Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Comparing changes

Choose two branches to see what's changed or to start a new pull request. If you need to, you can also compare across forks.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also compare across forks.
base fork: scipy/scipy
...
head fork: scipy/scipy
Checking mergeability… Don't worry, you can still create the pull request.
This comparison is big! We're only showing the most recent 250 commits
Commits on Mar 16, 2015
@raoulbq raoulbq Fix python3 compatibility errors 00586ed
@raoulbq raoulbq Small indent fixes e160114
@raoulbq raoulbq Get a small speedup by vectorized code c8240dd
@raoulbq raoulbq pep8 compliance 5983fae
@raoulbq raoulbq Use scipy own airy zero function 9c19e91
@raoulbq raoulbq Remove unused airy zero asymptotics 24e41f4
@raoulbq raoulbq Remove unused code 557d692
@raoulbq raoulbq Remove unused imports 2a00ef6
@alexbrc alexbrc MAINT: generalize a sign function to do the right thing for complex n…
…umbers
e9eb618
@ewmoore ewmoore MAINT: Further improve error measage in lfilter. 983d531
@raoulbq raoulbq Prefix internal functions with an underscore 2287a64
@raoulbq raoulbq Remove unnecessary tuple ff37615
@raoulbq raoulbq Precompute powers of phi 12f786c
@raoulbq raoulbq Put reference in proper place 6c25c53
@raoulbq raoulbq Remove unnecessary check 01e4d20
@raoulbq raoulbq Make selection of Ai zeros more explicit c9cf727
@raoulbq raoulbq Test Gauss-Hermite weights for large n ce0d5dc
@raoulbq raoulbq Test 3-item return from h_roots for large n 6a56995
@raoulbq raoulbq Replace flipud by plain index constructs 2660c6d
@ewmoore ewmoore BUG: restore pseudobroading to lfilter, correct zf shape when broadca…
…sting

lfilter requires that zi.ndim match x.ndim, and will not prepend
1s to acheive this.  It does allow dimensions of length 1, which
are expanded in the usual way.  It previously failed by always
forcing zf to have the same shape as zi, meaning it would generally
fail to return all of the final conditions if this broadcasting
behavior was used.  This commit corrects the shape of zf and
restores this pseudobroadcasting behavior that was removed while
fixing another bug.
8d45a63
Commits on Mar 17, 2015
@WarrenWeckesser WarrenWeckesser ENH: linalg: add the function 'solve_circulant' for solving a circula…
…nt system.
2535400
@Newman101 Newman101 Amended typo in apple_sgemv_fix.c 13bd09c
@argriffing argriffing Merge pull request #4639 from Newman101/master
Amended typo in apple_sgemv_fix.c
79cb69c
@alexbrc alexbrc MAINT: use lapack for scipy.linalg.norm 01efcf7
@alexbrc alexbrc BENCH: add matrix norm benchmarks f4fbb69
@rgommers rgommers Merge pull request #4614 from ewmoore/lfilter_conv
BUG: lfilter, the size of zi was not checked correctly for nd in.
a48063d
@alexbrc alexbrc BENCH: add an import 5baf1fa
@alexbrc alexbrc PEP8: whitespace before colon adb7a99
Commits on Mar 18, 2015
@insertinterestingnamehere insertinterestingnamehere BUG: Fixed declarations involving f2py fortranname keyword for
combatibility of Cython BLAS and LAPACK wrappers with OSX.
64319ab
@insertinterestingnamehere insertinterestingnamehere ENH: Added Cython API for all currently wrapped BLAS routines. ef21dfa
@insertinterestingnamehere insertinterestingnamehere ENH: Added Cython API for all currently wrapped LAPACK routines. a1b89cf
@insertinterestingnamehere insertinterestingnamehere TST: Test to make sure that the Cython BLAS wrappers work.
In particular, test that the functions that use the fortranname keyword in
fblas_l1.pyf.src work properly.
7ada589
@insertinterestingnamehere insertinterestingnamehere TST: Added a test for one of the Cython LAPACK wrappers. 445f9be
@insertinterestingnamehere insertinterestingnamehere ENH: Added the function get_linalg_dir to get the directory containin…
…g scipy.linalg.
5087adc
@insertinterestingnamehere insertinterestingnamehere DOC: Document Cython API for BLAS and LAPACK. 3cb1287
@insertinterestingnamehere insertinterestingnamehere MAINT: Ignore C files generated for the Cython BLAS and LAPACK wrappers
and their corresponding tests.
78798c4
@insertinterestingnamehere insertinterestingnamehere MAINT: Require Cython 0.21 in cythonize script. edd99b0
@insertinterestingnamehere insertinterestingnamehere DOC: Increase required Cython version listed in INSTALL.rst.txt. abb6482
@rgommers rgommers TST: add run_module_suite usage to tests for linalg Cython interface. 54cb608
@rgommers rgommers BLD: add Bento support for linalg Cython interface. 5aed08c
@insertinterestingnamehere insertinterestingnamehere TST: Refactored tests to move testing logic for Cython BLAS wrappers …
…into the testing directory.
44f4123
@insertinterestingnamehere insertinterestingnamehere TST: Refactored tests to move testing logic for Cython LAPACK wrappers
into the testing directory.
0bcc1b3
@insertinterestingnamehere insertinterestingnamehere ENH: Added subroutine wrappers for functions with return values expos…
…ed in

the Cython api for BLAS and LAPACK.
Also expanded and refactored the corresponding tests.
90b8da0
@insertinterestingnamehere insertinterestingnamehere BLD: Added Bento support for subroutine wrapper used in Cython
interface for BLAS and LAPACK.
f3a52b2
@pv pv ENH: Changed the Cython api for BLAS and LAPACK to link directly against
compiled fortran instead of relying on f2py.
ba438ea
@insertinterestingnamehere insertinterestingnamehere BUG: Fixed test failures for the Cython BLAS and LAPACK API.
Cleaned up naming conventions in the tests.
Fixed a text encoding bug on Python 3.
8ffe15d
@insertinterestingnamehere insertinterestingnamehere ENH: Added a full set of Cython wrappers for BLAS. d2c2259
@insertinterestingnamehere insertinterestingnamehere ENH: Added a full set of Cython wrappers for LAPACK. 7d35eb4
@insertinterestingnamehere insertinterestingnamehere DOC: Added Cython BLAS and LAPACK wrappers to the 16.0 release notes. 08fde40
@insertinterestingnamehere insertinterestingnamehere ENH: Added wrapper generator script for the Cython BLAS/LAPACK wrappers. 2fe8785
@insertinterestingnamehere insertinterestingnamehere ENH: Added signature file to allow generation of Cython BLAS wrappers. 6a3bb09
@insertinterestingnamehere insertinterestingnamehere ENH: Added signature file to allow generation of Cython LAPACK wrappers. 12dbf22
@insertinterestingnamehere insertinterestingnamehere ENH: Added a script to generate BLAS/LAPACK signature files used to make
the corresponding Cython wrappers.
85deb4f
@insertinterestingnamehere insertinterestingnamehere BUG: Fix build of Cython BLAS/LAPACK wrappers. 5a2657a
@insertinterestingnamehere insertinterestingnamehere STY: Cleaned up naming and simplified wrapper code. 9a4224d
@insertinterestingnamehere insertinterestingnamehere STY: Add comments at top of autogenerated files showing their origin. 31679c4
@insertinterestingnamehere insertinterestingnamehere STY: Fixed array dimensions in arguments for subroutine wrappers used in
the Cython LAPACK API.
d00ea63
@insertinterestingnamehere insertinterestingnamehere BUG: Exclude LAPACK routines with string arguments since the convention
for passing strings to Fotran routines is highly compiler dependent.
d84db5a
@insertinterestingnamehere insertinterestingnamehere BUG: RE-added linalg.pxd in order to expose the BLAS/LAPACK wrappers to
other packages.
82d8efb
@insertinterestingnamehere insertinterestingnamehere STY: Cleaned up handling of complex numbers to cut down on warning
messages when compiling Cython BLAS/LAPACK wrappers.
616b235
@insertinterestingnamehere insertinterestingnamehere STY: Simplified casting to further reduce the number of warnings gene…
…rated

when compiling the Cython LAPACK wrappers.
f7153e5
@insertinterestingnamehere insertinterestingnamehere BUG: Remove Cython wrappers for xerbla and zcgesv since their signatures
are not consistent across LAPACK versions.
a270f82
@insertinterestingnamehere insertinterestingnamehere ENH: Added wrappers for requested LAPACK auxiliary routines.
Also cleaned up code generation scripts to allow generation of wrappers
for all auxiliaries in LAPACK 3.1.0.
1199297
@insertinterestingnamehere insertinterestingnamehere ENH: Added wrappers for remaining LAPACK auxiliary routines included …
…with

LAPCK 3.1.0 that are present without interface changes in each version
through 3.5.0.
488682e
@rgommers rgommers BLD: fix Bento build for Cython BLAS/LAPACK wrappers.
Also two doc formatting tweaks included.
d599567
@insertinterestingnamehere insertinterestingnamehere DOC: Added a commented description of how to use the Cython wrappers for
BLAS and LAPACK within scipy.
6b2bde7
@insertinterestingnamehere insertinterestingnamehere DOC: Added a warning in the module-level docstrings for the Cython
BLAS/LAPACK wrappers mentioning that array alignment should be checked
before the wrappers are called.
710dba4
@insertinterestingnamehere insertinterestingnamehere DOC: Added Ian Henriksen to THANKS.txt for the Cython BLAS/LAPACK API. 555d6e3
@alexbrc alexbrc MAINT: treat c/fortran ordering and array contiguity more carefully 97a5ada
@alexbrc alexbrc MAINT: try reducing norm function call overhead 0844ac7
@alexbrc alexbrc BENCH: for matrix norm benchmarks allow non-square matrices and allow…
… negative matrix entries
63d9de7
@alexbrc alexbrc BENCH: add benchmarking of frobenius norm c0e367e
@alexbrc alexbrc MAINT: do not use *lange for frobenius norm dc85f2d
@ev-br ev-br DOC: remove :nowrap: in linalg tutorial a88c40d
@ev-br ev-br DOC: fix a LaTeX error in the linalg tutorial 8db3f91
Commits on Mar 20, 2015
@raoulbq raoulbq Remove unused variable cd9924e
Anne Archibald Added test for von Mises numerical underflow 56c7f61
Anne Archibald Fix von Mises underflow problem a8816a1
@ewmoore ewmoore DOC: Fix tukey docstring so figure appears in generated docs. 0b5fa50
@argriffing argriffing Merge pull request #4649 from aarchiba/vonmisesfix
Vonmisesfix
4f41d3a
@ewmoore ewmoore DOC: fixup place_poles examples 7154ffd
@maniteja123 maniteja123 DOC: Fix the error in convolve for same mode
In the tutorial for same mode, it was incorrectly mentioned that the convolve output has the same length as the largest input instead of the first input. fixes #4640
7f5f2f3
@rgommers rgommers Merge pull request #4647 from ev-br/linalg_tut
MAINT: fix doc pdf build
bd9eb67
@rgommers rgommers Merge pull request #4652 from maniteja123/issue4640
DOC: Fix the error in convolve for same mode
674d040
Commits on Mar 21, 2015
@WarrenWeckesser WarrenWeckesser Merge pull request #4650 from ewmoore/tukey_doc
Signal example clean up in Tukey and place_poles
0661109
@juliantaylor juliantaylor ENH: put polynomial functions in header and increase optimization of …
…ndtr

Allows inlining of the constant size polynomials with gcc.
a784547
@juliantaylor juliantaylor ENH: improve erf performance by unroll loops and __builtin_isnan
Unrolling polynomial evaluations and replace isnan with builtin to
double speed up erf(|x| < 1) and erfc by about 30%.
GCC does unfortunately not replace isnan with builtins automatically.
f9e381b
@pv pv BENCH: add special.erf benchmarks b745cba
@pv pv ENH: special: use cephes_is* everywhere instead of npy_is*, for bette…
…r performance
8c2b29b
@pv pv Merge pull request #4653 from juliantaylor/erf-improve
Allow inlining polynomials, combined with unrolling and isnan replace.
Do this for all Cephes functions.
f2023bc
Commits on Mar 22, 2015
@aeklant aeklant DEP: deprecate scipy.stats.histogram2 in favour of np.histogram2d
See issue #602
9784f40
Commits on Mar 23, 2015
@endolith endolith DOC: Add convolve() example and tight_layout()
Plots were displayed with overlapping labels on the scipy doc website
6fc2aa0
@lightcatcher lightcatcher Avoid extra copy for sparse compressed [:, seq] and [seq, :] slices. 43ce18e
@rgommers rgommers Merge pull request #4659 from endolith/convolve_example
DOC: Add convolve() example and tight_layout()
78390e3
@aeklant aeklant DEP: deprecate scipy.stats.signaltonoise
See issue #609
8a6e358
Commits on Mar 24, 2015
@ewmoore ewmoore Merge pull request #4637 from argriffing/complex-onenorm-sign-vs-dire…
…ction

MAINT: generalize a sign function to do the right thing for complex numbers
19e0b8a
@andreas-h andreas-h Update _monotone.py
Current `__init__` of `Akima1DInterpolator` doesn't support `axis` kwarg.
95f76fe
Commits on Mar 25, 2015
@ev-br ev-br Merge pull request #4664 from andreas-h/patch-2
Update _monotone.py
dde07a6
Commits on Mar 27, 2015
@ewmoore ewmoore Merge pull request #4021 from insertinterestingnamehere/cyblas
ENH: Cython api for lapack and blas
548a732
Commits on Mar 29, 2015
@pv pv MAINT: remove spurious add_data_dir
It had a side effect of including everything under benchmarks/* to the
built packages.
52354d4
Commits on Mar 30, 2015
@terrycojones terrycojones Added missing colon to :math: directive in docstring. 68a16ca
@alexbrc alexbrc ENH: sparse randn a3f9b68
Commits on Apr 01, 2015
@ev-br ev-br Merge pull request #4676 from terrycojones/fix-docstring-typo
DOC: cluster: Add missing colon to :math: directive in docstring.
bd2a354
@terrycojones terrycojones MAINT: cluster: Clean up duplicated exception raising code. cabb8a1
@AlexGrig AlexGrig Added wrappers for LAPACK least-squares solvers.
There are more efficient least-squares solvers available in LAPACK
which are not used in Scipy. This commit includes only the wrappers
for LAPACK functions. In the next commit the modification of the
existing LAPACK least-squares solvers is done.
  modified:   scipy/linalg/flapack.pyf.src
  modified:   scipy/linalg/tests/test_lapack.py
b107d0e
@AlexGrig AlexGrig Extra LAPACK wrappers for least-squares solvers.
Tolerance in the tests has been increased. Cosmetic code
modifications has been performed. Typos fixed. Everything
according to the pull #4471.
d30d7ec
@AlexGrig AlexGrig DOC: Information about new least-squares wrappers is added.
Info about new wrappers is added to scipy/linalg/lapack.py
5b89d2e
@AlexGrig AlexGrig ENH: Added functions which request lwork size for LAPACK least-square…
…s solvers.

Added functions which request sizes of auxiliary arrays for all least-squares LAPACK drivers.
This way is faster than previously, because the actual data is not copied to LAPACK
when sizes request is being done.
962aa89
@AlexGrig AlexGrig DOC: Added info of recently intoduced LAPACK wrappers into linalg.lap…
…ack.py
caaaafd
@AlexGrig AlexGrig BUG: Get rid of unused `minmn` variable in LAPACK wrappers. 3fb8be3
@ewmoore ewmoore MAINT: Don't return work array from newly wrapped lapack routines
Also, don't accept -1 for lwork in these routines, require the use
of the matched _lwork to calculate this size.  If possibe, check
that the lwork passed in is at least the minimum required size.
988901c
@ev-br ev-br Merge branch 'pr/4661'
Reviewed at #4478,
cleaned in #4661
03890cd
@ev-br ev-br Merge pull request #4686 from ev-br/pr/4680
MAINT: cluster: Clean up duplicated exception raising code.
faab714
Commits on Apr 02, 2015
@levelfour levelfour BUG: (gh-4596) check if input matrix is empty 4739653
@lesteve lesteve Improve is_distance_dm exception message
Converting tolerance to an int can be slightly misleading
905cfe7
@alexbrc alexbrc MAINT: use convenience functions for choosing blas and lapack routine…
… variants
65ad15d
@alexbrc alexbrc MAINT: increase flexibility of random sampling of sparse matrices 77af8f4
@alexbrc alexbrc DOC: add a sparse random docstring example e040094
@argriffing argriffing Merge pull request #4679 from argriffing/sparse-randn
ENH: sparse randn
4585b0f
@dlax dlax Merge pull request #4572 from rmcgibbo/lbfgs-hessinv
ENH: Extract inverse hessian information from L-BFGS-B
2e1dcc9
@vermelha vermelha MAINT: Correct odeint messages
odeint is writing messages to standard output. Change to use warnings for messages.
20d2384
Commits on Apr 03, 2015
@rgommers rgommers Merge pull request #4655 from aeklant/issue_#602
DEP: deprecate scipy.stats.histogram2 in favour of np.histogram2d
b961a9f
@rgommers rgommers Merge pull request #4656 from aeklant/issue_#609
DEP: deprecate scipy.stats.signaltonoise
9839ea5
Commits on Apr 04, 2015
@Andreea-G Andreea-G BUG: Fixes #4408: Vector-valued constraints in minimize() et al
Using optimize.fmin_cobyla or optimize.minimize with method 'Cobyla' with vector-valued constraints used to throw a ValueError "setting an array element with a sequence" at line 240 in cobyla.py. The constraints functions could only return numbers for Cobyla, while SLSQP was able to take in vector constraints. Now Cobyla can take either a single constraint or a sequence of constraints, where each constraint function can return either a number or a sequence of numbers. This is true for both fmin_cobyla and minimize.
a778560
@dlax dlax Merge pull request #4648 from Andreea-G/master
BUG: Fixes #4408: Vector-valued constraints in minimize() et al
301ba23
@ev-br ev-br Merge pull request #4662 from vermelha/ODEintMessages
MAINT: Correct odeint messages
a1bfecb
@ev-br ev-br MAINT: special_matrices: add a note about handling of empty sequences
Also pep8-ify the test.
d2537fa
@ev-br ev-br Merge branch 'pr/4672' 2b856a1
@ev-br ev-br Merge pull request #4642 from argriffing/use-wrapped-lapack-matrix-norms
MAINT: use lapack for scipy.linalg.norm
b015eda
@rgommers rgommers Merge pull request #4563 from rgommers/fixup-ppcc
ENH: rewrite of stats.ppcc_plot
e22f52a
@WarrenWeckesser WarrenWeckesser MAINT: stats: Simplify the calculation in tukeylambda._ppf 7e28b07
Commits on Apr 05, 2015
@aeklant aeklant ENH: added functionality to handle scalars in `stats._chk_asarray`
See issue #4550 for details.
ce9c441
@aeklant aeklant STY: style fix (blank line, PEP8) 5c0bfb1
@rgommers rgommers Merge pull request #4692 from WarrenWeckesser/tukeylambda-ppf
MAINT: stats: Simplify the calculation in tukeylambda._ppf
ba01a06
@rgommers rgommers Merge pull request #4688 from lesteve/patch-2
MAINT: spatial: improve is_distance_dm exception message
9e04e53
@anntzer anntzer Vectorization of Anderson-Darling computations.
Results in >1000x acceleration of scipy.stats.anderson_ksamp for
datasets of N~1000.
5ad6b7f
@aeklant aeklant MAINT: `_chk_asarray` and `spearmanr` grooming. c89f295
@aeklant aeklant TST: change to `assert_` in `_chk_asarray` using functions. d1056ff
Commits on Apr 06, 2015
@anntzer anntzer Added benchmark for anderson_ksamp. a0d3591
@aeklant aeklant TST: add scalar tests for `ttest_*` functions. 0d21f89
@argriffing argriffing Merge pull request #3157 from WarrenWeckesser/enh-linalg-solve-circ
ENH: linalg: add the function 'solve_circulant' for solving a circulant system
ea3bd82
@ewmoore ewmoore BUG: off by 1 error in (s,d,cs,zd)rot f2py wrappers
Tests in following commit
636596a
@ewmoore ewmoore ENH: wrap lapack's (sdcz)lartg, larfg, larf and (cz)rot f7812e5
@ewmoore ewmoore ENH: Add qr_delete function 0e8b86e
@ewmoore ewmoore ENH: Add qr_insert function ca8799f
@ewmoore ewmoore ENH: Add qr_update function 17131eb
@ewmoore ewmoore STY: pep8 fixes 10d8a7b
@ewmoore ewmoore BUG: Don't unnecessarily apply rotations to the zero part of R. 276f4c1
@ewmoore ewmoore ENH: Support rank 1 updates to economic QR decompositions.
This is currently using a routine I wrote to compute vector
2-norms.  This is to be replaced with calls to (s,d,sc,zd)nrm2
once that is wrapped.
f1fe2e9
@ewmoore ewmoore ENH: Support rank-k updates to thin QR decompositions
This is implemented by apply k consecutive rank-1 updates rather
than attempting to block this algorithm.

As part of this, the input validation code in qr_update was also
reworked.
85bdf2f
@ewmoore ewmoore ENH: Support check_finite in qr update routines 66778c1
@ewmoore ewmoore ENH: Support row deletion to thin QR decompositions a319330
@ewmoore ewmoore ENH: support column deletions with thin QR decompositions f7da055
@ewmoore ewmoore TST: bump test tolerance
Works for me, but travis seems to fail depending on the exact
configuration used.
5c93e38
@ewmoore ewmoore ENH: support row insertion into thin QR decompositions 3fb4357
@ewmoore ewmoore ENH: Support inserting columns in thin QR decompositions
This isn't so great yet for inserting a column vector that lies
in the span of q.dot(r).
7f0cd22
@ewmoore ewmoore MAINT: Clean up qr_insert entry point. 7d2e7bb
@ewmoore ewmoore STY: address review comments, improve error messages. etc. dce80e9
@ewmoore ewmoore MAINT: Remove numpy 1.5 work arounds 6c08676
@ewmoore ewmoore MAINT: Add _update_decomp extension to the bento build 85d857c
@ewmoore ewmoore ENH: add rcond to qr_insert, only used for col insert to thin qrs.
The output is not a QR decomposition if the column inserted lies
in the span of Q. This will now fail in that case instead of
returning an incorrect result.
307b4f3
@ewmoore ewmoore BUG: change overwrite default to False in qr update routines b4c313b
@ewmoore ewmoore MAINT: Use templates to remove some nearly duplicated code 6a8a428
@ewmoore ewmoore ENH: release the GIL around computation routines in _decomp_update d8b8f54
@aeklant aeklant MAINT: maintenance changes to `_chk_asarray` and `spearmanr` tests. b7b3ab0
@aeklant aeklant MAINT: fixed what seems to be an autocomplete typo in `ttest_rel` tests. 06bd2ec
@ewmoore ewmoore MAINT: rebase changes a31efd9
Commits on Apr 07, 2015
@ewmoore ewmoore BUG: fragile conversion from scalar to 1D array in lfilter
Use the ideomatic np.atleast_1d to check for and convert scalars
to 1D arrays in lfilter. Using np.isscalar fails if given a
decimal.Decimal scalar.
1a0a348
@ewmoore ewmoore BUG: fix broadcasting in lfilter with some singleton dims.
fixes gh-4681
643cd58
@ewmoore ewmoore MAINT: quiet warnings from cephes. d4aba4c
@wuhaochen wuhaochen Update doc of cluster.hierarchy.linkage
Indicate the ability to use customized metric in linkage function.
b809f84
@wuhaochen wuhaochen DOC: Remove outdated param milu
Remove outdated parameter in sparse.linalg.spilu.
3b5349e
@wuhaochen wuhaochen Merge pull request #1 from wuhaochen/doc_linalg.spilu
DOC: Remove outdated param milu
0275fc0
Commits on Apr 08, 2015
@ewmoore ewmoore Merge pull request #4587 from raoulbq/hermite_asymptotic
Hermite asymptotic
e353206
Commits on Apr 09, 2015
@raoulbq raoulbq Add citation of published paper
Fast computation of Gauss quadrature nodes and weights on the whole real line
0dce923
@alexbrc alexbrc MAINT: fix issues with np.matrix as input to functions related to kmeans 2152f79
Commits on Apr 10, 2015
@ewmoore ewmoore Merge pull request #4703 from raoulbq/hermite_citation
Add citation of published paper
450d1af
@ortylp ortylp Protect against crash in scipy.io.read() in case there are unknown ch…
…unks at the end of file

Real life case: _skip_unknown_chunk was called when fid.tell() == fsize condition was true
in this case reading of 4 bytes failed and call to struct.unpack() throws exception (and thus crashing the application)
ef3d462
Commits on Apr 11, 2015
@rgommers rgommers Merge pull request #4696 from ewmoore/lfilter_broadcast
BUG: fix singleton expansion in lfilter.
60e0028
@pv pv ENH: sparse.linalg: improve speed and memory usage of onenormest
Replace Python sort operations by Numpy sort operations (significantly
smaller memory usage and higher speed).  Make some operations
block-wise, to avoid large temporary arrays. Delete local variables when
possible to avoid keeping unused large arrays around.
fc8c969
@pv pv BENCH: make onenormest benchmark test also larger sparse matrices 5d1630f
@rgommers rgommers Merge pull request #4694 from anntzer/anderson-darling
ENH: vectorization of Anderson-Darling computations.
b4652f5
@ev-br ev-br MAINT: interpolate: no need to pass around dtype if it can be inferred ad4d4ec
Commits on Apr 12, 2015
@rgommers rgommers Merge pull request #4249 from ewmoore/updateqr2
ENH: add routines for updating QR decompositions
02ca7d2
@rgommers rgommers MAINT: polish signal.place_poles a bit - few incorrect tests, clarify…
… doc.

Thanks to Irvin Probst for pointing these out.
bf592c2
@rgommers rgommers MAINT: PEP8, pyflakes and test docstring->comment cleanups in test_lt…
…isys.py
4cc5d04
@Axiru Axiru BUG: fix issue with size-1 input to signal.invres/invresz ba1ff7b
@rgommers rgommers Merge pull request #4646 from axiru/master
BUG: fix issue with size-1 input to signal.invres/invresz
c0bb908
@rgommers rgommers DEP: deprecate mstats.signaltonoise, to match stats.signaltonoise dep…
…recation.

The stats.signaltonoise deprecation was done in gh-4656.
This also cleans up some testsuite noise.
92acd7e
@aeklant aeklant MAINT: maintenance changes to `stats.spearmanr` and corresponding tests. 5c2dc57
@rgommers rgommers Merge pull request #4693 from aeklant/_chk_asarray
ENH: added functionality to handle scalars in `stats._chk_asarray`
a731bdf
@ortylp ortylp Add documentation for skipping unpack() and seek() at EOF in _skip_un…
…known_chunk()
8ef942c
@ev-br ev-br ENH: stats: can specify f-shapes to fix in fitting by name
For instance, beta.shapes='a, b'. Then
beta.fit(..., fa=1) is exactly equivalent to beta.fit(..., f0=1)
93addf6
@ev-br ev-br Merge pull request #4716 from rgommers/testsuite-noise
DEP: deprecate mstats.signaltonoise ...
08bc625
@ev-br ev-br Merge pull request #4698 from ewmoore/special_warnings
MAINT: quiet warnings from cephes.
33af809
Commits on Apr 13, 2015
@f0k f0k Document that imresize converts the input to a PIL image 83d1a09
@ewmoore ewmoore MAINT: PyArray_BASE is not an lvalue unless the deprecated API is used.
Use PyArray_SetBaseObject for numpy 1.7 and up and use the older
API for numpy 1.6.2.

fixes gh-4719
7f2f092
@alexbrc alexbrc MAINT: use _asarray_validated instead of ad-hoc validation 8e22d9b
@rgommers rgommers Merge pull request #4722 from ewmoore/signal_build
MAINT: PyArray_BASE is not an lvalue unless the deprecated API is used.
7d6f47d
@rgommers rgommers MAINT: add linalg/_decomp_update.pyx to .gitignore
[ci skip]
8fd588d
@rgommers rgommers Merge pull request #4727 from rgommers/update-gitignore
MAINT: add linalg/_decomp_update.pyx to .gitignore
b259eaa
@alexbrc alexbrc DOC: add poch to the list of scipy special function descriptions c113c50
@WarrenWeckesser WarrenWeckesser Merge pull request #4728 from argriffing/document-poch
DOC: add poch to the list of scipy special function descriptions
e23a37c
Commits on Apr 14, 2015
@rgommers rgommers Merge pull request #4707 from argriffing/vq-np-matrix
MAINT: fix issues with np.matrix as input to functions related to kmeans
c0f54db
ldamewood MAINT: Fix gengamma _nump failure at large values of `a`
fixes gh-4724
a7ded12
@ev-br ev-br Merge branch 'pr/4725'
Reviewed at #4725
4f65e8e
@ev-br ev-br MAINT: stats: avoid (a spurious) division-by-zero in skew
Use _lazywhere instead of ``np.where(m2==0, 0, m3/m2**1.5)``
d338161
@rgommers rgommers Merge pull request #4735 from ev-br/np_dev_err
MAINT: stats: avoid (a spurious) division-by-zero in skew
26ac155
@aeklant aeklant TST: silence runtime warnings for some corner cases in `stats` tests. c65ff98
@alexbrc alexbrc BLD: try to build numpy instead of using the one on TravisCI 7aa911f
@WarrenWeckesser WarrenWeckesser DOC: Update some docstrings with 'versionadded'.
Also added 'reverse_cuthill_mckee' and 'maxmimum_bipartite_matching' to the
list of the functions in the 'scipy.sparse.csgraph' module docstring.
ea5a2a6
@ev-br ev-br Merge pull request #4739 from argriffing/fix-relaxed-strides
BLD: try to build numpy instead of using the one on TravisCI
0731013
@ev-br ev-br BLD: make sure that relaxed strides checking is in effect on TravisCI
Fail the build otherwise. (EG, if it has been disabled by TravisCI
env changes, gh-4737, gh-4739)
5479239
@argriffing argriffing Merge pull request #4742 from ev-br/assert_relaxed_strides
BLD: make sure that relaxed strides checking is in effect on TravisCI
db06d73
Commits on Apr 15, 2015
@rgommers rgommers Merge pull request #4738 from aeklant/issue_#4734
TST: silence runtime warnings for some corner cases in `stats` tests.
cce72f6
@argriffing argriffing Merge pull request #4740 from WarrenWeckesser/versionadded
DOC: Update some docstrings with 'versionadded'.
d9ff7c4
Commits on Apr 16, 2015
@argriffing argriffing Merge pull request #4712 from pv/onenormest-memfix
Reduce onenormest memory usage
6885006
@argriffing argriffing Merge pull request #4715 from rgommers/maint-place-poles
MAINT: polish signal.place_poles and signal/test_ltisys.py
6c59141
@ev-br ev-br DOC: special: TeX typesetting of rel_entr, kl_div and pseudo_huber
closes gh-4732
4b9f1d4
Commits on Apr 17, 2015
@argriffing argriffing Merge pull request #4660 from lightcatcher/avoid_extra_slice_copy
Avoid extra copy for sparse compressed [:, seq] and [seq, :] slices.
3e82f1a
@alexbrc alexbrc BENCH: add sparse null slice benchmark 72c614a
@argriffing argriffing Merge pull request #4750 from ev-br/spec_backsl
DOC: special: TeX typesetting of rel_entr, kl_div and pseudo_huber
e92ac8f
@insertinterestingnamehere insertinterestingnamehere MAINT: Require Cython 0.22 in the installation docs and the cythonize
script.
Also use Cython 0.22 in the benchmark suite.
9f4903e
@insertinterestingnamehere insertinterestingnamehere BUG: Fixed compilation failures with recent versions of Cython on win…
…dows.
1eb34a2
@argriffing argriffing Merge pull request #4699 from wuhaochen/master
DOC: cluster.hierarchy.linkage and sparse.linalg.spilu
6cdd087
@argriffing argriffing Merge pull request #4713 from ev-br/no_dtype
MAINT: interpolate: no need to pass dtype around if it can be inferred
989c75f
Commits on Apr 20, 2015
@alexbrc alexbrc DOC: optimize.leastsq default maxfev clarification 4702e2b
@alexbrc alexbrc MAINT: better generic initialization of location and scale parameters…
… for mle
4d60815
@argriffing argriffing Merge pull request #4718 from ev-br/fshapes
ENH: stats: can specify f-shapes to fix in fitting by name
4a41fff
Commits on Apr 21, 2015
@larsmans larsmans Merge pull request #4751 from argriffing/bench-sparse-null-slice
BENCH: add sparse null slice benchmark
8c2ba28
@dlax dlax Merge pull request #4758 from argriffing/leastsq-docstring
DOC: optimize.leastsq default maxfev clarification
acd131d
Commits on Apr 22, 2015
@insertinterestingnamehere insertinterestingnamehere MAINT: Specify Cython version in Travis build. 2ccca3a
@rgommers rgommers Merge pull request #4753 from insertinterestingnamehere/compile_fix
BUG: Fixed compilation with recent Cython versions.
14a6a5b
@alexbrc alexbrc DOC: linprog copyediting 90986de
Commits on Apr 23, 2015
@alexbrc alexbrc MAINT: fix superficial linprog bugs 361ddfd
@dlax dlax Merge pull request #4763 from argriffing/linprog-cleanup
fix small linprog bugs
f9810b9
@Andreea-G Andreea-G BUG: Fixes #4594: scipy.optimize.linprog IndexError when a callback i…
…s provided

The bug appeared at the beginning of Phase 2, if Phase 1 successfully terminated with some of the artificial variables as pivots. These artificial variables were zero because the corresponding element in the last column of T was zero, which allowed all artificial variables to be eliminated and Phase 1 to terminate successfully. However, basis[:m] contained elements that were greater than the length of solution, which resulted in IndexError at solution[basis[:m]]. The remaining true (non-artificial) variables that are not included in basis[:m] are also zero. Therefore simply setting the length of solution to be large enough solves the problem.
9438fb7
@Andreea-G Andreea-G BUG: Fixes #4746: linprog returns solution violating equality constraint
This adds to pull request #4685. The problem happens if some artificial variable(s) remain basic at the end of Phase 1. PR #4685 solved the problem if the row in T associated with the artificial variable was zero (for all columns in T that correspond to a non-artificial variable). When there are non-zero elements in that row, then one of the non-zero elements is picked for pivot, and the artificial variable is replaced in the basis by the non-artificial variable. This is done until there are no more artificial variables in the basis that have non-zero rows in T.
260f6a8
Commits on Apr 24, 2015
@argriffing argriffing Merge pull request #4762 from Andreea-G/linprog_artificial
BUGS: Fixes #4746 and #4594: linprog returns solution violating equality constraint and IndexError when a callback is provided
bca1d9c
Commits on Apr 25, 2015
@Andreea-G Andreea-G Fixes #4726: test_cobyla.test_vector_constraints
Set rhoend and tol as 1e-5 in test_cobyla for fmin_cobyla and minimize.
5b802a1
@dlax dlax Merge pull request #4769 from Andreea-G/cobyla_rhoend
Fixes #4726: test_cobyla.test_vector_constraints
67dce17
Commits on Apr 26, 2015
@astrofrog astrofrog BUG: io.idl: don’t crash if a pointer reference is not found in the h…
…eap.

Instead, set variable to None and emit a warning.  Closes gh-4613.
e83e981
@astrofrog astrofrog TST: added regression test for io.readsav issue with invalid pointers. d0645f7
@rgommers rgommers Merge pull request #4710 from astrofrog/fix-idl-pointer
scipy.io.idl: make reader more robust to missing variables in heap
100ce74
@rgommers rgommers Merge pull request #4721 from f0k/patch-1
DOC: document that imresize converts the input to a PIL image
e4b5787
@rgommers rgommers Merge pull request #4711 from ortylp/master
Fix crash for unknown chunks at the end of file
49f0cb3
@daggerdagr daggerdagr edited scipy/stats/stats.py to fix doctest for fisher_exact 5edc08c
@rgommers rgommers Merge pull request #4771 from daggerdagr/master
DOC: fix documentation error for fisher_exact
bcf5fcd
Commits on Apr 27, 2015
@alexbrc alexbrc TST: remove chi from list of failing fits 87d128c
@ev-br ev-br Merge pull request #4759 from argriffing/stats-fit-support
improved stats mle fit
fce7061