.
- gh-12214: DOC: fix docstring of "sd_beta" of odr.Output.
- gh-12211: DOC: Dice similiarity index
- gh-12208: wavfile bugfixes and maintenance
- gh-12207: BUG: fix covariance matrix of ODR to be scaled by the residual variance
- gh-12206: MAINT:lstsq: Switch to tranposed problem if the array is tall
- gh-12204: ENH: Add overwrite argument for odr.ODR() and its test.
- gh-12200: ENH: add FIR and IIR gammatone filters to scipy.signal
- gh-12199: ENH: match np.result_type behaviour in some scipy.signal functions
- gh-12198: TST: Fix str + bytes error
- gh-12197: BUG: fix moments method to support arrays and list
- gh-12188: BUG: fix splprep seg fault for task=-1
- gh-12186: MAINT: generalize tests in SphericalVoronoi
- gh-12183: ENH: Use fromfile where possible
- gh-12177: MAINT: Correctly add input line to ValueError
- gh-12172: Bounds for the Powell minimization method
- gh-12170: ENH: scipy.stats: add negative hypergeometric distribution
- gh-12161: ENH: quadratic assignment problem
- gh-12156: BUG: Fix permutation of distance matrices in scipy.stats.multiscale_graphcorr
- gh-12154: ENH: scipy.signal: allow lists in gauss_spline
- gh-12150: DOC: update docstring and add Examples for functions in signal.bsplines.py
- gh-12146: DOC: add docs to explain behaviour of newton's mehod on arrays
- gh-12144: DOC: Update ndimage.morphology.distance_transform_*
- gh-12143: BUG: Correctly use `axis` in `scipy.stats.tmean`
- gh-12142: [MRG] DOC: add a note in poisson distribution when mu=0 and k=0 in pmf method
- gh-12139: TST: Reduce flakiness in lsmr test
- gh-12125: TST: Add a test for stats.gmean with negative input
- gh-12120: TST: gh12111, best fun should be OptimizeResult.fun
- gh-12112: DOC: Improve integrate.nquad docstring
- gh-12086: MAINT: Cleanup PyMODINIT_FUNC used during 2to3
- gh-12085: ENH: special: support ILP64 Lapack
- gh-12080: MAINT: Remove suppress_warnings
- gh-12077: MAINT malloc return value checks for cython
- gh-12071: Update _procrustes.py to handle complex matrices (#1)
- gh-12069: DOC: Add developer guidelines for implementing the nan_policy parameter.
- gh-12067: MAINT: Cleanup OLDAPI in ndimage/src/_ctest.c
- gh-12066: ENH: Exposes factor and logs WRITE in nnls
- gh-12061: Check parameter consistensy in signal.iirdesign
- gh-12049: [WIP] custom leaves ordering for dendrogram
- gh-12043: ENH: optimize: add HiGHS methods to linprog - continued
- gh-12042: Add z-score table and z-score functions
- gh-12029: BUG FIX: pearson3.stats returning moments as arrays rather than scalars.
- gh-12027: ENH: hamming distance for string comparison
- gh-12000: ENH (WIP): Adds Poisson binomial distribution support (See #6000)
- gh-11966: MAINT: fix mypy errors after adding py.typed marker
- gh-11955: MAINT: type sph vor function signatures
- gh-11954: ENH: faster ARGUS rv generation in scipy.stats
- gh-11911: ENH: Added bounds to boxcox and boxcox_normmax
- gh-11846: Add ignore_nan flag to scipy.optimize.curve_fit
- gh-11829: Fixes safe handling of small singular values in svds.
- gh-11823: avoid integer overflow in NI_MinOrMaxFilter
- gh-11815: ENH: Add verbose as optional argument to solve_ivp to print time step info
- gh-11712: BUG: trust-constr evaluates function out of bounds
- gh-11650: Fix trust-constr function evaluations outside of bounds
- gh-11618: BUG: linprog returns 'infeasible' for unbounded case (fix for #11617)
- gh-11604: MAINT: swap int for long type for bdt{r,ri,rc} functions
- gh-11603: ENH: Optimise _cs_matrix._set_many when new entries are all zero
- gh-11563: ENH: exact p-value in stats.kendalltau() for sample sizes > 171
- gh-11491: [WIP] avoid copy for dtype conversion when possible
- gh-11477: FIX: bounded parameter in cdfchn.f gives bad results
- gh-11453: BUG: sparse: convert csc/csr indices to int64 as needed
- gh-11413: ENH: Multivariate t-distribution
- gh-11345: MAINT: constants: revise way 'exact' values are recomputed
- gh-11344: ENH: diagonal banded form in solve banded
- gh-11263: MAINT: closes gh11258 specified rel_step
- gh-11211: ENH: Adds Carlson's elliptic integrals in `scipy.special`
- gh-11150: [BUG] fix rvs output shape in multivariate distributions
- gh-11147: [WIP] fix some dimensionality inconsistencies
- gh-11144: Fix wavfile import for scipy.io
- gh-11119: ENH: add Cramer-von-Mises (one-sample) test to scipy.stats
- gh-11068: ENH: Adds kwarg xcorr to fftconvolve.
- gh-11010: TST: add Azure stages
- gh-10860: ENH: option to reuse scipy.interpolate.RegularGridInterpolator with new values
- gh-10844: ENH: add stats.qmc module with quasi Monte Carlo functionality
- gh-10840: MAINT: use the define for NPY_NO_DEPRECATED_API, fix where needed
- gh-10778: ENH: add differential-evolution to minimize
- gh-10730: ENH: Add new method in order to perform a balanced cut tree
- gh-10504: ENH: Carlson symmetric elliptic integrals.
- gh-10129: ENH: add method that creates new spline for a partial derivative of bivariate spline
- gh-10015: skip unnecessary function calls in derivative()
- gh-9974: ENH: Left truncated weibull (minimum) distribution
- gh-9948: DOC: Updated documentation for nan_policy=propagate as per #9947
- gh-9932: Add pdf_marginal and logpdf_marginal to kde
- gh-9706: DOC: Correction for scipy.stats.poisson - interval documentation
- gh-9678: ENH: Add Yuen T test
- gh-9523: ENH: improvements to the Stable distribution
- gh-9473: Compute KS p-values lazily.
- gh-9462: Cythonize ks_2samp.
- gh-9361: ENH: Add Kendall's tau-a and tau-c variants to scipy.stats.kendalltau()
- gh-9314: ENH: WIP Add support for LowLevelCallable to optimize.zeros.brenth et al
- gh-9291: Fix slow stats.fisher_exact(x) (#9231)
- gh-9075: ENH: added query_range method to KDTree class
- gh-9072: ENH: Added rectangular integral to multivariate_normal
- gh-9019: MAINT: code cleanup in optimize/linesearch
- gh-8766: Methods for pdf computation, parameter estimation, and variable sampling in stats.levy_stable. Support for different parameterizations.
- gh-8727: BUG: vq.kmeans() compares signed diff to a threshold.
- gh-8552: Scipy PEP for class based Optimizers
- gh-8529: Read wavfiles of size > 4GB
- gh-8404: relax cdflib precision to 1e-15
- gh-8305: Circular convolution
- gh-8289: DOC/ENH: Explain limitations of stats.levy_stable more explicitly
- gh-8213: MAINT: Rename sosfreqz to freqz_sos
- gh-8151: BUG: adding Buhring's analytic continuation method to special.hyp2f1; #8054
- gh-8110: BUG: special.hyp2f1 now allows neg int c value
- gh-8108: Bugfix #8103: scipy.stats.boxcox - np.nan produces warnings
- gh-8032: Add in taylor window common in Radar processing
- gh-7946: ENH: add a unique function for sparse matrices
- gh-7257: ENH: add rv_scatter distribution
- gh-7205: ENH: implemented kendall tau distance
- gh-6931: ENH: add weight-awareness to many functions in scipy.stats and scipy.…
- gh-6800: ENH: stats: Mixture distribution
- gh-6386: Improve Pinv
- gh-6345: ENH: Vectorized computations in approx_derivative
- gh-6012: Implement signal.planck (Planck-taper window)
- gh-5763: MAINT: spatial: reimplement distance_wrap in Cython
- gh-5564: BUG: fix KDTree for input with many identical values
- gh-5512: Corrected definition of pearsonr for masked arrays
- gh-5318: Added the BFGS(Hess) Algorithm for minimization independent of target function.
- gh-5025: BUG: Fix of dlsim when the system is characterized has no dynamic.
- gh-5006: MAINT: Rewrote scipy.ndimage.find_object function in cython
- gh-4872: ENH: tf2zpk positive or negative powers
- gh-4824: Permutation Ttest (new PR)
- gh-4577: MNT: Use dense array for solution of spsolve
- gh-4519: Fast permutation one-way anova
- gh-4406: ENH: Implemented methods 'extended' and '3/8-th' in simps function. fixes gh-4264
- gh-4122: remezord implementation
- gh-3213: ENH Implement Jensen-Shannon Divergence
Same as this github search, with updated PRs, WIPs, and review:changes-requested excluded.
Needs decision [17]
- gh-11785: ENH: Add functionality to accept descending input for scipy.interpolate.interpn and RegularGridInterpolator
- gh-11609: ENH: add CITATION.cff
- gh-11346: ENH: add fast toeplitz matrix multiplication using FFT
- gh-11193: (Draft) 64-bit BLAS/LAPACK
- gh-11165: BUG: io.matlab: Squeeze 0d to numpy scalars, not python ones
- gh-10449: Add a Gitter chat badge to README.rst
- gh-10112: ENH: minimize success should be False if solution is infeasible
- gh-9701: make p0 (initial values) required for curve_fit() instead of defaulting to 1
- gh-8773: spsolve: separate logic for choosing method and method itself
- gh-8723: BUG: sparse.linalg: preserve column vector kind in iterative solvers
- gh-8306: Optional Pythran support for scipy.spatial._hausdorff
- gh-6923: optimize: API to control iteration in differential evolution
- gh-6907: differential_evolution: improve callback
- gh-6419: Added Breit-Wigner distribution
- gh-5747: [ENH] circvar calculated simply as 1-R
- gh-4904: Add root finding ode subclass. (not using the native solver's capabilities, just retracing and bisecting **in python**)
- gh-3163: ENH: add distance gufuncs to `scipy.spatial`
Same as this github search, with updated PRs and WIPs excluded.
Needs work etc. [97]
- gh-12184: (Draft) ENH: Avoid re-factorizing covariance matrix in multivariate_normal_frozen
- gh-12097: (Draft) ENH: Add Rayleigh MLE fit method
- gh-11985: WIP: Add wrapper for ?pbsvx
- gh-11890: WIP: Python wrapper for ?gbsvx
- gh-11888: WIP, arffread.py -- redundant if-condition
- gh-11855: WIP, MAINT: Test cython
- gh-11754: (Draft) WIP/ENH: stats: Updates to `variation`.
- gh-11502: WIP: alpha shapes
- gh-11352: (Draft) ENH: stats: Add crosstab function.
- gh-11257: WIP: sparse redundancy removal routine optimizations
- gh-11212: Improved Radial Basis Function Interpolation
- gh-11158: (Draft) WIP: Remove 0d/1d special-casing from _numdiff
- gh-11148: WIP: DOC: examples for docstrings in scipy.stats
- gh-11143: WIP: Add constants.svg tree
- gh-11129: WIP: DOC: convert hyperlinks to interlinks
- gh-11030: WIP: Adjust digital filter design warnings
- gh-10871: handle duplicate x_min values in interp1d
- gh-10625: Make appendmat actually append ".mat".
- gh-10623: ENH: Simplify filter_type in *ord() functions
- gh-10519: [DOC] More accurate docstring of nan policy=propagate in some descriptive statistics
- gh-10511: WIP, ENH: Reduce fft sizes for zero-padded arrays
- gh-10454: ENH: Extend find_peaks_cwt to take numbers and iterables for widths argument.
- gh-10375: (Draft) ENH: sparse/linalg/arpack: Upgrade arpack-ng to 3.7.0
- gh-10170: (Draft) MAINT: stats: Implement cdf and ppf as ufuncs for the cosine distribution.
- gh-10102: FIX: now can calculate distances correctly, regardless of type
- gh-10039: MAINT: malloc checks for C source files
- gh-10035: MAINT: malloc return value checks for cython
- gh-9973: (Draft) ENH: speed up multivariate_normal.logpdf for nonsingular matrices
- gh-9957: MAINT: fix sparse matrix comparison operator overloading
- gh-9839: ENH: scipy.optimize.minimize with method=SLSQP returns KKT multipliers
- gh-9735: WIP: discrete Frechet distance function in scipy.spatial.distance
- gh-9689: ENH: add johnsonsl distribution to scipy.stats
- gh-9430: ENH: extend stats.bartlett to 2-D arrays
- gh-8896: FIX: generating random sparse matrix with size > max(int32)
- gh-8835: ENH/TST: Update Newton's method.
- gh-8818: WIP/ENH: Allow r to be an array for cKDTree.query_ball_point
- gh-8779: WIP: Run benchmarks
- gh-8775: ENH: added accept_rate as a keyword to basinhopping
- gh-8707: (Draft) WIP/ENH: datasets: Create the datasets subpackage.
- gh-8679: Stop astype from overwriting arrays when copy=True
- gh-8525: ENH: scipy.sparse: add method for (faster) lil matrix vector multiplication
- gh-8472: ENH: in scipy.sparse.linalg.norm implement spectral norm
- gh-8414: WIP: Class based Optimizers
- gh-8360: BUG: Issue #4464 Fixed residuez which failed for Complex Numbers
- gh-8286: WIP: ENH: Add rank-1 Cholesky update/downdate
- gh-7903: ENH: add more interpolation methods to RegularGridInterpolator.
- gh-7819: WIP: Basin hopping improvements
- gh-7800: WIP: Adding a valgrind tool script.
- gh-7702: WIP: ENH: Add Skewed Cauchy Distribution to scipy.stats
- gh-7569: WIP: Implement LineBufferIterator class and wrappers.
- gh-7425: ENH: optimize: more complete callback signature
- gh-7310: ENH: Add Hamilton's FFTLog.
- gh-7127: ENH: Add infrastructure for sparse in-place interface.
- gh-6870: _cs_matrix: warn about bigger data/indices in check_format
- gh-6852: WIP: adding 24-bit support to io.wavfile
- gh-6837: ENH: Add lanczos window to scipy.signal.windows
- gh-6832: WIP: SOS LTI systems
- gh-6739: added Normal inverse gamma distribution
- gh-6736: DOC, ENH: run linkchecker on html docs after refguide check & fix broken links
- gh-6617: ENH: instantiate sparse.lil_matrix with data
- gh-6482: ENH: add a function for Gauss-Hermite quadrature
- gh-6466: WIP: arbitrary continuous distribution
- gh-6331: WIP: sparse.linalg: Convert `spsolve` to `solve_banded` for DIA matrix inputs
- gh-6089: Make distance functions broadcasting
- gh-6066: Seperate out the creation of a Gaussian kernel from the filtering
- gh-6053: BUG: savemat does not save keys starting with underscore
- gh-6004: Add fast_lil_matrix sparse matrix
- gh-5952: Allow for absolute finite difference step
- gh-5886: WIP - BUG: scipy.signal.lsim overflow error
- gh-5819: Fixes for #2324 rankdata returns wrong results on masked arrays
- gh-5775: ENH: Add infrastructure for sparse in-place @ with gemm interface
- gh-5737: ENH: Added transfer function estimation
- gh-5619: Faster block_diag when blocks are equally shaped
- gh-5591: Fixed SparseMatrixOperator to correctly implement adjoint.
- gh-5327: BUG: io: Fix savemat compression OverflowError
- gh-5288: BUG: multivariate_normal returns a pdf for values outside its support when covariance is singular
- gh-5109: MAINT: Cython version of distance_transform_bruteforce function of mo…
- gh-5087: Remove more uses of inv() and replace with cho_solve()
- gh-5031: Extend to_writable method to array-like
- gh-5012: ENH: add 'alternative' and 'exact' keywords to `stats.ansari`
- gh-4933: Update the Mann-Whitney-Wilcoxon test
- gh-4697: Extrapolate Anderson-Darling p-values linearly.
- gh-4674: WIP: ENH: interpolate: add rational interpolation
- gh-4607: WIP: add Chirp Z-transform, zoom FFT
- gh-4460: WIP: add SQUAREM for fixed point problems
- gh-4459: ENH: optimize: add stagnation check to df-sane
- gh-4444: BF: use UTF16 for saving unicode to mat files
- gh-4384: ENH: optimize, returning True from callback function halts minimization
- gh-4279: Kdtree subset query
- gh-4166: A faster DFS implementation: (i) linear (ii) releases gil. Some assumpti...
- gh-3597: Use real names when loading MATLAB opaque objects.
- gh-3535: sparse arff
- gh-3472: Hankel transforms for scipy.signal
- gh-3260: This adds root finding to pchip as this is possible to do optimal because pchip is a monotonic interpolation.
- gh-2862: Enable simulation of MIMO systems
- gh-2787: DOC: Fixing a KrylovJacobian class - newton_krylov function's **kw and changing deprecated parameters
- gh-354: Faster expm
Same as this and this github search, with updated PRs excluded and WIPs included.
Needs champion [15]
- gh-6470: A better Brent's method implementation
- gh-5141: ENH: Parallel differential evolution
- gh-5120: ENH: use derivatives for boxcox transformation parameter and error estimation
- gh-5119: More efficient median binned stat calculation and new min & max binned stats
- gh-5113: MAINT: more data validation for boxcox transformation
- gh-5062: MAINT: fancy indexing of a sparse matrix now returns a sparse matrix
- gh-4934: WIP: fix an expm typo.
- gh-4886: MAINT: svd-based mahalanobis distance
- gh-4689: MAINT: simplify some indexing code in sparse matrix sampling
- gh-4294: MAINT: pchanial's improvement of input validation in sparse's __mul__
- gh-4115: ENH: associated Legendre function of second kind with types 2 and 3
- gh-3659: added cramer von-mises two sample test
- gh-3406: clarify scipy.cluster.hierarchy.linkage usage
- gh-2974: ENH: Alex Leach's wrapper of Roger Sidje's EXPOKIT
- gh-435: MAINT: special: deprecate truncate-to-int behavior
Same as this github search.
Needs backport [16]
- gh-11757: ENH: improve binned_statistics_2d performance.
- gh-11639: BUG: interpolate/interp1d: fail gracefully on all-nan inputs
- gh-11631: Revert to old behaviour for constant cost matrices in `linear_sum_assignment`
- gh-11534: FIX: Allow non-native byte order inputs to scipy.fft
- gh-11459: BUG: fix augmentation being broken in maximum_bipartite_matching
- gh-11382: MAINT: remove error throw in binned_statistic_dd() on non-finite values or samples
- gh-11343: BUG: constants: fix 'exact' values table
- gh-11330: MAINT: Update vendored pypocketfft version
- gh-11324: BLD: Don't assume -pthread is already set on compiler
- gh-11320: BUG: Include `int64_t` declaration to `rectangular_lsap.cpp`
- gh-11300: MAINT: Update vendored uarray version
- gh-11269: Fix: Sparse matrix constructor data type detection changes on Numpy 1.18.0
- gh-11265: Fix: Can no longer slice csr_matrix in 1.3.0
- gh-11208: BUG: #11207 _minimize_scalar_bounded: reference before assignment
- gh-11097: BUG: stats: fixing chisquare to return correct results for arrays of type int32 with large values
- gh-10540: MAINT: Fix Travis and Circle
Same as this github search.
Generated by github-needs-work