townforge/math
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
5,028 Commits 1 Branch 0 Tags 151 MiB
cc
Commit Graph

2240 Commits

Author SHA1 Message Date
Paul A. Bristow
4dd9452c22 Corrected failure to check for NaN in quantiles. UNsure if df should include zero. 
[SVN r78776]
 2012-05-31 09:35:52 +00:00
John Maddock
392fbeeddd Add extra error tests for negative_binomial to laplace distros. 
Refs #6934.

[SVN r78773]
 2012-05-30 18:15:41 +00:00
John Maddock
41bd4bd442 Update to use new out-of-bounds error checking. 
Refs #6934.

[SVN r78771]
 2012-05-30 16:52:18 +00:00
Paul A. Bristow
a72f149301 Changed range to infinity if supported. 
[SVN r78768]
 2012-05-30 16:35:59 +00:00
Paul A. Bristow
7ef962fc6b Changed range to include infinity if supported. 
[SVN r78766]
 2012-05-30 16:31:18 +00:00
Paul A. Bristow
187960b9a7 Corrected checks missing (). 
[SVN r78765]
 2012-05-30 16:30:45 +00:00
John Maddock
bc2dc1d4dd Add boilerplate code for testing most domain errors in the distributions. 
Update the non-central distributions to make use of it.

[SVN r78751]
 2012-05-29 17:43:16 +00:00
John Maddock
3e9242ade2 Improve testing of error conditions and fix any errors detected. 
Takes care of distributions normal-weibull.
Refs #6937.

[SVN r78726]
 2012-05-28 17:42:50 +00:00
John Maddock
a1789c4766 Tighten Weibull distro tests. 
Fix corner cases in Weibull distro.
Fixes #6939.
Fixes #6938.

[SVN r78723]
 2012-05-28 16:27:00 +00:00
John Maddock
02787878e6 Update concept checks to unmask issues reported in #6934. 
Add some distributions to the concept tests that were missing before.
Add skew_normal_distribution to distributions.hpp.
Fix some errors in calculating custom policies when the defaults have been changed.
Fix the errors and warnings that result from the above.
Refs #6934.

[SVN r78711]
 2012-05-28 11:05:43 +00:00
Paul A. Bristow
5e0aa58152 Added Policy as a template parameter to functors and #6934 test program now compiles and runs. 
Need to commit changes to concept checks to confirm.

[SVN r78579]
 2012-05-24 11:13:54 +00:00
John Maddock
1a2d4d2c5e Move default arguments to forward declaration. 
Fixes #6930.

[SVN r78559]
 2012-05-23 17:18:49 +00:00
John Maddock
382d525d15 Add basic versions of the Airy functions. 
Add some extra jacobi, hankel and owens_t tests and fix resulting errors.
Allow remez code to work with expression template types.
Regenerate docs.

[SVN r78527]
 2012-05-21 09:45:48 +00:00
John Maddock
056745573f Add the Jacobi elliptic special functions. 
[SVN r78492]
 2012-05-17 09:19:57 +00:00
John Maddock
2fdb5f551f Quash some minor warnings. 
[SVN r78422]
 2012-05-11 17:13:03 +00:00
John Maddock
9834bafe7e Add needed forward declarations (keeps GCC-4.7.0 happy). 
[SVN r78400]
 2012-05-10 09:16:26 +00:00
John Maddock
be63e63891 Fix a pair of typos (one a clear bug). 
[SVN r78316]
 2012-05-03 17:20:24 +00:00
Paul A. Bristow
b0021e4c28 Added constant ln_ten 
[SVN r78288]
 2012-05-01 10:42:14 +00:00
John Maddock
aa64d8dc00 Add forwarding-policy to hankel functions. 
Updated docs with links to more info.

[SVN r78256]
 2012-04-29 16:54:10 +00:00
John Maddock
be93fe3b95 Rename hankel_hX to hankel_X. 
Update tests and docs for Hankel functions.

[SVN r78253]
 2012-04-29 12:39:28 +00:00
John Maddock
d099a2bbf0 Add first draft of Hankel functions. 
Add some Math lib docs that were previously missing.

[SVN r78245]
 2012-04-28 17:59:37 +00:00
John Maddock
ac0d76900a Add code to force initialization of constants prior to the start of main(). 
Rename "do_nothing" to "force_instantiate" which better describes what it does.
Fix up the resulting failures.

[SVN r78193]
 2012-04-25 17:16:09 +00:00
John Maddock
e1fc40cf11 Yet another try at getting constexp usage correct! 
Refs #6820.

[SVN r78160]
 2012-04-23 15:13:10 +00:00
John Maddock
73bd07a44d Fix constexp usage. 
[SVN r78133]
 2012-04-22 11:17:08 +00:00
John Maddock
ec742b4b80 Merge Owen's T function and Skew Normal code from sandbox and fix whatever failures show up. 
Regenerate docs.

[SVN r78130]
 2012-04-22 10:23:03 +00:00
John Maddock
ad520dbbcb Disable real_concept tests for IBM's compiler. 
[SVN r78108]
 2012-04-21 08:58:09 +00:00
John Maddock
f149d4cf53 Fix double-definition of float_t in tr1.hpp. 
Tweak behaviour of BOOST_MATH_BIG_CONSTANT.
Fix expected error rates for a bunch of tests (mostly on Mingw).
Fix paths in owens_t.qbk.

[SVN r78107]
 2012-04-21 08:43:06 +00:00
John Maddock
15baa10fac Add checks to discrete quantile finder to handle cases where there is no true answer. 
Fixes #6805.

[SVN r78090]
 2012-04-20 10:51:55 +00:00
John Maddock
85d55b2ea5 Update test_data.hpp to improve definition of SC_ so it's compatible with multiprecision testing. 
Update BOOST_MATH_BIG_CONSTANT to use a long double when the result is long double or smaller.

[SVN r78042]
 2012-04-17 16:05:52 +00:00
John Maddock
d94751ba33 Fix numeric error in bessel_jy.hpp. 
Remove some dead code from bessel_jy.hpp.
Add additional test cases for above bug.
Update and regenerate docs to match.

[SVN r77513]
 2012-03-24 11:59:21 +00:00
John Maddock
14b97d2dd4 Add missing include. 
[SVN r77350]
 2012-03-16 16:55:22 +00:00
John Maddock
940c3824c8 Change put_num_and_fill to work on CharType not char, so that formatting of signed zeroes is wide character safe even if '0' is not used for zeroes. 
[SVN r77286]
 2012-03-09 18:36:29 +00:00
John Maddock
b745e80ede Fix warning with a cast rather than a blanket disable. 
[SVN r77283]
 2012-03-09 09:35:36 +00:00
Paul A. Bristow
fec1f2407e Version that passes new tests for both default and signed_zero flag set based on modifications by KR Walker. 
BUT had to suppress MSVC warning "C4244 conversion wchar_t* to char* may cause data loss." 

[SVN r77268]
 2012-03-08 18:12:28 +00:00
John Maddock
6fc9eb72a9 Fix GCC some warnings, plus one failure. 
[SVN r76821]
 2012-02-01 13:06:17 +00:00
Paul A. Bristow
a94442f5de Added tests for new constants two_div_pi and root_two_div_pi 
[SVN r76820]
 2012-02-01 12:32:22 +00:00
John Maddock
20f26bd52c Use underscores in attribute names. 
Fixes #6411.

[SVN r76773]
 2012-01-29 12:27:53 +00:00
John Maddock
59eb0f44df Don't be too quick to disable long double functions if __NO_LONG_DOUBLE_MATH is set by GLIBC. 
Fixes #6459.

[SVN r76764]
 2012-01-28 19:13:44 +00:00
John Maddock
682d60c363 Merge new math constants code from the sandbox. 
Regenerate docs.
Fix build failures caused by merge.

[SVN r76736]
 2012-01-28 12:13:17 +00:00
John Maddock
887f590fcd Refactor the tests some more: allow tables of data to be of a different type to the type being tested - allows us to have tables of POD's (fast to compile) rather than tables of UDT's which compile very slowly on some compilers (usually Intel). 
[SVN r76544]
 2012-01-16 18:25:02 +00:00
John Maddock
c5bdd41539 Correct overflow logic in Bessel Y and add new test case. 
Fixes #6367.

[SVN r76373]
 2012-01-08 18:51:23 +00:00
John Maddock
fecae229b0 Simplify/improve zeta function still further. 
[SVN r76337]
 2012-01-07 16:57:51 +00:00
John Maddock
532ddf8cb0 Change template argument "L" to "Lanczos" to avoid conflicts with L as a macro. 
[SVN r76278]
 2012-01-02 17:44:19 +00:00
John Maddock
9005ade2a3 Updated big_lanczos.hpp to use new BIG_CONSTANT macros. 
Fix expression template issue in zeta.hpp.

[SVN r76251]
 2012-01-01 11:34:50 +00:00
John Maddock
9922e9134c Remove unused constant. 
Fixes #5982.

[SVN r76099]
 2011-12-21 17:58:28 +00:00
John Maddock
4d57f3e642 Disable SSE2 instruction support under the CUDA compiler. 
Fixes #6234.

[SVN r76097]
 2011-12-21 17:43:55 +00:00
John Maddock
a4a1cca095 Change overflow tests to not generate an infinity. 
Fixes #6307.

[SVN r76095]
 2011-12-21 16:58:36 +00:00
John Maddock
02e5447867 Delete unneeded variable. 
Fixes #6307.

[SVN r76094]
 2011-12-21 16:46:46 +00:00
John Maddock
1a304dccb0 Apply patch from #6192. 
Plus fix a few other issues that arise when BOOST_NO_LIMITS_COMPILE_TIME_CONSTANTS is set.
Fixes #6192.

[SVN r76093]
 2011-12-21 16:37:32 +00:00
John Maddock
af0c106290 Update zeta implementation to use much faster algorithm. 
Change zeta test to use external template instances.

[SVN r76092]
 2011-12-21 16:07:37 +00:00
John Maddock
70f902d332 Misc small workarounds to allow multiprecision tests to function. 
[SVN r76080]
 2011-12-20 16:21:44 +00:00
John Maddock
2902c06d6c Add missing using declarations. 
[SVN r76041]
 2011-12-18 16:05:55 +00:00
John Maddock
381bddafa0 Finish moving test cases into headers. 
Fix array declarations so GCC doesn't warn about them.
Declare constants in headers so they can be used by UDT's larger than type long double.
Suppress a few warnings and fix a couple of bugs that showed up when testing with UDT's.

[SVN r75960]
 2011-12-15 11:23:35 +00:00
John Maddock
434d53bc83 Fix complex number routines to work with signed zeros, changes involve: 
* Use boost::math::signbit rather than comparison to zero.
* Use boost::math::changesign rather than unary negation (unary negation fails for Intel on Linux when the argument is a zero).
* Update to use boost::math::isnan/isinf rather than the old code.
* Update to use boost::math constants.
Fixes #6171.

[SVN r75933]
 2011-12-13 18:11:46 +00:00
John Maddock
8d9372fac9 Begin refactoring math lib test cases as template so they can used by other numerical types. 
Change constant declarations in the Math lib headers to give full digits when used with UDT's that are constructible from strings.

[SVN r75820]
 2011-12-06 09:38:16 +00:00
Paul A. Bristow
cbe12e231d Added "two_thirds" and deprecate "twothirds" Trac 6199 
[SVN r75773]
 2011-12-02 15:01:40 +00:00
Paul A. Bristow
09c2238e7a Corrected complement cdf with Yan Zhou patch. 
[SVN r75591]
 2011-11-21 12:57:33 +00:00
John Maddock
c1e5af9291 Warning suppression. 
[SVN r75585]
 2011-11-21 09:46:33 +00:00
John Maddock
2d5ba01092 Tentative fix for platforms that have fenv.h but no fexcept_t. 
[SVN r75486]
 2011-11-14 12:20:13 +00:00
John Maddock
9373287faa More fixes for expression template types. 
[SVN r75403]
 2011-11-08 12:49:27 +00:00
John Maddock
a8d7299861 Allow distribution parameter checking to be eluded if it's already taken place. 
[SVN r75196]
 2011-10-31 16:38:16 +00:00
John Maddock
c43891eab0 Allow precision.hpp to work with number types in base 10. 
Fix binomial_distribution quantile when success_fraction is 1.
Fixes #6069.

[SVN r75168]
 2011-10-29 17:26:26 +00:00
John Maddock
34a635b621 Fix Clang build issue. 
[SVN r75152]
 2011-10-28 08:29:55 +00:00
John Maddock
681b9c1288 Add missing #include. 
Fixes #6017.

[SVN r74937]
 2011-10-13 17:31:00 +00:00
John Maddock
b7df32ac03 Apply patch from ticket #6014. 
Fixes #6014.

[SVN r74927]
 2011-10-12 17:09:48 +00:00
John Maddock
48e179266b Fix error checking for trunc and round functions. 
Update tests to match.

[SVN r74925]
 2011-10-12 11:06:09 +00:00
John Maddock
f617e8e284 Change nextafter and related functions to handle infinities as arguments the same way as GLIBC and MSVC. 
Fixes #5823.

[SVN r74918]
 2011-10-11 17:28:13 +00:00
John Maddock
a5941a2179 GCC workaround when using TR1 tuple. 
Fixes #5934.

[SVN r74891]
 2011-10-10 12:09:12 +00:00
John Maddock
303503d876 Remove unused constants. 
Fix expression for tolerance in eps_tolerance.
Add include guards to prime.hpp.
Fixes #5982.
Fixes #5927.
Fixes #6001.

[SVN r74890]
 2011-10-10 11:52:05 +00:00
John Maddock
f52587e616 More fixes for UDT's as arguments. 
[SVN r74802]
 2011-10-08 12:30:50 +00:00
John Maddock
8f22d5ebe5 Minor fixes for expression-template types. 
[SVN r74572]
 2011-09-26 11:01:03 +00:00
Paul A. Bristow
a74236ad41 std qualified strlen in int width = (int)std::strlen(prefix) + (int)std::strlen(body); 
to comply with C++ standard.

[SVN r74480]
 2011-09-20 18:03:22 +00:00
John Maddock
6e78e00321 Fix mpreal #includes. 
Add some more expression-template fixes.
Add fix for Cray platform in SSE2 support selection.

[SVN r74389]
 2011-09-15 17:16:20 +00:00
John Maddock
be97e03f4e Fix expression template compilation issues. 
[SVN r74355]
 2011-09-12 08:36:54 +00:00
John Maddock
acb59c8512 Fix usage with expression template types. 
Fix fpclassify code when both BOOST_MATH_DISABLE_STD_FPCLASSIFY and isnan are defined.
Remove dead code.

[SVN r74185]
 2011-09-01 18:03:31 +00:00
John Maddock
990c183d8d Fix error handling arguments. 
[SVN r74184]
 2011-09-01 17:47:01 +00:00
John Maddock
c416cd2772 Fix bug in e_float bindings - negative integers were incorrectly converted. 
[SVN r73647]
 2011-08-11 11:02:12 +00:00
Paul A. Bristow
92b23c16a6 Corrected wrong code in uniform.hpp for >upper or < lower 
wrong test values in uniform cdf complemented.

Trac ticket #5733

[SVN r73360]
 2011-07-25 16:49:34 +00:00
John Maddock
38b6a4f3b8 Revert BSD fix - it breaks Boost.TR1 and was nonsense anyway! :( 
[SVN r72821]
 2011-06-30 16:04:54 +00:00
John Maddock
4cf83544cd Add new series evaluation forms for the Bessel J and Y functions. 
Also take extra care when evaluating to avoid spurious overflow/underflow.
Changed check_series_iterations to know the real-number type so messages are properly formatted.
Updated docs to match and regenerated.
Fixes #5560.

[SVN r72415]
 2011-06-05 16:56:17 +00:00
John Maddock
3b71dc32eb Update Bessel functions to use series representations when x is small compared to v and also to prevent the generation of NaN's in such cases. 
Update docs to match.
Fixes #5560.

[SVN r72221]
 2011-05-27 17:24:08 +00:00
John Maddock
ef7f0bb940 Remove unnecessary macros. 
[SVN r71989]
 2011-05-16 17:30:52 +00:00
John Maddock
74d540d59a Move TR1 C-compatible macros to separate file so they don't get indexed by the docs. 
[SVN r71988]
 2011-05-16 17:28:48 +00:00
John Maddock
1f6642b388 Add initial version of e_float support. 
[SVN r71674]
 2011-05-02 18:21:28 +00:00
John Maddock
a5239fa567 Suppress a whole bunch of new gcc-4.6 warnings. 
Add code to verify that the radix of the type is 2 wherever that assumption is made.
Add some more instrumentation code.

[SVN r71672]
 2011-05-02 15:36:51 +00:00
John Maddock
d27407f320 Tentative fix for FreeBSD failure. 
[SVN r71534]
 2011-04-27 16:00:00 +00:00
John Maddock
1982cb4a4c Fix remaining Intel regressions. 
Add prototypes to fwd header.
Add new functions to the concept checks.

[SVN r71432]
 2011-04-23 08:17:10 +00:00
John Maddock
7f1a45870a Change fp_traits to handle sign manipulation separately from fp-classification. 
Fixes some Linux/GCC regressions but not yet all.

[SVN r71405]
 2011-04-21 18:11:37 +00:00
Paul A. Bristow
15c9b19ac1 Added more comments 
[SVN r71404]
 2011-04-21 17:53:58 +00:00
Paul A. Bristow
5cb894721b Added native_tag specialisation to try to cure failures on gcc. 
[SVN r71389]
 2011-04-20 08:49:02 +00:00
Paul A. Bristow
caf05d27cd Added boost::math::detail::changesign_impl(x, method()); to try to get to work with gcc. 
[SVN r71381]
 2011-04-19 18:04:58 +00:00
Paul A. Bristow
e3e838da6c nonfinite facets 
[SVN r71371]
 2011-04-18 16:19:48 +00:00
Paul A. Bristow
1da961d8e7 Changesign addded. 
[SVN r71370]
 2011-04-18 16:19:26 +00:00
John Maddock
c2faf71f7d Fix value returned for atanh(1). 
Fixes #5385.

[SVN r70692]
 2011-03-29 11:17:58 +00:00
John Maddock
bac00ef969 Cast arguments to quaternion/octonion constructors so they still work when there's no implicit conversion available. 
Fixes #5369.

[SVN r70569]
 2011-03-26 12:04:02 +00:00
John Maddock
4b3cdb252d Don't include fenv.h when BOOST_NO_FENV_H is set. 
Fixes #5356.

[SVN r70392]
 2011-03-22 18:14:00 +00:00
John Maddock
13bd3e54cc MSVC warning suppression. 
[SVN r69509]
 2011-03-03 11:09:23 +00:00
John Maddock
5b6c4bfeb6 Change the name of the extern "C" TR1 functions so they no longer clash with (possibly inline) std lib versions. 
Fixes #4828.

[SVN r68653]
 2011-02-05 10:49:43 +00:00
John Maddock
91471146f1 Fix acosh bug, and update tests and docs accordingly. 
Fixes #5113.

[SVN r68421]
 2011-01-24 19:04:24 +00:00
John Maddock
9744b02e98 Add more debug/diagnostic info and fix bug that manifests itself on SSE x64 machines only. 
Fixes #5095

[SVN r68346]
 2011-01-21 09:42:51 +00:00
John Maddock
9635919e54 Fix for IMB C++ on Linux. 
Fixes #5090.

[SVN r68290]
 2011-01-19 15:05:18 +00:00
Steven Watanabe
3524499fc0 Revert [67111] (addition of boost/detail/iomanip.hpp) and all the commits that depend on it. ([68137], [68140], [68141], [68154], and [68165]). 
[SVN r68168]
 2011-01-15 08:11:51 +00:00
Bryce Adelstein-Lelbach
f11d1c3c2b Replacing the use of <iomanip> with <boost/detail/iomanip.hpp> across Boost. 
On Linux, GNU's libstdc++, which is the default stdlib for icc and clang,
cannot parse the <iomanip> header in version 4.5+ (which thankfully neither
compiler advises the use of yet), as it's original C++98-friendly
implementation has been replaced with a gnu++0x implementation.
<boost/detail/iomanip.hpp> is a portable implementation of <iomanip>, providing
boost::detail::setfill, boost::detail::setbase, boost::detail::setw,
boost::detail::setprecision, boost::detail::setiosflags and
boost::detail::resetiosflags. 



[SVN r68140]
 2011-01-14 02:35:58 +00:00
John Maddock
0ad713690d Fix bug in special-case handling of ibeta_inv. 
Fix concept test failures with mpfr_class in legendre.hpp and non_central_t.hpp.
Fixes #5015.

[SVN r67469]
 2010-12-27 18:37:15 +00:00
Paul A. Bristow
9d8407f38d Add comment after assert for integer template type 
[SVN r67362]
 2010-12-20 15:16:19 +00:00
Paul A. Bristow
595c61e976 Add assert (and comment) for integer template type 
[SVN r67361]
 2010-12-20 15:14:31 +00:00
John Maddock
309e2c3ec5 Add support for "mpreal" wrapper for mpfr. 
Add concept check for mpreal type.
Add previously missing doc file.
Fix type promotion rules to do the right thing when one type is a class type that's implicitly convertible to a real.

[SVN r67282]
 2010-12-17 16:04:51 +00:00
John Maddock
0bfae472d5 Fix min/max usage. 
[SVN r67279]
 2010-12-16 18:37:50 +00:00
John Maddock
555f3d6736 Improve performance of non-central T by simplifying the arithmetic where possible. 
Change error handlers to pass the target type down to "rounding errors".
Change the rounding error handlers to return a well defined value (the maximum representable value in the target type).
Update and complete docs on rounding errors.
Fix typo in test/test_beta_dist.cpp.
Regenerate the docs.
Remove illegal comment from test_constants.cpp.
Fixes #4977.

[SVN r67231]
 2010-12-14 19:22:52 +00:00
Paul A. Bristow
1495ca1ccc Removed inverse_uniform.hpp 
[SVN r67215]
 2010-12-13 16:38:06 +00:00
Paul A. Bristow
7c13ff9124 Removed inverse_uniform.hpp 
[SVN r67214]
 2010-12-13 16:37:33 +00:00
Paul A. Bristow
90fcebc87b use raise_overflow error in place of return NaN 
[SVN r67213]
 2010-12-13 16:36:46 +00:00
Paul A. Bristow
6d604ab6a9 use raise_overflow error in place of return NaN 
[SVN r67212]
 2010-12-13 16:35:59 +00:00
John Maddock
870a1e233b Suppress warnings from external headers. 
[SVN r67149]
 2010-12-10 17:43:38 +00:00
John Maddock
8177da58be Add non-functional stub versions of the skewness and kurtosis to the non-central beta. 
Update tests accordingly, and suppress some compiler warnings.

[SVN r66986]
 2010-12-03 17:55:00 +00:00
John Maddock
7ff88c69f1 Added concept checks for new distros, and fixed resulting errors. 
Added mean and variance to non-central beta.

[SVN r66985]
 2010-12-03 16:23:55 +00:00
Paul A. Bristow
bb7b3d6540 Enabled mean, variance ... for dotnetexample (TODO return NaN?) 
[SVN r66957]
 2010-12-01 18:48:12 +00:00
Paul A. Bristow
68e1194c63 disable 4512 
[SVN r66956]
 2010-12-01 18:47:46 +00:00
Paul A. Bristow
274028b383 comment 
[SVN r66892]
 2010-11-30 14:37:28 +00:00
Paul A. Bristow
1b63ffe8f0 Corrected missing casts int 0 to RealType 
[SVN r66891]
 2010-11-30 14:36:45 +00:00
Paul A. Bristow
4680ece7af using boost/math/tools/tuple.hpp instead. 
[SVN r66822]
 2010-11-28 15:03:13 +00:00
Paul A. Bristow
bab0f2ce24 Changed make_tuple from std::tr1:: to boost::math 
qualified numeric_limits with std.

[SVN r66797]
 2010-11-27 16:11:27 +00:00
Paul A. Bristow
aaaad79bca Changed make_tuple from std::tr1:: to boost::math 
[SVN r66793]
 2010-11-27 16:00:46 +00:00
Paul A. Bristow
93670a3942 removed diagnostics 
[SVN r66766]
 2010-11-26 10:28:58 +00:00
Paul A. Bristow
d0c5796bce Removed wald (now called inverse_gaussian 
[SVN r66754]
 2010-11-25 16:44:14 +00:00
Paul A. Bristow
c0a3413240 Removed inverse_normal 
[SVN r66751]
 2010-11-25 12:11:19 +00:00
Paul A. Bristow
ef4dbab713 Update for new distributions 
[SVN r66731]
 2010-11-24 16:53:25 +00:00
Paul A. Bristow
4fbea7ba6f Added new handling check. 
[SVN r66730]
 2010-11-24 16:51:36 +00:00
Paul A. Bristow
523e4d78f9 Corrections 
[SVN r66729]
 2010-11-24 16:50:41 +00:00
Paul A. Bristow
a3d2de43de Added new distributions. 
[SVN r66728]
 2010-11-24 16:50:06 +00:00
Paul A. Bristow
ec88511ead Update for new constants 
[SVN r66727]
 2010-11-24 16:48:36 +00:00
Paul A. Bristow
2726a04151 Update for new distributions 
[SVN r66726]
 2010-11-24 16:47:58 +00:00
John Maddock
ded3b78a05 Update distribution concept checks to test ad-hoc member functions as well as the generic ones. 
Fix typo in docs and regenerate.

[SVN r66689]
 2010-11-22 19:00:55 +00:00
John Maddock
aee7090820 Fixes #4879. 
[SVN r66685]
 2010-11-22 16:31:42 +00:00
Bryce Adelstein-Lelbach
6d780cab92 Added a detail header for fenv. Please use boost/detail/fenv.hpp in Boost code 
to avoid problems with Clang/glibc.



[SVN r66458]
 2010-11-08 17:10:25 +00:00
Bryce Adelstein-Lelbach
c78863943a Clang hand-holding, mostly for issues caused by Clang defining GCC macros. 
[SVN r66442]
 2010-11-07 20:05:04 +00:00
John Maddock
022bb7b8f1 Tentative fix for WinCE failures. 
[SVN r66070]
 2010-10-18 11:19:06 +00:00
John Maddock
addb71354b Add more workaround code for numeric underflow/overflow. 
[SVN r65714]
 2010-10-01 15:54:03 +00:00
John Maddock
bbc5edfe1c Add concept checks for new distributions. 
Fix Linux gcc failures for new distributions.

[SVN r65694]
 2010-09-30 17:16:14 +00:00
Paul A. Bristow
2707de07b1 checked with better tests. 
[SVN r65658]
 2010-09-29 08:57:51 +00:00
Paul A. Bristow
1e0c9feaf1 layout and comments etc 
[SVN r65657]
 2010-09-29 08:56:06 +00:00
John Maddock
9c9e9eecef Fix integer argument promotion, to simplify code and suppress warnings. 
[SVN r65631]
 2010-09-27 14:36:18 +00:00
Ralf W. Grosse-Kunstleve
49a9039689 boost/math/distributions/inverse_chi_squared.hpp: bug fix (compilation errors with g++ 4) 
[SVN r65516]
 2010-09-21 16:03:23 +00:00
Paul A. Bristow
8d2cfeada9 Add inverse chi square distribution 
[SVN r65485]
 2010-09-20 15:52:41 +00:00
John Maddock
b85756d2cc Improve error handling and fix Linux errors. 
[SVN r65444]
 2010-09-17 18:13:45 +00:00
Paul A. Bristow
05bd2c0043 Add inverse_chi_squared_distribution to fwd list 
[SVN r65425]
 2010-09-15 15:40:44 +00:00
Paul A. Bristow
5c354f4354 1st draft 
[SVN r65424]
 2010-09-15 15:39:43 +00:00
Paul A. Bristow
c8f818f4b8 added x = 0 special case. Example runs OK. Need to rebuild trunk libs to test. 
[SVN r65329]
 2010-09-07 13:57:12 +00:00
John Maddock
0c633f0b6e Change the use of tuples, so that we use std::tuple when available. 
Fix bug in calculation of initial guess in the incomplete gamma inverse.
Fix illegal comment in inverse_gamma.hpp source file.
Update docs to reflect tuple changes.
Regenerate docs.
Add Jamfile for the tools directory and make sure everything still builds cleanly.

[SVN r65320]
 2010-09-06 15:13:23 +00:00
Paul A. Bristow
5dffa5f7ad Add several new distributions 
[SVN r65122]
 2010-08-30 09:02:02 +00:00
John Maddock
434414f58e Added workaround for Intel compiler, which has sporadic issues with fpclassify on Linux. 
Fixes #4592.

[SVN r65015]
 2010-08-26 12:29:12 +00:00
John Maddock
8d719c3ff4 Prevent macro expansion of isfinite. 
[SVN r64983]
 2010-08-24 17:13:04 +00:00
Paul A. Bristow
4d2e34c523 new inverse gamma distribution 
[SVN r64969]
 2010-08-23 11:58:39 +00:00
Paul A. Bristow
80911e889a added inverse-gamma distribution 
[SVN r64968]
 2010-08-23 11:55:51 +00:00
Paul A. Bristow
bbff4a61c6 T result = 0; // Just to avoid warning C4701: potentially uninitialized local variable 'result' used 
[SVN r64889]
 2010-08-18 15:52:13 +00:00
Paul A. Bristow
35655664a7 warning C4244 suppressed by using static_cast<realType>(0) 
[SVN r64888]
 2010-08-18 15:23:37 +00:00
Paul A. Bristow
50dd77b5a8 warning C4244 suppressed by using static_cast<realType> 
[SVN r64886]
 2010-08-18 13:10:19 +00:00
John Maddock
4372c68f55 Fix gcc-3.2.x compiler errors. 
Fixes #4507.

[SVN r64789]
 2010-08-13 17:26:28 +00:00
John Maddock
b11676fa67 Added test cases for issue #4504. 
Tidied up inverse-Student's t evaluation a touch.
Fixes #4504.

[SVN r64597]
 2010-08-04 15:47:30 +00:00
John Maddock
94a309749f Fix typo in polynomial approximations. 
Refs #4504.

[SVN r64577]
 2010-08-03 17:31:20 +00:00
John Maddock
999f37a21e Fix signed/unsigned comparison warnings when char is unsigned. 
[SVN r63496]
 2010-07-01 17:47:51 +00:00
John Maddock
f9562e1c92 Oops missed this file in last commit. 
[SVN r63424]
 2010-06-29 08:24:29 +00:00
John Maddock
2ed2a883ba Make the compile time gcd facilities work with greatest integer types, and be mpl compatible. 
[SVN r63408]
 2010-06-28 15:32:19 +00:00
Paul A. Bristow
95d3d4d406 Corrected support minimum to 0 (was 1). 
[SVN r62380]
 2010-06-02 14:49:22 +00:00
John Maddock
ac3a1ef164 Change to use new symbol visibility macros. 
[SVN r62200]
 2010-05-25 14:45:54 +00:00
John Maddock
200b0f7717 Fix cbrt's internal promotion rules. 
Update tests to match.

[SVN r62087]
 2010-05-18 17:24:01 +00:00
John Maddock
7ab7d21eae Fix failures when used with an expression-template enabled number type such as the gmpxx clases. 
Add additional concept check for integer code using gmp classes.
Fixes #4139.

[SVN r61837]
 2010-05-07 15:36:09 +00:00
John Maddock
c7e0965ce9 Previous fix doesn't work for later GCC versions, revert. 
[SVN r61815]
 2010-05-06 12:33:40 +00:00
John Maddock
0dd75b6c47 Prevent multiple definitions of BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS. 
[SVN r61807]
 2010-05-06 10:23:56 +00:00
John Maddock
45db115f9d Disable long double support when the C++ runtime does not support it either. 
Fixes #3825.

[SVN r61806]
 2010-05-06 10:21:18 +00:00
Douglas Gregor
81c8795897 Eliminate redundant (and incorrect) using declarations at function scope. 
[SVN r61700]
 2010-04-30 07:10:56 +00:00
Douglas Gregor
9257a12d1b Eliminate a redundant BOOST_MATH_STD_USING; we aren't allowed to redeclare declarations in function scope 
[SVN r61595]
 2010-04-26 22:57:02 +00:00
John Maddock
7a7e393485 Fix boost::math::round to handle halfway cases correctly. 
Update tests accordingly.
Update boost::math::fmod docs to make sign and rounding direction clear.

[SVN r61252]
 2010-04-13 16:50:46 +00:00
John Maddock
cd64856ff5 Improve the performance of the Bessel functions, and update docs. 
[SVN r59274]
 2010-01-27 13:16:14 +00:00
John Maddock
ea35733550 Add tgamma1pm1 speed test. 
Suppress some more gcc warnings: in particular make sure that our C99 functions have the same throw specification as the functions in math.h.

[SVN r59129]
 2010-01-18 17:07:03 +00:00
John Maddock
1a5044497d Change cbrt implementation to use a better performing algorithm. 
Fix a few warnings along the way.

[SVN r59095]
 2010-01-17 17:28:34 +00:00
John Maddock
488701a7e3 Suppress unused parameter warning. 
[SVN r57509]
 2009-11-09 12:26:02 +00:00
John Maddock
75f47f2cad remove surplus const-qualifiers. 
[SVN r57354]
 2009-11-04 12:15:05 +00:00
John Maddock
ba698c2297 Improve ibeta error handling, update and regenerate docs. Update tests. 
Fixes #3533.

[SVN r57045]
 2009-10-21 15:53:28 +00:00
John Maddock
cc52a19fb4 Improve ibeta error handling, and add new tests. 
[SVN r57040]
 2009-10-21 12:27:38 +00:00
John Maddock
ebc86f03d8 Update performance test results. 
Adjust msvc performance tuning.

[SVN r56818]
 2009-10-14 12:04:07 +00:00
John Maddock
21cd5f0d9f Update workarounds for Sun's latest compiler release. 
[SVN r56711]
 2009-10-11 12:20:21 +00:00
John Maddock
581f74e09b Fix duplicate definitions when BOOST_NO_EXPLICIT_FUNCTION_TEMPLATE_ARGUMENTS is defined. 
[SVN r56679]
 2009-10-09 16:52:27 +00:00
K. Noel Belcourt
fe4001165c Add PGI macro to get proper alignment. 
[SVN r56653]
 2009-10-08 17:52:10 +00:00
John Maddock
799b88be52 Change iteration limits to use get_epsilon(). 
[SVN r56650]
 2009-10-08 16:03:25 +00:00
John Maddock
2e07572faf Add a check on iterations when Halley iterating. 
Correct igamma inverse initial calculations, and adjust digits needed for convergence.
Fix erf calculation limits.
Adjust non-convergence behaviour in Halley iteration.
For issue #3408.

[SVN r56598]
 2009-10-05 17:36:56 +00:00
John Maddock
71940bec79 Another round of performance tweaks for issue #3408. 
These should make our igamma implementation comparable in performance to the dcdflib FORTRAN routine - at least as far as MSVC is concerned.

[SVN r56503]
 2009-10-01 16:03:42 +00:00
John Maddock
8418e9d68e A few more minor performance tweaks for issue #3407. 
[SVN r56455]
 2009-09-28 17:06:39 +00:00
John Maddock
056d083a2f Changes relating to issue #3408. 
Add hooks for the dcdflib to the incomplete gamma tests.
Add hooks for the dcdflib to the igamma performance tests.
Some small performance enhancements.

[SVN r56370]
 2009-09-24 11:23:52 +00:00
John Maddock
f99607da5d Fix lookup ambiguity. 
Fixes #3207.

[SVN r54284]
 2009-06-23 16:39:50 +00:00
John Maddock
6feccbee3d Added a couple of static assertions to prevent misuse. 
[SVN r53338]
 2009-05-28 08:52:05 +00:00
Jeremiah Willcock
99b1ee34f7 Fixed most tab and min/max issues from trunk inspection report 
[SVN r53141]
 2009-05-20 19:19:00 +00:00
John Maddock
b36f2d6bb6 Fix for no long double math functions. 
[SVN r52522]
 2009-04-21 08:43:36 +00:00
John Maddock
0b017bc74c Fixes #2961. 
GCC-4.0 workaround.

[SVN r52513]
 2009-04-20 15:59:58 +00:00
John Maddock
bedc5b058a Add more instrumentation code, along with some AMD64/Linux fixes. 
[SVN r52335]
 2009-04-11 15:26:59 +00:00
John Maddock
3d45185864 Sort circuit evaluation when the result will be 0. 
[SVN r52287]
 2009-04-09 09:10:45 +00:00
John Maddock
97049ddaa7 Fix use of C99 macro names. 
[SVN r51951]
 2009-03-24 10:05:46 +00:00
John Maddock
9c124ea11c Fix bug in cyl_bessel_i that hits when v=0.5 and x is small. 
Fix return type of signbit to match C99 std.
Update and regenerate docs.
Fixes #2877.

[SVN r51890]
 2009-03-21 18:44:20 +00:00
John Maddock
a935d5e1cd Updated the sign functions to use Johan Rade's fp-utilities code. 
Added tests for the sign functions.
Added docs for the sign functions and moved some sections around.

[SVN r51793]
 2009-03-16 13:25:22 +00:00
John Maddock
e38209bdc0 Add instrumentation code and some FPU control options. 
[SVN r51755]
 2009-03-13 17:56:28 +00:00
John Maddock
f6a4cbad2e Misc. small platform specific fixes and expected error rate adjustments. 
[SVN r51648]
 2009-03-08 10:39:29 +00:00
John Maddock
f10e098a4f Adjust fudge-factors yet again. 
[SVN r51578]
 2009-03-03 18:18:17 +00:00
John Maddock
2246b77395 Change tests to link to external Boost.Test library as this speeds build times by several factors when building with Intel C++. 
Change fpclassify to use our code rather than std::fpclassify with Intel C++ as it doesn't always compile the latter.

[SVN r51576]
 2009-03-03 17:16:56 +00:00
John Maddock
cf84efb70b Adjust fudge factor upwards to try and fix some of the current regression test failures. 
Add option to disable use of std::fpclassify.

[SVN r51501]
 2009-03-01 12:19:04 +00:00
John Maddock
fef466be54 remove spurious euro sign in comment. 
[SVN r51474]
 2009-02-27 17:06:21 +00:00
John Maddock
323a108149 Merges changes from Sandbox version: 
MPFR Support.
New distributions: laplace, logistic and hypergeometric.
Added PCH support to the Jamfiles for faster build times.
Added C# usage example.
Updated and regenerated docs.
Also fixes #2799 and updates the tests to catch this failure in future.

[SVN r51455]
 2009-02-26 17:34:52 +00:00
Paul A. Bristow
cc7e9904cb Added reference to Wolfram in comments. 
[SVN r51137]
 2009-02-09 14:51:25 +00:00
John Maddock
5fad227b2a Fix bug in bessel_jn for n == -1. 
Add new test case.
Checked that the other Bessel functions do not have the same issue.
Checked that real-valued -1 argument is fixed OK as well as integer argument.
Fixes #2733.

[SVN r51059]
 2009-02-06 17:26:30 +00:00
John Maddock
34049f0624 Fix unused parameter warnings. 
[SVN r50940]
 2009-02-01 10:16:20 +00:00
John Maddock
840bbe720f Improve performance of non-central distributions and the logistic quantile. 
Updated performance results.
Fixed a few bugs in performance testing code.
Added short-cut macros for new distributions in math.qbk.

[SVN r50896]
 2009-01-30 11:00:12 +00:00
John Maddock
e595c6b837 Fix min/max usage. 
[SVN r50895]
 2009-01-30 10:48:17 +00:00
Paul A. Bristow
0b8d45eb88 New copyright date. 
[SVN r50738]
 2009-01-23 09:46:27 +00:00
John Maddock
0901fbb44b Fix remaining quantile tests with the hypergeometric. 
[SVN r50723]
 2009-01-22 10:01:54 +00:00
John Maddock
160379d147 Fixes #2668. 
[SVN r50698]
 2009-01-21 09:39:50 +00:00
John Maddock
20052dcfa0 Fix some compiler issues, and mostly get the quantiles working better (round-to-nearest now works). 
[SVN r50686]
 2009-01-20 16:58:52 +00:00
John Maddock
dc392f0b3f Add some previously missing tests, and fix up the remaining concept failures for the new distributions. 
[SVN r50658]
 2009-01-17 18:14:03 +00:00
John Maddock
a635a894c7 Add easier support for mpfr and ntl. 
Add laplace to concept tests.
Fix misc. bugs and warnings, from concept tests.

[SVN r50653]
 2009-01-17 15:47:50 +00:00
Paul A. Bristow
273ae81f7a Added Doxygen doc comments for file and isNaN, Isinf, Isfinite 
[SVN r50636]
 2009-01-16 17:40:43 +00:00
Paul A. Bristow
b2d577bc6f change of variable name from location to scale (also in docs) but no changed to code (& re-tested) 
[SVN r50631]
 2009-01-16 15:41:17 +00:00
John Maddock
30d61923f6 Fix a few bugs, and more test data and update docs. 
[SVN r50608]
 2009-01-15 12:34:26 +00:00
John Maddock
2c00494ab7 Tweak config, fix signed/unsigned comparison warnings. 
[SVN r50561]
 2009-01-13 13:05:53 +00:00
John Maddock
d5b039fec8 Initial commit 
[SVN r50511]
 2009-01-08 10:21:18 +00:00
John Maddock
7a07d765ba Reconfigure STLPort. 
Fixes #2631.

[SVN r50504]
 2009-01-07 18:26:20 +00:00
John Maddock
5a2d1aae79 Split test_policy_2.cpp into multiple smaller tests. 
Fix use of long long so that code compiles with g++ -pedantic.
Update docs to match.
Fixes #1451.

[SVN r50403]
 2008-12-29 17:55:53 +00:00
John Maddock
df646686e5 Fix expm1 error handling, and add additional tests to verify the correct behaviour. 
[SVN r50344]
 2008-12-21 13:50:52 +00:00
John Maddock
94181567fa Fix some Sun-5.9 on Linux failures. 
[SVN r50341]
 2008-12-21 10:23:52 +00:00
John Maddock
8d68670436 More tests, plus fixes for issues that showed up. 
[SVN r50340]
 2008-12-21 10:14:19 +00:00
John Maddock
017a3cfe64 Added hypergeometric to more tests and to the "include everything" headers. 
Suppressed a few warnings.

[SVN r50316]
 2008-12-18 13:38:56 +00:00
John Maddock
eb786e1828 Added initial version of hypergeometric distribution. 
[SVN r50313]
 2008-12-18 12:27:41 +00:00
John Maddock
fa263945e2 Added informational comment. 
[SVN r50278]
 2008-12-15 16:03:56 +00:00
John Maddock
5f9cb93d99 Reorganise std::fpclassify usage so that functions are only declared if they're used: otherwise compilers can issue an error regarding std::fpclassify usage, even though the template is never instantiated. 
[SVN r50274]
 2008-12-15 09:31:44 +00:00
John Maddock
a69840ed47 Add a couple more exemptions from using std::fpclassify. 
[SVN r50245]
 2008-12-12 10:32:01 +00:00
John Maddock
b01c45c9c7 Patch for gcc-3.2 compiler errors. 
[SVN r50223]
 2008-12-09 16:34:59 +00:00
John Maddock
af0b3d744e Disable use of std::fpclassify for more platforms. 
[SVN r50218]
 2008-12-09 10:15:46 +00:00
John Maddock
adb945b074 Fix Intel/Linux compile failures, and reinstate some old workarounds for broken std::fpclassify implementations. 
[SVN r50161]
 2008-12-06 16:06:15 +00:00
John Maddock
bf5388177d Integrated Johan Rade's floating point utilities code. 
[SVN r50134]
 2008-12-05 16:05:26 +00:00
John Maddock
29aa65cc3f Add missing \. 
[SVN r50089]
 2008-12-03 18:12:16 +00:00
John Maddock
66e1cfa214 Adjust usage logic of log1pl and expm1l. 
[SVN r50055]
 2008-12-01 13:31:39 +00:00
John Maddock
22c59e55eb Added extra instrumentation. 
[SVN r50050]
 2008-12-01 10:34:56 +00:00
Thijs van den Berg
7718ec6ba5 First version of the Laplace distribution 
[SVN r49985]
 2008-11-28 13:36:24 +00:00
John Maddock
d13615acf3 Fixed bug in elliptic integral periodic-reduction code. 
Added some more debugging/tracing aids.

[SVN r49828]
 2008-11-18 19:47:50 +00:00
John Maddock
c15b211f2f Integrated lognormal distro into the concept check tests. 
[SVN r49486]
 2008-10-29 13:42:53 +00:00
John Maddock
74559257f6 Added logistic distribution. 
[SVN r49485]
 2008-10-29 13:10:08 +00:00
John Maddock
c3367bf1bb Added some specializations of the policy classes to reduce compile times. 
Updated history.

[SVN r49405]
 2008-10-20 16:54:24 +00:00
John Maddock
1169a0b20f Fixes for Sun C++ on Linux. 
[SVN r49141]
 2008-10-06 12:43:46 +00:00
John Maddock
df7ad868e7 Added mpfr docs, and updated the NTL usage docs. 
Updated bindings tests to run automatically when the appropriate libraries are present.

[SVN r49110]
 2008-10-02 09:37:29 +00:00
John Maddock
42f52d1993 Fix for GLIBC platforms that don't have long double support. 
[SVN r48964]
 2008-09-25 12:21:04 +00:00
John Maddock
6f07ae4be5 Add support for ellint_3 in the difficult previously unsupported region. 
[SVN r48899]
 2008-09-19 16:42:41 +00:00
John Maddock
dda431bf58 Fix missing #include. 
[SVN r48883]
 2008-09-19 09:12:33 +00:00
John Maddock
5cb4de411a Added mpfr support, and updated NTL::RR support. 
Added concept checks for those.
Added type casts where required to get mpfr_class support working OK.

[SVN r48737]
 2008-09-11 17:58:11 +00:00
John Maddock
9dd7c7ffe3 Fix for EDG-245 compilers on DECCXX and Tru64. 
[SVN r48681]
 2008-09-09 09:32:54 +00:00
John Maddock
de4fb969ad Fix definition to match fwd declaration. 
[SVN r48607]
 2008-09-05 10:39:14 +00:00
John Maddock
bb3859054f Apparently no full long double support on NetBSD. 
[SVN r48566]
 2008-09-03 17:42:45 +00:00
John Maddock
7b3bf1cc4c Fix gcc-4.x errors. 
[SVN r48510]
 2008-09-01 11:08:33 +00:00
John Maddock
7c545120df Fully qualify function calls to keep Borland happy. 
[SVN r48421]
 2008-08-28 16:06:08 +00:00
John Maddock
fb02817e67 Fix Intel-10.1 test failures, by adjusting the function signatures for the inverse hyper-geometric functions. 
[SVN r48420]
 2008-08-28 15:57:06 +00:00
John Maddock
2459d0f2a6 Merged changes from sandbox, and regenerated the docs. 
[SVN r48294]
 2008-08-22 12:51:40 +00:00
John Maddock
3f6394139c Merge changes from Trunk. 
[SVN r48293]
 2008-08-22 12:33:21 +00:00
John Maddock
394ed0afef Fixes for gcc compilation. 
Added sanity checks to tests.

[SVN r48289]
 2008-08-22 08:51:22 +00:00
John Maddock
c9e8b4d192 Updated inverse hypergeometric functions with better tests and more accurate implementation. 
[SVN r48250]
 2008-08-20 11:02:05 +00:00
John Maddock
8cbd8c5f72 Added <cmath> support to the TR1 library. 
Added Borland and Sun workaround headers to the TR1 lib.
Added a "no tr1" version of cmath to Boost.Config and made use of it throughout Boost.Math.

[SVN r48054]
 2008-08-10 09:12:02 +00:00
John Maddock
2db92b7a2a Changed some variable names to suppress -Wshadow warnings, also fixes #2167. 
[SVN r47995]
 2008-08-06 08:47:31 +00:00
John Maddock
7d56e18fed Prevent macro expansion of TR1 functions. 
[SVN r47931]
 2008-08-01 14:53:00 +00:00
Daryle Walker
952a07b6f8 Tweaked files to pass inspection report; one file had some min/max violations; but all had to have 'boostinspect:nolicense' set because a copyrighter, Paul Moore, has disappeared, so his files can never be changed to the main Boost license 
[SVN r47847]
 2008-07-27 23:43:04 +00:00
John Maddock
f451e1d7e4 Add dependency tracking info. 
[SVN r47745]
 2008-07-24 09:55:07 +00:00
John Maddock
1f8b7ae201 Fix VC 7.1 failures. 
Added minimal docs on the numeric constants.
Regenerated HTML pages.

[SVN r47541]
 2008-07-18 10:25:16 +00:00
John Maddock
2501ff283e Fix for std libs that don't have denorm support in numeric_limits. 
[SVN r47421]
 2008-07-14 16:22:03 +00:00
John Maddock
a8e2c5f910 Fix a few regressions with compiler/platform specific workarounds. 
[SVN r47420]
 2008-07-14 15:26:44 +00:00
John Maddock
980a2cd605 Fixes issue #2033. 
[SVN r47318]
 2008-07-11 17:57:21 +00:00
John Maddock
d816b0e2ff Merge changes from the Sandbox to Trunk. 
Rebuilt docs.

[SVN r47080]
 2008-07-04 17:14:37 +00:00
John Maddock
f284acaf50 Merge changes from Trunk. 
[SVN r47079]
 2008-07-04 17:08:58 +00:00
John Maddock
9cf8afd57c Remove/replace non-ASCII characters. 
[SVN r46814]
 2008-06-28 16:59:32 +00:00
John Maddock
deaa36b8d7 Fix copyright declaration. 
[SVN r46805]
 2008-06-28 11:24:02 +00:00
John Maddock
5d6396ffbe Added SSE2 optimised Lanczos approximations. 
Reordered the tgamma function to reduce the number of comparisons needed.

[SVN r46289]
 2008-06-10 08:12:17 +00:00
Bruno Lalande
285615995e Changed undeterminate to indeterminate. 
[SVN r46206]
 2008-06-06 23:57:24 +00:00
Bruno Lalande
73b62fc8b9 Completed policies and error handling test cases. 
Sanitized floating point test values in pow tests.

[SVN r46205]
 2008-06-06 22:33:39 +00:00
Bruno Lalande
7fbf7dd47f Added undeterminate_result_error, and used it for the pow<0>(0) case. 
[SVN r46182]
 2008-06-05 22:03:27 +00:00
John Maddock
4a4271a594 Added missing #include. 
Prevented macro substitution of fpclassify.

[SVN r45913]
 2008-05-29 16:03:26 +00:00
John Maddock
72ba7bce59 Added rational approximations for log1p. 
[SVN r45898]
 2008-05-29 12:38:04 +00:00
John Maddock
38812b8a96 Added missing #include. 
[SVN r45824]
 2008-05-27 17:38:49 +00:00
Bruno Lalande
7c1a1362c7 Reduced again the number of template instanciations 
[SVN r45768]
 2008-05-26 10:00:02 +00:00
John Maddock
97bdf5e56d Fixed partial specialization. 
[SVN r45585]
 2008-05-20 16:54:15 +00:00
John Maddock
211d58523c Rename some of the extern "C" C99 functions so we don't get infinite recursion. 
Disabled long double testing when not available.

[SVN r45550]
 2008-05-19 18:03:09 +00:00
John Maddock
4ce850dd8c Merged additions/changes from the Sandbox to Trunk. 
[SVN r45482]
 2008-05-18 09:18:22 +00:00
John Maddock
0aa1a09902 Merged changes from Trunk. 
[SVN r45478]
 2008-05-18 08:29:50 +00:00
John Maddock
f47f9de505 Sunpro and HPUX fix. 
[SVN r45395]
 2008-05-15 17:04:31 +00:00
John Maddock
b13d51b1fe Added float_advance + docs. 
Changed pow<N> to reduce the number of template instantiations.

[SVN r45357]
 2008-05-14 18:35:52 +00:00
John Maddock
721d6a0d62 Fixed float_next and added extra tests to detect the bug. 
Fixed expected assoc_legendre results.

[SVN r45332]
 2008-05-13 16:43:46 +00:00
John Maddock
2d19ade9fd Made float_distance return a signed value. 
Added docs for float_next etc.
Updated test_next to cope with signed results.
Added informational output to test_tr1.

[SVN r45220]
 2008-05-08 17:14:17 +00:00
John Maddock
2986ee8bf6 Renamed next_* functions to float_*. 
[SVN r45216]
 2008-05-08 16:16:51 +00:00
John Maddock
e909163e72 Fixed import/export code, and added auto-linking support. 
[SVN r45214]
 2008-05-08 16:07:22 +00:00
John Maddock
4873f71729 Added TR1 header, and tidied up ellint_3 declarations. 
[SVN r45127]
 2008-05-05 11:09:28 +00:00
John Maddock
d7e23fd2a4 Added first cut of nextafter family of functions. 
[SVN r44878]
 2008-04-29 12:01:22 +00:00
John Maddock
6012b10c87 Changed long long to boost::long_long_type and unsigned long long to boost::ulong_long_type. 
A couple of other typo corrections, to get the code compiling with g++ -pedantic.

[SVN r44877]
 2008-04-29 10:05:11 +00:00
John Maddock
ffbb796900 Apply VC-7.1 fixes: sometimes ADL fails, and we need a using declaration in order for the correct overload to be found. 
Add missing #include. to t_distribution_inv.hpp.
Suppress unnecessary instantiations in instantiate_all.cpp.

[SVN r44748]
 2008-04-24 09:40:31 +00:00
John Maddock
3ccbd30ab3 Merged patches from Trunk. 
[SVN r44695]
 2008-04-21 13:04:30 +00:00
John Maddock
bde28692c3 Moved real_concept and std_real_concept overloads of the rounding and truncation functions into the same namespace as the classes so that they can be found via ADL. 
Updated the tests to cope with the change.
Updated library configuration to cope with the change.
Made sure that the rounding and truncation functions are all correctly forward declared.

[SVN r44612]
 2008-04-20 10:58:20 +00:00
John Maddock
c3acff62f7 Reinstated real_cast to that STLPort tests can pass. 
[SVN r44452]
 2008-04-16 11:57:29 +00:00
John Maddock
fb220c20dd Add some explicit casts from the FP_* macros to type int, so that comparisons actually work when building with GCC, this is GCC bug report: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20589. 
[SVN r44434]
 2008-04-15 17:30:03 +00:00
John Maddock
8b89996bd5 Add non central distro's to fwd.hpp. 
Added needed #include to bessel_ik.hpp.

[SVN r44350]
 2008-04-12 16:27:11 +00:00
John Maddock
cd13eecbfa Added needed #includes. 
[SVN r44123]
 2008-04-09 11:35:36 +00:00
John Maddock
669bfb3991 Merged changes from sandbox to Trunk: 
New special functions for truncation and rounding, plus exponential integrals and zeta.
New non central distributions.
Updated equation png's so that they are all consistent.

[SVN r44091]
 2008-04-07 15:58:51 +00:00
John Maddock
1b7401b4c0 Merged changes from Trunk. 
[SVN r44090]
 2008-04-07 15:43:09 +00:00
John Maddock
501072165e Updated minimax docs, and program. 
Changed erf/erfc approximations to more efficient versions.

[SVN r43908]
 2008-03-28 11:27:38 +00:00
John Maddock
d02ae84866 Added new pow function from Bruno Lalande. 
Fixed a few bugs in the non-central distro's that could cause infinite looping.
Added non central distros to the performance test app.

[SVN r43800]
 2008-03-23 10:32:21 +00:00
Paul A. Bristow
2261e271d2 Added more comment and correct spelling mistake. 
[SVN r43588]
 2008-03-13 10:56:48 +00:00
Paul A. Bristow
c91e03129a Corrected correction in uniform :-( = for ==, again. 
[SVN r43587]
 2008-03-13 10:55:35 +00:00
Paul A. Bristow
5d73fa8ef1 Changed logic of error as suggested by johneddy101@comcast.net similar to uniform 
[SVN r43551]
 2008-03-10 17:53:15 +00:00
Paul A. Bristow
f25f6a6d89 Changed logic of error as suggested by johneddy101@comcast.net 
[SVN r43545]
 2008-03-08 17:47:07 +00:00
John Maddock
c4eb8bfc55 Regenerated graphs: bar graphs now have a fill! 
Added new tests for non-central distributions.
Added new tests for rounding/truncation functions and fixed the bugs exposed by them.
Deleted no-longer used graph and equation files.

[SVN r43529]
 2008-03-06 16:32:02 +00:00
John Maddock
4fb47c61e4 Apply fixes to issue #1658 which fixes some broken URL's. 
[SVN r43434]
 2008-02-29 09:49:42 +00:00
John Maddock
2396fa311c Fix mode of the weibull distribution. 
Add distribution new graphs as PNG and SVG, and change quickbook to reference them.
Update dist graph generator.

[SVN r43430]
 2008-02-28 19:12:00 +00:00
John Maddock
190c26240b Added non central distros to fwd.hpp. 
Tightened up error handing in the NC beta and T.
Added NC T docs and equations.
Updated NC T tests.

[SVN r43401]
 2008-02-23 09:52:40 +00:00
John Maddock
b652dc390c Fix bug that effects the behaviour when the result is negative. 
[SVN r43400]
 2008-02-23 09:46:43 +00:00
John Maddock
5e22dad52e Added non-central T to distributions.hpp. 
Updated generic_mode to use additive stepping where appropriate.
Improved NC-Beta PDF calculation and termination conditions.
More or less got non-central T finished off: quantiles now work, PDF and mode almost correct.
Fixed typo in toms748_solve.hpp
Updated NC-T tests to include mode and quantile.

[SVN r43333]
 2008-02-20 18:11:43 +00:00
John Maddock
3e6b00970c Added some fixes to check for iterations >= the limit rather than ==. 
Added extra common error handler.
Factored out some generic quantile code.
Killed some warnings in NC Beta.
Added initial non central T.
Fixed a few doc typos.
Added initial NC T tests.


[SVN r43324]
 2008-02-19 16:52:02 +00:00
Paul A. Bristow
6dd0f0d42c Corrected range to permit df == 2 
and Added comment(-ed code) about mode with df < 2.

No reply to question about this on Wikipedia 16 Feb 08. 

[SVN r43275]
 2008-02-16 10:36:02 +00:00
John Maddock
d77728bffb Ooops, fix comment. 
[SVN r43223]
 2008-02-11 17:30:28 +00:00
John Maddock
12df5b2419 Updated non central beta cdf to accept both x and 1-x on input: allows greater accuracy when using these routines in the non central F and T distributions. 
Updated non central F to take advantage of this.
Added some non central distributions to the graph generator.

[SVN r43222]
 2008-02-11 17:21:14 +00:00
John Maddock
8e6ce7d00c Made the non-central chi-squared PDF more robust. 
Set generic mode calculation to throw if the original guess yields a zero PDF.
Updated tests to match.

[SVN r43098]
 2008-02-04 18:57:37 +00:00
John Maddock
698f280920 Added non central F distribution. 
Tidied up non-central beta and Chi squared distribution by factoring out common generic-mode code.
Updated RR bindings.

[SVN r43075]
 2008-02-03 17:17:32 +00:00
John Maddock
38e3301e09 More or less finished off the non central beta. 
[SVN r43020]
 2008-01-30 12:42:26 +00:00
John Maddock
b45ba0d428 Added initial commit of non-central beta distribution: CDF and quantile work OK, but nothing else! 
[SVN r42999]
 2008-01-28 18:29:37 +00:00
John Maddock
04a99ac565 Optimise the sums when we're going to be subtracting the result from 1. 
[SVN r42998]
 2008-01-28 18:28:16 +00:00
John Maddock
928b8d8a63 Added error handling to the rounding functions. 
Added better error handling to the non-central chi squared, and updated the tests.

[SVN r42920]
 2008-01-23 10:41:06 +00:00
John Maddock
bd2518ee9e Add needed <iostream> include. 
[SVN r42851]
 2008-01-18 16:56:57 +00:00
John Maddock
c73de6a963 Finish off Noncentral Chi Squared docs: except for a graph. 
Update tests to cope with other platforms.
Disabled a few tests that can never pass.
Added nccs to concept checks.
Added warning fixes and special case handling to implementation.

[SVN r42849]
 2008-01-18 16:35:56 +00:00
John Maddock
8081dc49df Commit of more or less complete non-central chi-square distribution: docs still to come. 
[SVN r42820]
 2008-01-16 19:02:45 +00:00
Andrew Sutton
ccb9b8c656 Added a modification to the extreme value distribution. This allows the specification 
of a sign to determine the minimum or maximum case. Note that the original extreme value
distribution refers to the maximum case. This class will need some work. Discussion
leaned towards the use of an enumerated parameter rather than a real-typed sign
for the determination of min/max case.


[SVN r42762]
 2008-01-14 13:54:49 +00:00
Andrew Sutton
8e24be2ffe Added stubs for new distributions. These are neither fully tested nor documented and, 
in some cases, incomplete. However, migrating them into the sandbox will hopefully
help motivate their completion.


[SVN r42760]
 2008-01-14 13:36:23 +00:00
John Maddock
fe3db87cc4 Change concepts classes so they can be used when there is no long double support. 
Added first cut of the non-central chi squared distribution.
Removed almost all occurrences of real_cast: replaced with calls to the truncation/rounding functions instead.

[SVN r42744]
 2008-01-14 09:27:44 +00:00
John Maddock
2f7fadc594 Change Poisson PDF to use gamma_p_derivative as it avoids overflows and other nasties. 
[SVN r42371]
 2007-12-30 13:09:12 +00:00
John Maddock
db2ffd21c2 Added trunc, round and fmod family of functions. 
[SVN r42244]
 2007-12-22 19:01:37 +00:00
John Maddock
b27b39532b Added expint docs. 
Updated zeta and expint tests with hooks for other libraries.
Added #pragma once to header files.

[SVN r42140]
 2007-12-18 18:32:50 +00:00
Paul A. Bristow
c327e0ac0f Changed support range to avoid a discontinuity at zero. 
[SVN r42124]
 2007-12-17 15:58:23 +00:00
Paul A. Bristow
72e5db15f6 Changed support range to avoid a discontinuity at zero. 
[SVN r42123]
 2007-12-17 15:51:53 +00:00
Paul A. Bristow
99cf02892c Changed support range to avoid a discontinuity at zero. 
[SVN r42122]
 2007-12-17 15:51:08 +00:00
John Maddock
6ae334402a Fix error messages so they work with Boost.Format. 
[SVN r42074]
 2007-12-15 17:10:03 +00:00
John Maddock
fc5f875671 Fixes #1454. 
Previous fix was incomplete, so added new tests, confirmed problem report, and added fixes.

All the tests now compile cleanly with g++ -pedantic -Wall -Wno-long-long (checked with gcc-4.4.1 and 3.4.4).

[SVN r42012]
 2007-12-13 18:45:58 +00:00
John Maddock
128f2adc14 Fixes #1454. 
[SVN r41966]
 2007-12-11 16:37:22 +00:00
John Maddock
ab9017cf29 Added expint and zeta functions, plus tests. 
Fixed ntl::RR iostream operators.
Updated minimax program with zeta and expint approximations.

[SVN r41958]
 2007-12-10 18:35:56 +00:00
John Maddock
a9fa7b4924 Fix << iostream operator for the NTL bindings. 
[SVN r41272]
 2007-11-21 09:46:36 +00:00
John Maddock
e504edf582 aCC on PA-RISC emits hard errors if a numeric constant underflows: enable existing workaround for that compiler. 
[SVN r41251]
 2007-11-20 11:12:36 +00:00
John Maddock
38fc694025 Oops: added missing template specialisation argument. 
[SVN r41218]
 2007-11-19 10:02:16 +00:00
John Maddock
9a0b70f42b Added missing template argument to specialisations. 
[SVN r41199]
 2007-11-18 12:23:37 +00:00
John Maddock
605c55e00e Added workarounds for IBM xlc C++: the compiler can't resolve an overloaded function template to a function pointer unless the template arguments are explicitly provided. 
Fixed some bugs uncovered by the above workaround: some forward declarations didn't match the actual definition!
Hopefully fixed remaining Sun compiler issues: mostly fixed by above fixes anyway.

[SVN r41142]
 2007-11-16 11:30:43 +00:00
John Maddock
5c7b465759 Fix some compiler warnings when building with the Intel and aCC compilers. 
[SVN r41103]
 2007-11-15 09:52:26 +00:00
John Maddock
8c7cd40a78 Added WinCE workaround. 
[SVN r41088]
 2007-11-14 12:41:32 +00:00
John Maddock
2b1fac126f Merged changes from the Sandbox to Trunk. 
[SVN r41086]
 2007-11-14 12:14:19 +00:00
John Maddock
e7db05cc1a Merged changes from Trunk. 
[SVN r41085]
 2007-11-14 12:11:23 +00:00
John Maddock
850b746f1c Ooops, disable long double overloads when there is no long double support (Borland fix). 
[SVN r41062]
 2007-11-13 16:15:27 +00:00
John Maddock
69031ebe9d Added static assertion to catch use of long double function overloads when BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS is defined. 
Made sure that if BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS is defined, then no long double overloads are actually instantiated by the tests.

[SVN r41007]
 2007-11-11 14:05:50 +00:00
John Maddock
d37562d2ac Added needed macro suppression to use of isfinite. 
[SVN r40969]
 2007-11-09 16:57:42 +00:00
John Maddock
af3f1e97f5 Fix calls to fpclassify so they don't trigger macro expansion when fpclassify is a macro. 
[SVN r40879]
 2007-11-07 10:09:50 +00:00
John Maddock
fdac2f83cd Ooops, got the pp-logic wrong and disable long double support for all Unix platforms, instead of just Intel on Linux prior to version 10. 
[SVN r40878]
 2007-11-07 10:07:08 +00:00
John Maddock
8ff3199598 Disabled static assertions for compilers that don't support them. 
[SVN r40843]
 2007-11-06 16:58:46 +00:00
John Maddock
f9f9023daa Disabled long double support for Intel on Linux prior to version 10. 
[SVN r40780]
 2007-11-05 12:07:47 +00:00
John Maddock
098513f585 Added workaround for Sunpro not compiling the traits classes. 
Added workaround for platforms where numeric_limits<>::denorm_min() does not return a denorm!
Updated Solaris error rates.

[SVN r40727]
 2007-11-03 18:32:54 +00:00
John Maddock
25a040f5c2 Hopefully, take care of the remaining Tru64 failures: adjust >= test to > in some of the tests, and fix one expected error level for the incomplete beta function. 
[SVN r40719]
 2007-11-03 12:36:55 +00:00
John Maddock
4dad82d26a Added needed #include and using declaration. 
[SVN r40693]
 2007-11-02 18:26:47 +00:00
John Maddock
e875aee99a Changed concept code so there are no null-references any more (the code breaks on EDG based compilers otherwise). 
[SVN r40499]
 2007-10-27 11:27:19 +00:00
Paul A. Bristow
f731fd35d0 Removed obselete reference to parameter arity. Previous deletion caused lots of errors. Hope this is now OK. 
[SVN r40402]
 2007-10-24 08:56:38 +00:00
Paul A. Bristow
8cf0e97cb7 Removed obselete reference to parameter arity 
[SVN r40369]
 2007-10-23 18:54:39 +00:00
John Maddock
54ce0c0959 Fix for negative guesses. 
[SVN r40368]
 2007-10-23 18:41:17 +00:00
John Maddock
a35f87c6ea Removed redundant #define. 
[SVN r40353]
 2007-10-23 13:10:32 +00:00
Boris Gubenko
f54e683f16 fix typo in rev. 40321 
[SVN r40323]
 2007-10-23 03:44:33 +00:00
John Maddock
352d2bab85 Use __hppa to detect HP-UX on PA-RISC. 
[SVN r40269]
 2007-10-21 16:02:34 +00:00
John Maddock
722a2d848d Merged changes in Trunk. 
[SVN r40161]
 2007-10-18 18:05:13 +00:00
John Maddock
82eb0636b9 Workaround for real_concept when there are no long double math functions. 
Added HP-UX on PA RISC to list of platforms with no long double overloads.
Added macro expansion suppression code to declarations of sign and to it's usages (an AIX fix).

[SVN r40155]
 2007-10-18 15:37:01 +00:00
John Maddock
3313f991c0 Fixes for the Sun-5.9 compiler: don't try and encode constants smaller than LDBL_MIN, the compiler chokes on it. 
Fixes for STLport: long double stream operators are broken (streaming in appears not to work at all, streaming out can segfault if the value is too small).

[SVN r40083]
 2007-10-16 15:41:57 +00:00
John Maddock
4ba0027365 Fixed lots of "inspect" issues: bad hyperlinks, tabs in source and missing licenses. 
Tweaked some error levels again.

[SVN r40079]
 2007-10-16 09:32:28 +00:00
John Maddock
b6aa138a79 Fixed most of the remaining Borland issues, and removed dependency to Boost.Lambda. 
Added tentative fix for the Sunpro compilers.

[SVN r39971]
 2007-10-13 16:23:18 +00:00
John Maddock
f51e500eea Lots of Borland specific patches: quite a few of the special-function tests do now do actually pass. 
[SVN r39958]
 2007-10-12 10:01:36 +00:00
John Maddock
aebe417ed0 Workaround for msvc+stlport. 
[SVN r39900]
 2007-10-10 19:12:46 +00:00
John Maddock
d4c1bb24c7 Added Compaq CXX long long workaround to real_concept.hpp. 
Added missing forward declaration to ellint_1.hpp.
Adjusted native log1p support for aCC.
Removed bad forward declaration of fpclassify: correct declaration appears later in the file.
Adjusted compile_test/test_compile_result.hpp to not return a NULL reference.
Adjusted permitted error-limits for new platforms.
Split some of the tests into smaller units so we don't get compiler timeouts when building (hopefully!)

[SVN r39893]
 2007-10-10 16:05:26 +00:00
John Maddock
8342ec648e Tentative Sunpro workaround. 
[SVN r39884]
 2007-10-10 09:31:44 +00:00
John Maddock
991567fe80 More Borland workarounds. 
[SVN r39855]
 2007-10-09 18:12:06 +00:00
John Maddock
2983887486 Changed call signatures to keep Borland happy. 
[SVN r39842]
 2007-10-09 14:49:04 +00:00
John Maddock
f154f8bf6f Disabled long double support on some platforms. 
[SVN r39837]
 2007-10-09 12:26:27 +00:00
John Maddock
8dc2b5673f Fix problem with GCC on Linux not finding std_real_concept overload for fmod. 
[SVN r39834]
 2007-10-09 11:15:11 +00:00
John Maddock
772c8ae7ca Initial math-toolkit commit. 
Includes all code and tests, but no docs yet.

[SVN r39792]
 2007-10-08 16:58:15 +00:00
John Maddock
84dc707e2a Misc fixes for minimal Borland support. 
[SVN r39779]
 2007-10-08 09:46:32 +00:00
John Maddock
3e62ef28f0 Fix Borland infinite looping issue. 
[SVN r39723]
 2007-10-06 11:12:33 +00:00
John Maddock
3a02a8b625 Fix #include case. 
Add some needed forward declarations.

[SVN r39697]
 2007-10-05 09:56:06 +00:00
John Maddock
899f4fbfd1 Added some more descriptive information to error messages. 
[SVN r39667]
 2007-10-03 09:32:35 +00:00
John Maddock
1c5fffdd28 Added workaround for apparently broken std::fmod(long double,long double) on Darwin. 
Added more tracing macros to try and track down remaining Darwin issues.

[SVN r39650]
 2007-10-02 09:30:41 +00:00
John Maddock
01604e79d3 Added some more trace messages. 
[SVN r39637]
 2007-10-01 10:24:11 +00:00
Paul A. Bristow
5d4d94b512 Surpressed unreachable code warning. 
[SVN r39636]
 2007-10-01 10:05:21 +00:00
John Maddock
164bbbf473 Updated debug-tracing macros. 
Fixed epsilon so it works on MacOS with a 106-bit long double now.

[SVN r39593]
 2007-09-29 09:40:16 +00:00
John Maddock
700ea50732 Fixes for VC-7.1. 
[SVN r39478]
 2007-09-22 17:00:29 +00:00
John Maddock
54ff3c6535 Fixed gcc-compiler errors (added missing typename's). 
Fixed pesky warning in normal.hpp

[SVN r39476]
 2007-09-22 11:35:26 +00:00
John Maddock
3d9e81a67e Added needed using declaration to cauchy.hpp. 
Added needed #include to config.hpp.
Tightened up compile-tests.
Fixed compile tests so they build with VC++ earlier than VC8sp1.

[SVN r39471]
 2007-09-22 09:17:21 +00:00
Paul A. Bristow
41b7db31cf Warning squashing and estimate to find 
[SVN r39452]
 2007-09-21 17:39:04 +00:00
Paul A. Bristow
8daab07912 Warning squashing and estimate to find 
[SVN r39451]
 2007-09-21 17:38:38 +00:00
John Maddock
542dbbce01 Added needed header include. 
[SVN r39420]
 2007-09-20 17:29:40 +00:00
John Maddock
3e53b99430 MSVC specific fix: gets thing building with /Za. 
[SVN r39407]
 2007-09-20 12:36:32 +00:00
Paul A. Bristow
4bd52bfe7a Changechanged to use 
if((boost::math::isinf)(x))
{
  if(x < 0) return 0;
  return 1;
}
 to get rid of 4127 warnings

but still need to supress 4127


[SVN r39402]
 2007-09-20 11:37:57 +00:00
Paul A. Bristow
611653a1f8 changed to use 
if((boost::math::isinf)(x))
{
  if(x < 0) return 0;
  return 1;
}
 to get rid of 4127 warnings

[SVN r39345]
 2007-09-17 10:05:26 +00:00
Paul A. Bristow
abc0337ab9 4127 warning squashed 
[SVN r39343]
 2007-09-17 09:17:42 +00:00
Paul A. Bristow
fdef88709d changed to allow pdf and cdf infinity. 
[SVN r39310]
 2007-09-16 10:09:51 +00:00
Paul A. Bristow
0f852a9292 Try to cure test failures. 
[SVN r39299]
 2007-09-15 15:20:44 +00:00
John Maddock
76acd7ddf3 Fix for Boost.TR1 compatibility: it prevents recursive includes if we do things this way. 
[SVN r39297]
 2007-09-15 11:20:24 +00:00
John Maddock
c820ea58fc Fix for failing example: added heuristic to up the multiplication factor if we don't bracket the root quickly. 
[SVN r39296]
 2007-09-15 11:19:20 +00:00
John Maddock
6bd903d454 Updated NTL::RR support with a thin wrapper class: RR can now be used unmodified via the supplied wrapper. 
[SVN r39239]
 2007-09-13 16:56:24 +00:00
Paul A. Bristow
e334918936 added error messages, but need complement version with policy still (and name of distribution & policy?) 
[SVN r39232]
 2007-09-13 11:21:17 +00:00
Paul A. Bristow
c394aa0065 cosmetic 
[SVN r39210]
 2007-09-12 09:42:58 +00:00
Paul A. Bristow
95bc17616e added check location and check_x (random deviate) 
and changed check scale to not permit 0

[SVN r39209]
 2007-09-12 09:40:10 +00:00
Paul A. Bristow
bebff763db 1st version, but problem with sign in complement 
[SVN r39184]
 2007-09-10 17:50:32 +00:00
John Maddock
486f5a1d3e Changed: "using namespace std;" to a more specific using declaration. 
[SVN r39110]
 2007-09-03 11:51:07 +00:00
John Maddock
df3533c2d7 Added explanatory notes. 
[SVN r39077]
 2007-08-30 17:21:49 +00:00
John Maddock
d0aa223204 Added some type traits so we can do concept checking where required. 
Updated find_location to use the traits.
Added tests for traits.hpp.
Updated Jamfile.v2 with new tests.

[SVN r39076]
 2007-08-30 17:08:46 +00:00
Paul A. Bristow
7a6b00e663 tests now pass. 
[SVN r39071]
 2007-08-30 14:05:43 +00:00
John Maddock
9cdc280e69 Instrument erf better. 
[SVN r39068]
 2007-08-30 10:20:34 +00:00
John Maddock
762e728d95 Added some instrumentation to help track down some Mac OS failures. 
[SVN r39066]
 2007-08-30 10:13:13 +00:00
Paul A. Bristow
b523f0e3b7 updated copyright date 
[SVN r39050]
 2007-08-29 14:09:34 +00:00
John Maddock
51c6df2843 Disabled broken includes, will re-enable later when those headers are complete. 
[SVN r39049]
 2007-08-29 09:02:53 +00:00
John Maddock
7bbaeabe6b Fix for Darwin: not all versions have 106-bit long doubles. 
[SVN r39048]
 2007-08-29 09:00:28 +00:00
John Maddock
95559eb26a Updated isnan so that it works even when the system does a "#define isnan isnan". 
Updated Jamfiles to reflect new file names, and suppress some more warnings.

[SVN r39031]
 2007-08-28 12:34:14 +00:00
John Maddock
0a62121792 Made max-iterations permitted before we give up a policy rather than a macro or - worse - an ad-hoc value. 
At the same time cleared up a lot of gcc-specific signed/unsigned comparison warnings.

[SVN r38970]
 2007-08-26 17:39:53 +00:00
Paul A. Bristow
a7f80bca1d 1st cut - but complement does NOT work. 
[SVN r38967]
 2007-08-26 16:38:39 +00:00
Paul A. Bristow
f8578189de added copyright notice & comment & includes find_scale & location 
[SVN r38933]
 2007-08-25 12:10:03 +00:00
Paul A. Bristow
871921e322 added copyright notice & comment 
[SVN r38932]
 2007-08-25 12:09:22 +00:00
Paul A. Bristow
9ef29b5580 Copyright & cosmetic 
[SVN r38931]
 2007-08-25 12:02:41 +00:00
Paul A. Bristow
69e15fb6dc Removed warning suppression push'n'pop 
[SVN r38930]
 2007-08-25 11:59:56 +00:00
Paul A. Bristow
a76dd8ee89 Copyright & cosmetic 
[SVN r38929]
 2007-08-25 11:57:22 +00:00
Paul A. Bristow
c915c8b18c added synonyms location & scale for mean & sd 
[SVN r38925]
 2007-08-25 09:02:13 +00:00
John Maddock
bb1931edec Changed "policy" directory to "policies" to match the namespace name. 
[SVN r38780]
 2007-08-20 12:43:32 +00:00
John Maddock
9cf861c275 Hopefully fixed Intel-regression in release mode: if the data points are very close to linear the cubic interpolation may fail and generate a NaN in highly optimised code. 
[SVN r38778]
 2007-08-20 09:31:28 +00:00
John Maddock
934c402c14 Fixed some GCC-4.2 warnings in unchecked_factorial.hpp. 
Fixed some Intel-10 compile errors in error_handling.hpp.
Updated performance/main.cpp to print configuration info.

[SVN r38762]
 2007-08-19 16:37:56 +00:00
Paul A. Bristow
15d515f8b8 layout & comments only. 
[SVN r38719]
 2007-08-16 14:18:05 +00:00
Paul A. Bristow
8bdb25f7c7 More comments changed - and commented-out an unreachable code warning suppression - believed not now needed. 
[SVN r38718]
 2007-08-16 14:11:07 +00:00
John Maddock
e6a8118e0c Fixed a few more regressions, and quashed a few warnings. 
[SVN r38712]
 2007-08-16 12:22:22 +00:00
John Maddock
6d5fb5a5d6 Updated include tests, to instantiate the respective code. Fixed any errors that cropped up and suppressed rather a lot of warnings. 
Updated docs as well.

[SVN r38692]
 2007-08-15 18:31:18 +00:00
John Maddock
56852c989f Fix gcc-mw compilation. 
Added performance tuning options to performance app.

[SVN r38660]
 2007-08-14 16:29:39 +00:00
John Maddock
06f22e5d79 Reinstated tools::digits<>(): it has it's uses as a default backup used by policy::digits<>(), and makes the conceptual requirement a lot easier. 
[SVN r38649]
 2007-08-14 10:38:54 +00:00
Paul A. Bristow
09ade40af2 Correction to comment, now has a typedef for binomial_distribution. No change to code. 
[SVN r38610]
 2007-08-12 14:58:34 +00:00
John Maddock
c3ffe27f16 Changed the name of some of the discrete quantile policies, and more or less finished the policy tutorial. 
[SVN r38609]
 2007-08-12 09:24:08 +00:00
John Maddock
e2e94afe64 Minor fix to previous policy update. 
Lots of warning suppression.
Updated docs, and added the start of a policy tutorial.

[SVN r38415]
 2007-08-03 13:14:29 +00:00
Paul A. Bristow
cb82bea11e Paul's Big policy revison 
[SVN r38413]
 2007-08-03 10:13:47 +00:00
John Maddock
2c4e1860f0 Fixed gamma-function regressions. 
Tidied up function calls in t_distribution_inv.hpp.
Tidied up casts in erf.hpp

[SVN r38396]
 2007-08-02 12:06:13 +00:00
John Maddock
fa15867c49 Quashed a few warnings. 
[SVN r7592]
 2007-07-30 12:56:13 +00:00
Paul A. Bristow
1e4265c829 warn 4100 unref formal parameter supressed by making name a comment 
[SVN r7584]
 2007-07-30 11:10:16 +00:00
Paul A. Bristow
e19d19ec66 Suppressed wqarning loss of accuracy conversion to T valuetype by explicit T() 
result = T(z * 1.125 + z * 0.003379167095512573896158903121545171688L);

[SVN r7583]
 2007-07-30 09:31:20 +00:00
Douglas Gregor
d39e962ee0 Fix eol-style and mime-types 
[SVN r7564]
 2007-07-27 00:28:00 +00:00
Paul A. Bristow
f1f6a22a6a warn 4100 unref formal parameter supressed by making name a comment 
[SVN r7551]
 2007-07-26 16:14:11 +00:00
Paul A. Bristow
e543368b8d suppressed 4100 unref formal parameter 
after over-enthusiastic changing name(s) to comment.

[SVN r7550]
 2007-07-26 16:12:32 +00:00
John Maddock
e602f59025 Added typedef to binomial: there is no longer any name clash with the "binomial_coefficient" function. 
Updated distribution docs to bring them into synch with the policy based code.  Still a few "TODO" sections at present.

[SVN r7545]
 2007-07-26 12:50:29 +00:00
Paul A. Bristow
2a64206d94 added copyright and a few comments 
[SVN r7544]
 2007-07-26 11:31:17 +00:00
John Maddock
cbb289228f Mostly fixed real_concept failures on HP-UX. 
[SVN r7538]
 2007-07-25 18:05:46 +00:00
John Maddock
472614473a Merged policy code. 
[SVN r7536]
 2007-07-25 11:51:46 +00:00
Paul A. Bristow
2d15d4c712 layout only 
[SVN r7426]
 2007-07-14 11:22:18 +00:00
John Maddock
9d6b58927e Added VC7.1 fix so that the header interacts OK with Boost.Rational. 
[SVN r7335]
 2007-07-02 09:46:52 +00:00
Paul A. Bristow
1454c06994 T x = 0; // Set to a safe zero to avoid a 
// MSVC 2005 warning C4701: potentially uninitialized local variable 'x' used
   // But code inspection appears to ensure that x IS assigned whatever the code path.


[SVN r7171]
 2007-06-26 14:01:06 +00:00
Paul A. Bristow
2ca7a75bf3 Warning suppressed by comment out unused parameter name 
[SVN r7170]
 2007-06-26 13:59:16 +00:00
Paul A. Bristow
113f82d441 Control of compile options via macro - to be superseded by policy 
[SVN r7169]
 2007-06-26 13:42:30 +00:00
John Maddock
6972d2d32b Fix for failing release mode test. 
[SVN r7110]
 2007-06-20 15:23:10 +00:00
John Maddock
e74e2ffe1d Added some boost::math:: qualifiers to disambiguate between ::expm1 and boost::math::expm1. 
[SVN r7084]
 2007-06-17 15:43:56 +00:00
Douglas Gregor
f98b790432 Set MIME types and end-of-line styles globally 
[SVN r7018]
 2007-06-12 22:09:40 +00:00
John Maddock
cd9dfe5139 Performance optimisations: added Cornish-Fisher style expansions to give much better initial approximations to the inverses. Optimised Student's T quantile so that only one Halley step is taken at double precision. 
[SVN r4489]
 2007-06-07 17:42:22 +00:00
John Maddock
75a49073f6 Added needed using declaration. 
[SVN r4486]
 2007-06-07 17:37:23 +00:00
John Maddock
8b93bb6c96 Added acos, and changed asin to use Halley iteration rather than Newton Raphson. 
[SVN r4485]
 2007-06-07 17:36:31 +00:00
John Maddock
8b8ad2cd07 Oops, previous commit on this file introduced a bug, this is the fix. 
[SVN r4484]
 2007-06-07 17:34:19 +00:00
John Maddock
2f783813ba Bug Fix: If we reach a denormalised result and it's still not small enough, return zero as we're not going to find a better value. 
[SVN r4452]
 2007-06-05 09:20:23 +00:00
John Maddock
23b52e6a89 Bug fix: fancy logic only works for arguments > 1, just use straight sin/cos evaluation for small args. 
[SVN r4451]
 2007-06-05 09:15:22 +00:00
John Maddock
6e7d2e4984 Bug fix: denormalised values are just as bad for the subsequent logic as a zero. 
[SVN r4450]
 2007-06-05 09:14:03 +00:00
John Maddock
b548cab386 Remove unneeded ; which trips up gcc-4.x 
[SVN r4413]
 2007-06-01 12:41:47 +00:00
John Maddock
882f26f04f Added "inline" specifier to lots of functions that really should always have had it.... 
[SVN r4227]
 2007-05-24 17:36:46 +00:00
John Maddock
4bdf0dd8f0 Added alternative polynomial and rational function evaluation methods. 
Added new optimisation config options (still need documenting).
Tidied up use of instrumentation code so they all use BOOST_MATH_INSTRUMENT now.
Various tweaks to inverse incomplete beta and gamma to reduce number of iterations.
Changed incomplete gamma and beta to calculate derivative at the same time as the function (performance optimisation for inverses).
Fixed MinGW failures.
Refactored and extended rational / polynomial test cases.

[SVN r4172]
 2007-05-22 08:52:48 +00:00
Paul A. Bristow
6d0fab3dc7 Corrected push warning disable and added missing pop at end. 
[SVN r4049]
 2007-05-14 10:47:36 +00:00
John Maddock
655956d4a0 Fix for VC7.1 
[SVN r3996]
 2007-05-03 11:12:03 +00:00
John Maddock
cb0194297b Adjusted Mac OS error levels. 
Fixed up pareto failures.
Added some debugging code to track remaining Mac OS failures.


[SVN r3926]
 2007-04-11 11:44:43 +00:00
Paul A. Bristow
d5144d43e0 Aded comments 
[SVN r3813]
 2007-03-28 11:46:08 +00:00
Paul A. Bristow
acd36eaf15 corrected missing explicit conversion to RealType exposed by concept check. 
[SVN r3812]
 2007-03-28 11:44:55 +00:00
Paul A. Bristow
ec1691f1c6 warning suppressed by comment out unreferenced parameter 
[SVN r3806]
 2007-03-20 19:18:06 +00:00
Paul A. Bristow
7bc8929bed Message corrected 
[SVN r3805]
 2007-03-20 19:09:45 +00:00
John Maddock
c1b675c2e4 Get the number of digits right on MacOS X. 
[SVN r3798]
 2007-03-19 18:48:49 +00:00
John Maddock
f95f292d2d Fixed bracketing and comma's in initialiser lists. 
Some fixes for MacOS X's strange long double.
Updated and fixed docs.


[SVN r3795]
 2007-03-19 13:20:38 +00:00
Paul A. Bristow
c71f1ced10 cosmetic updates 
[SVN r3779]
 2007-03-16 14:55:12 +00:00
Paul A. Bristow
97167c2767 1st try 
[SVN r3778]
 2007-03-16 14:53:11 +00:00
John Maddock
5044157439 A couple of minor using declaration adjustments. 
Updated docs: adjusted titles and added more information on the kurtosis.


[SVN r3729]
 2007-02-22 16:52:38 +00:00
John Maddock
900b409d80 Fix #pragma push/pops. 
[SVN r3691]
 2007-02-15 16:45:51 +00:00
Paul A. Bristow
5c32c37b60 Warnign suppressions when used with dotnet example. 
[SVN r3685]
 2007-02-14 10:15:26 +00:00
Paul A. Bristow
c64e355e2f Spurious warning zerodivide suppressed as agreed. 
[SVN r3651]
 2007-02-07 16:17:37 +00:00
John Maddock
4bb07211d8 Final lot of updated interfaces and docs. 
[SVN r3646]
 2007-02-06 13:32:59 +00:00
John Maddock
87fa2cc1c1 Updated elliptic integral interfaces 
[SVN r3644]
 2007-02-05 10:34:44 +00:00
John Maddock
c87d0706ec Updated interface for error functions, and polynomials. 
[SVN r3642]
 2007-02-03 16:52:18 +00:00
John Maddock
6056403f72 Updated beta and factorial function interfaces. 
[SVN r3641]
 2007-02-03 13:33:20 +00:00
John Maddock
aed5ae8f6f Updated gamma functions with new argument-promotion rules and docs. 
[SVN r3637]
 2007-02-02 17:20:14 +00:00
John Maddock
8bd3649e52 Correct bessel-asymptotic usage. 
Added first cut at docs.


[SVN r3634]
 2007-01-31 16:31:05 +00:00
John Maddock
10eeb62895 Fix Linux and HP-UX regressions. 
[SVN r3633]
 2007-01-29 11:35:10 +00:00
Paul A. Bristow
8ef269fc93 added default values 
[SVN r3631]
 2007-01-22 13:54:33 +00:00
John Maddock
4ef10fbe2c Added updated SOC Bessel function code. 
Changed overlow_error to have a default second argument, updated client code to take advantage of this.


[SVN r3629]
 2007-01-19 17:05:21 +00:00
Paul A. Bristow
d52ef0920e Added default argument 1 and 1 for 'uniform beta distribution. 
[SVN r3627]
 2007-01-18 18:48:59 +00:00
John Maddock
0f32112e12 Few minopr changes to Rayleigh. 
Added missing #include to bernoulli.hpp.
Reorganised instantiate.hpp.


[SVN r3612]
 2007-01-08 13:37:40 +00:00
John Maddock
0b8c7d1fab Fix potential overflow problem. 
[SVN r3611]
 2007-01-08 12:16:50 +00:00
Paul A. Bristow
c7d542e37c 1st cut. 
[SVN r3607]
 2007-01-07 15:51:42 +00:00
Paul A. Bristow
4ccf8a528f Added comments explaining reasons for macro format. 
[SVN r3601]
 2007-01-05 12:31:40 +00:00
John Maddock
dc76889b85 Streamlined includes, removed MS-specific #pragma. 
[SVN r3599]
 2007-01-04 16:22:58 +00:00
John Maddock
27f5e4d59e Updated hypot with simpler formula. 
Added sinh and cosh to ntl.hpp.
Fixed FreeBSD test failures (no real long long support).


[SVN r3598]
 2007-01-04 16:20:49 +00:00
John Maddock
2e94601868 Conceptual updates for hypot. 
regig config setup.


[SVN r3597]
 2007-01-03 19:05:10 +00:00
Paul A. Bristow
95d7be4127 Added more constants (see also test_constants) 
[SVN r3594]
 2007-01-03 17:36:40 +00:00
Paul A. Bristow
a61d8016a4 Correctee kurtosis and made k neither 0 nor 1 an error. 
[SVN r3593]
 2007-01-03 15:30:47 +00:00
John Maddock
d3bc78cdac Fix linker errors from concept test. 
Update credits.


[SVN r3592]
 2007-01-02 11:15:16 +00:00
John Maddock
47d6f40306 Update hypot and docs. 
[SVN r3590]
 2007-01-01 16:30:52 +00:00
John Maddock
b1f54f9fe2 Updated concept checks, fixed resulting errors. 
[SVN r3589]
 2007-01-01 13:58:49 +00:00
Paul A. Bristow
27786145ea added comments(?) 
[SVN r3579]
 2006-12-30 10:59:59 +00:00
Paul A. Bristow
5867c943c1 1st commit, tests OK 
[SVN r3578]
 2006-12-30 10:58:19 +00:00
John Maddock
22dce1381e Integrated existing Boost.Math special functions, updated them to meet our conceptual requirements, and use our error handlers. 
Lots of warning suppression changes.
Added new tests to verify header includes (plus fixes where these new tests failed!).
Documentation updates.


[SVN r3575]
 2006-12-29 18:27:29 +00:00
John Maddock
bb1fbff202 Misc updates and fixes to the code, including a re-ordering of parameters to ensure TR1 compatibility. 
Updated test data.
Ellint docs added.


[SVN r3573]
 2006-12-28 11:04:39 +00:00
John Maddock
1b86b19af6 Changed gamma_P/gamma_Q to gamma_p/gamma_q. 
[SVN r3570]
 2006-12-23 17:28:30 +00:00
John Maddock
9cb3ddd158 Added concept checks for new functions. 
Fixed compiler errors.


[SVN r3560]
 2006-12-23 10:38:06 +00:00
John Maddock
eec6aaf0a0 Fixed distributions so new accessors actually compile! 
Added elliptic integrals plus tests (no docs yet).
Added some refs to Remez docs. Added comment about Erlang=gamma distribution.


[SVN r3559]
 2006-12-23 10:30:16 +00:00
Paul A. Bristow
3cd72e6865 Corrected using std::pair mistake. 
[SVN r3557]
 2006-12-22 13:04:30 +00:00
John Maddock
de9322f427 Added needed using declaration 
[SVN r3555]
 2006-12-22 10:06:37 +00:00
Paul A. Bristow
57963398de Added range and support to distributions. 
[SVN r3551]
 2006-12-21 17:39:22 +00:00
Paul A. Bristow
4302268541 Commented out the undefined median to allow derived accessor to rpovide quantile(half). 
[SVN r3549]
 2006-12-21 13:11:22 +00:00
Paul A. Bristow
8dcd29185e Layout & comments. 
[SVN r3547]
 2006-12-20 18:59:08 +00:00
Paul A. Bristow
ca818226af ? changed 
[SVN r3546]
 2006-12-20 18:58:04 +00:00
Paul A. Bristow
222836bded added ln2 etc 
[SVN r3545]
 2006-12-20 18:54:26 +00:00
Paul A. Bristow
570d5bdf5f added median as quantile(half) 
[SVN r3539]
 2006-12-20 15:31:13 +00:00
Paul A. Bristow
7b6fe3921a added median to all distributions. 
[SVN r3538]
 2006-12-20 15:29:43 +00:00
John Maddock
10ec229b15 Added missing inline specifiers. 
[SVN r3533]
 2006-12-18 11:04:37 +00:00
Paul A. Bristow
3362331ddc Corrections to kurtosis. 
[SVN r3526]
 2006-12-13 19:13:11 +00:00
Paul A. Bristow
a7f162346d Corrections to kurtosis. 
[SVN r3525]
 2006-12-13 19:03:22 +00:00
John Maddock
d7653488e8 Added needed parenthesis. 
[SVN r3524]
 2006-12-13 17:48:40 +00:00
John Maddock
c6f65d5ad3 Big initial commit of the Polynomial and associated functions from TR1. 
[SVN r3522]
 2006-12-12 17:25:29 +00:00
Paul A. Bristow
7d4795a44d Layout improvements 
[SVN r3520]
 2006-12-12 10:28:57 +00:00
Paul A. Bristow
a0baf23d99 infinite x removed. 
[SVN r3518]
 2006-12-12 10:05:12 +00:00
Paul A. Bristow
c47170b91b infinity removed. 
[SVN r3516]
 2006-12-12 09:46:24 +00:00
Paul A. Bristow
b274b9b464 changed to disallow lower == upper and more tests. infinite a, b, and x still allowed. 
[SVN r3514]
 2006-12-11 14:35:04 +00:00
Paul A. Bristow
cf6f16d4be 1st cut but questions on infinity? 
[SVN r3510]
 2006-12-10 18:44:29 +00:00
Paul A. Bristow
9e9da20092 Changed to more accurate kurtosis realType(-6)/5 but still problem with infinity. 
[SVN r3504]
 2006-12-06 18:06:26 +00:00
Paul A. Bristow
8f8065eb43 warning 4130) // '==' : logical operation on address of string constant. suppressed. 
[SVN r3503]
 2006-12-06 14:13:19 +00:00
Paul A. Bristow
bc9d1cfec0 Changed to more accurate kurtosis realType(-6)/5 but still problem with infinity. 
[SVN r3501]
 2006-12-06 09:47:27 +00:00
Paul A. Bristow
f878c291d9 1st cut 
[SVN r3495]
 2006-12-04 17:54:35 +00:00
John Maddock
306748dcee Small fixes to beta and beta distribution. 
Brought docs into line.


[SVN r3487]
 2006-12-02 19:21:13 +00:00
John Maddock
4358aba0ee Finished off binomial and neg binomial docs (hopefully). 
Renamed param estimators for (negative) binomial sample sizes.
Tweaked root finding algorithms.
Tweaked ibeta inverse.


[SVN r3483]
 2006-11-30 18:26:30 +00:00
John Maddock
ccfb864daf Added Jeffreys prior method to the binomial confidence limits. 
Fixed a few typos in the neg binomial.


[SVN r3482]
 2006-11-30 10:47:26 +00:00
John Maddock
1002d7d090 Minor coments/changes to code. 
Proof reading changes to docs, mostly rewritten neg binomial worked example.
Removed dead test.


[SVN r3473]
 2006-11-28 14:12:26 +00:00
John Maddock
475d4a0280 Misc small fixes and warning suppressions. 
[SVN r3472]
 2006-11-28 10:54:09 +00:00
Paul A. Bristow
69ecd8d2a0 Boundary values for quantile corrected and tests added but still don't deal with infinity well. TODO still. 
[SVN r3469]
 2006-11-27 19:02:37 +00:00
Paul A. Bristow
2fdd390169 Promotion no longer relevant so removed.. 
[SVN r3468]
 2006-11-27 15:40:28 +00:00
John Maddock
ed1b8ae751 Finally got the parameter estimators right, and checked against "known goods". 
Misc small fixes to the docs and example programs.


[SVN r3462]
 2006-11-26 09:03:49 +00:00
John Maddock
7fc815029e Fixed up neg binomial estimators. 
Enabled neg binomial tests and adjusted arguments.
Added extra comment to binomial.


[SVN r3460]
 2006-11-24 19:14:32 +00:00
John Maddock
d15bf712e5 Fixed binomial estimators and added stronger tests. 
Typos in concepts.
Enabled native log1pl on more platforms.
Fixed binomial example.
Added image.


[SVN r3459]
 2006-11-24 18:54:52 +00:00
Paul A. Bristow
867dd1d8e6 added code for special case prob < cdf(0) for quantile and complement. 
[SVN r3458]
 2006-11-24 17:53:10 +00:00
Paul A. Bristow
4a6cd24f9c suppressed error 4702 unreachable code. 
[SVN r3451]
 2006-11-24 12:23:32 +00:00
Paul A. Bristow
48b5f13838 if (p <= pow(1 - dist.success_fraction(), dist.trials())) 
{ // p <= pdf(dist, 0) == cdf(dist, 0)
   return 0; // So the only reasonable result is zero.
} // And root finder would fail otherwise.
Added, and similarly for complement


[SVN r3445]
 2006-11-23 17:46:15 +00:00
John Maddock
78244650dc More fixes to precision.hpp 
More or less finished concept docs.
Lots of proof-reading and editorial fixes.


[SVN r3440]
 2006-11-22 18:48:28 +00:00
Paul A. Bristow
24af492a2d Correction to estimate parameters. 
[SVN r3437]
 2006-11-22 18:28:23 +00:00
Paul A. Bristow
1192252ba9 Correced typos and added <16834> size for 80 and 128 bit 
[SVN r3436]
 2006-11-22 16:48:52 +00:00
John Maddock
12984831f0 More concept docs, updated code to match. 
[SVN r3423]
 2006-11-21 13:19:08 +00:00
John Maddock
b89f35783e More editorial and proof reading fixes. 
Added initial draft of Remez background information.


[SVN r3414]
 2006-11-19 18:29:06 +00:00
Paul A. Bristow
a6b95e985f Some corrections but still not right. 
[SVN r3409]
 2006-11-16 12:46:45 +00:00
Paul A. Bristow
7cd4885c79 Nano optimisation of skewness. 
[SVN r3398]
 2006-11-15 12:43:32 +00:00
Paul A. Bristow
3242af54a7 Minor revisions 
[SVN r3389]
 2006-11-14 15:40:12 +00:00
Paul A. Bristow
8452c720ba conditional expression is constant warning 4127 suppressed. (produced 4 pages of messages!) 
[SVN r3384]
 2006-11-14 11:48:19 +00:00
Daryle Walker
37e8422e8a Made the GCD specializations for '(unsigned) long long' and '(unsigned) __int64' mutually exclusive, following the lead of <boost/cstdint.hpp>. 
[SVN r35999]
 2006-11-10 23:15:49 +00:00
Paul A. Bristow
f84fef66c8 More tests including test_spot and handful of spot values. 
[SVN r3362]
 2006-11-09 17:33:17 +00:00
John Maddock
3ea5308b45 Fixed problem with chi-squared sample size estimator. 
Updated test program accordingly.
Added chi-square tutorial code.


[SVN r3360]
 2006-11-08 17:14:39 +00:00
Paul A. Bristow
32be9aa3f3 corrected quantile to use ibeta_inv. test_spot still todo. & parameter estimate?? 
[SVN r3358]
 2006-11-08 15:27:01 +00:00
Paul A. Bristow
a4c6015e0b 1st cut with a few simple tests for double only. 
[SVN r3354]
 2006-11-07 12:21:22 +00:00
Paul A. Bristow
24d36b78f6 Spelling correction, and comments copied to concepts.qbk. 
[SVN r3352]
 2006-11-07 09:35:58 +00:00
Daryle Walker
b7001da56b Changed GCD for built-in integer types to use the binary-GCD algorithm 
[SVN r35850]
 2006-11-05 07:10:42 +00:00
John Maddock
9e41ae39b4 Minor fixes for gcc in poisson.hpp 
Tidied up fraction.hpp.
Updated linux acceptable error rates.


[SVN r3346]
 2006-11-04 18:28:24 +00:00
Paul A. Bristow
d9d5f2b21e tests revised to try to cover failures in 80 and 128 bit FP cases. Added throw checks 
[SVN r3345]
 2006-11-04 10:55:12 +00:00
John Maddock
feacc055d3 Added gamma distribution and tests. 
Fixed a lot of gcc-linux issues: mostly missing "using namespace std" declarations.
Added new concept check to test for these kinds of failures at compile time.


[SVN r3344]
 2006-11-03 18:42:23 +00:00
Paul A. Bristow
041d476768 Removed median. Allowed hazard & chf to be provided by derived accessors, and removed any mention of std::numeric_limits<T>::quiet_NaN() 
[SVN r3343]
 2006-11-03 15:04:38 +00:00
John Maddock
8441e358fb Rewritten tgamma1pm1 in terms of lgamma rational approximations: now more efficient and more accurate! 
Minor patch to exponential distribution tests.
Removed dead code from Remez app.


[SVN r3331]
 2006-10-31 10:50:11 +00:00
John Maddock
7d003a280f Fix typo and regressions. 
[SVN r3330]
 2006-10-31 10:48:26 +00:00
John Maddock
2132f92495 Changed tgamma to use factorial lookup where applicable. 
Updated docs accordingly and added new docs for the new factorial and binomial functions.


[SVN r3327]
 2006-10-28 16:30:39 +00:00
John Maddock
bcbb6f3ca9 Added double_factorial plus rising and falling factorials and binomial coefficients. 
Added new tests for the above.
Refactored factorials header.
Optimised tgamma_delta_ratio for integer arguments.


[SVN r3326]
 2006-10-28 13:00:10 +00:00
Paul A. Bristow
da6a644de8 Minor comments edits, for more tests OK with builint float double but fail neg binomial for real concept. 
[SVN r3324]
 2006-10-26 13:12:20 +00:00
John Maddock
82548699a5 Improved and generally tidied up sqrt1pm1 and powm1. 
Renamed sqrtp1m1 as sqrt1pm1 to match log1p etc.
Added docs for misc power functions, plus weibull distribution.
Updated tests to match the above changes.


[SVN r3321]
 2006-10-26 11:51:59 +00:00
John Maddock
527bf0d227 Updated erf implementation selection logic. 
Provided rational expansions for expm1.
Provided faster form for log1p.
Updated remez minimax code.


[SVN r3319]
 2006-10-25 18:10:13 +00:00
John Maddock
9b16307ae7 Fixed a bug in incomplete gamma, and refactored and refined internal selection logic. 
Added tests for incomplete gamma for int and half int arguments.


[SVN r3312]
 2006-10-25 10:19:34 +00:00
John Maddock
24066f892f Updated to call overflow_error when the result is infinite. 
[SVN r3311]
 2006-10-25 09:38:24 +00:00
John Maddock
2f964f6ef6 Added finite sum to incomplete beta, plus tests and docs. 
Linked in lognormal docs.


[SVN r3310]
 2006-10-25 09:03:04 +00:00
Paul A. Bristow
76f2122bad Comments. 
[SVN r3309]
 2006-10-24 18:05:48 +00:00
Paul A. Bristow
84bdaaa6ee Comments and checks that no return of infinity that should call overflow_error. 
[SVN r3308]
 2006-10-24 17:58:30 +00:00
Paul A. Bristow
25570182cb Tests now pass, but more tests to do still. 
[SVN r3307]
 2006-10-24 17:47:40 +00:00
Paul A. Bristow
504e9da847 Layout & comments only. 
[SVN r3303]
 2006-10-24 09:36:36 +00:00
John Maddock
7a0ac09ba8 Misc fixes and better error handling and special case treatment. 
[SVN r3300]
 2006-10-23 12:55:19 +00:00
Paul A. Bristow
0a4dfe1bc5 complement and derived accessors still TODO 
[SVN r3299]
 2006-10-23 12:01:03 +00:00
Paul A. Bristow
0af1942639 Added complement.hpp and expm1 to get it compile, but fails tests as yet. 
[SVN r3297]
 2006-10-23 11:39:43 +00:00
Paul A. Bristow
d29c25757c Added some checks for scale and prob. But fails tests. 
[SVN r3292]
 2006-10-23 10:09:57 +00:00
Paul A. Bristow
c7c1e86552 Several "return static_cast<V>( ... )" added to avoid warning C4244 conversion from 'const double' to 'float'. 
Not all tested by test_chi_squared usage :-(  so a mistake might cause compile failures elsewhere?


[SVN r3283]
 2006-10-22 16:05:07 +00:00
Paul A. Bristow
ecd6a66fee Added default RealType double (except binomial - where commented out). 
(Changes to test in binomial reflect this, others in progress).


[SVN r3278]
 2006-10-21 18:34:13 +00:00
Paul A. Bristow
860e361ef7 default Real type double 
[SVN r3277]
 2006-10-21 18:07:46 +00:00
John Maddock
b6529add8a Added preliminary lognormal and weibull implementations. 
Updated common error handling.
Added more test values.


[SVN r3265]
 2006-10-18 15:18:15 +00:00
Paul A. Bristow
c4de333200 JM's best so far optimisation. 
[SVN r3263]
 2006-10-17 09:08:47 +00:00
Paul A. Bristow
6bdc0ec840 Notes & comments only. 
[SVN r3261]
 2006-10-17 09:06:18 +00:00
John Maddock
09f5ff738f Brought distributions into the same conceptual line and updated tests accordingly. 
[SVN r3256]
 2006-10-15 17:57:02 +00:00
Paul A. Bristow
a9cfe8beec corrections and improvements (hopefully) to pdf. 
Quantile and complement handle mean = 0 case ith domain error.


[SVN r3254]
 2006-10-15 15:35:11 +00:00
John Maddock
1555938251 Added scale and location parameters to Cauchy. 
[SVN r3253]
 2006-10-14 18:03:03 +00:00
John Maddock
5f5d873826 Added concept checks and concept archetypes. 
Uprated tests and distributions to match concept requirements.


[SVN r3251]
 2006-10-13 17:37:27 +00:00
John Maddock
4edc3d19a1 Added cauchy, exponential, and extreme value distributions. 
Updated style sheet.


[SVN r3250]
 2006-10-12 17:28:09 +00:00
Paul A. Bristow
0010c8c1c6 misc trivia, spelling refs, punctation. 
[SVN r3248]
 2006-10-10 15:34:15 +00:00
Paul A. Bristow
5eb7b02d56 1st version of Poisson as distribution. parameter estimation missing? 
[SVN r3246]
 2006-10-10 15:25:18 +00:00
John Maddock
d988a98c59 Added new asymptotic expansion for incomplete gamma functions P(a,x) and Q(a,x), used for large a, and a ~ x. 
[SVN r3239]
 2006-10-06 14:52:53 +00:00
John Maddock
5b95219e09 Fixed digamma 128-bit error rates with a more accurate value for the root. 
Updated png's: hopefully they'll all render OK now.
Tweeked docs to reflect recent advances.


[SVN r3237]
 2006-10-04 09:57:05 +00:00
John Maddock
dc9165592e Updated digamma docs and code. 
Added needed include to test_data.hpp.
Changed Doc Jamfile to generate TOC with 0.2 version FOP.


[SVN r3232]
 2006-10-02 16:58:49 +00:00
John Maddock
5997e3249c Added digamma. 
Added needed #include to evaluation_type.
Updated minimax and lanczos generators to take account of digamma.


[SVN r3231]
 2006-10-01 09:45:44 +00:00
Paul A. Bristow
0ef909cbb5 Comments and added real_cast s to int unsigned int and long. 
[SVN r3227]
 2006-09-30 18:02:01 +00:00
Paul A. Bristow
0672bcb1a1 Comments & cosmetic. 
[SVN r3226]
 2006-09-29 16:22:28 +00:00
Paul A. Bristow
7b9e7a922d warning suppression for MSVC using push 'n' pop. 
[SVN r3225]
 2006-09-29 14:39:34 +00:00
Paul A. Bristow
19d0f37013 comment on non-integer k and discrete distributions. 
[SVN r3224]
 2006-09-29 14:13:28 +00:00
Paul A. Bristow
dbacc2276e trivia. 
[SVN r3223]
 2006-09-29 14:12:10 +00:00
John Maddock
0275810ec6 Updated fpclassify code to optimise isnan. 
Removed reference to interval lib.


[SVN r3220]
 2006-09-27 11:41:06 +00:00
John Maddock
11c1946e1e Fix minor bug in test_data.hpp. 
Updated docs with minima and minimax docs.
Also reorganised to separate out the more experimental components.


[SVN r3218]
 2006-09-26 18:24:13 +00:00
John Maddock
1335a727f6 Bug fix. 
[SVN r3216]
 2006-09-24 09:23:33 +00:00
John Maddock
b9e4fd1418 Updated old code to use compile-time polynomial size detection. 
Fixed template deductiojn bug in evaluate_rational.
Updated lanczos generation to match.


[SVN r3215]
 2006-09-24 09:23:09 +00:00
John Maddock
4679d4e421 Updated rational function evaluation with compile time detection of array size. 
Added tests.
Updated Jamfile.


[SVN r3214]
 2006-09-22 18:03:57 +00:00
Paul A. Bristow
153e797b74 Commented out variable name to avoid C4100 unref parameter, and comment on its use to select Lanczos. 
[SVN r3213]
 2006-09-22 11:33:23 +00:00
John Maddock
5fff71d70c Fix to get new code compiling with Borland. 
[SVN r3212]
 2006-09-22 09:43:52 +00:00
John Maddock
306c96a5e1 Adjusted limits. 
[SVN r3211]
 2006-09-22 09:29:19 +00:00
John Maddock
5b9e2ec5a8 Added new (better) forms for evaluating polynomials. 
Added more Remez code, and appoximations for erf/erfc inverses.
Updated and refactored erf code to use new approximations.
Added more test cases.


[SVN r3210]
 2006-09-22 09:28:40 +00:00
John Maddock
379e6a5e07 Oops, fix bug. 
[SVN r3203]
 2006-09-16 18:29:39 +00:00
John Maddock
47690421b6 Added remez and minima tests. 
Added some mimimal fixes to get the code to compile with Borland-5.8.2, although frankly it's more trouble than it's worth.


[SVN r3200]
 2006-09-16 16:41:18 +00:00
John Maddock
6ee3cba62b Added Remez minimax code. 
[SVN r3197]
 2006-09-15 16:18:07 +00:00
John Maddock
6a9623c9b2 Added missing inline specifiers: we get duplicate symbol errors without them! 
[SVN r3195]
 2006-09-15 16:10:21 +00:00
John Maddock
e9b217437e Added rational approximations for lgamma and small arguments: greatly reduced errors near the roots of lgamma. 
[SVN r3194]
 2006-09-15 16:09:12 +00:00
John Maddock
6c3c521a3c More linux fixes: finally got AMD64 machines to behave correctly. 
[SVN r3186]
 2006-09-02 17:56:22 +00:00
John Maddock
84829f74e3 Numerous minor patches to get tests passing on Linux x86 (32-bit) and HP-UX / itanium. 
[SVN r3183]
 2006-09-02 15:05:43 +00:00
John Maddock
567dc1bd7d Updated error checks. 
Removed dead code from error-handling test.


[SVN r3176]
 2006-09-01 09:19:24 +00:00
Paul A. Bristow
aa5fca355c added newline at end of last line #endif // BOOST_MATH_DISTRIBUTIONS_CHI_SQUARED_HPP 
to avoid fatal compile error.

Noted failure to throw 12 expected exceptions - will investigate why.


[SVN r3175]
 2006-08-31 17:41:19 +00:00
Paul A. Bristow
5577b49a33 Spelling and a few comments & refs. 
[SVN r3174]
 2006-08-31 16:56:28 +00:00
John Maddock
dc70a06809 Added F distribution. 
[SVN r3173]
 2006-08-31 16:21:17 +00:00
Paul A. Bristow
3f5d624b05 Comment added about big df and df = +infinity needing normal distribution. 
[SVN r3171]
 2006-08-31 14:32:05 +00:00
John Maddock
a151830ed4 Fixed corner cases. 
Implemented parameter estimator.


[SVN r3167]
 2006-08-30 15:48:21 +00:00
John Maddock
00a347bee0 Removed remaining old style distribution code (we can get it back from cvs if necessary). 
Updated error handling.
Added better chi-squared tests.


[SVN r3165]
 2006-08-29 16:41:07 +00:00
John Maddock
9a3ade1320 Refactored error handling a little more. 
Added chi-squared distribution and tests.


[SVN r3164]
 2006-08-28 12:52:08 +00:00
John Maddock
fa86c0c490 Updated docs. 
Rewritten students-t parameter estimator.
Added binomial examples.


[SVN r3162]
 2006-08-26 18:33:46 +00:00
John Maddock
9d7123eecf More or less completed normal/binomial/students-t distributions. 
Updated docs and tests.


[SVN r3160]
 2006-08-25 18:03:48 +00:00
John Maddock
99929b1023 Fixed binomial quantile, added complements, various subtle changes to corner cases, tidied up and removed dead code. 
Fixed error messages in beta and gamma inverses.
Updated binomial test program, refacted cdf test cases to test cdf, quantile and complements in one hit.
Added accuracy tests for ibeta_inva and ibeta_invb.
Added more docs.


[SVN r3159]
 2006-08-20 09:28:27 +00:00
John Maddock
0d03447123 Added overflow check. 
[SVN r3157]
 2006-08-18 09:39:30 +00:00
Paul A. Bristow
a23806eb61 1st draft - pdf and cdf probably OK, but quantile etc are not. 
[SVN r3155]
 2006-08-17 17:40:55 +00:00
John Maddock
d40fa2fb99 Changed students-t quantile to be free from cancellation error. 
Added ibeta_inv overloads that return both x and y.
Updated docs to match.


[SVN r3154]
 2006-08-16 17:33:24 +00:00
John Maddock
ef43e16e47 Adjusted termination conditions to cope with very large first derivatives in some regions. 
[SVN r3152]
 2006-08-16 12:05:08 +00:00
John Maddock
f965af9da3 Updated incomplete beta code to calculate power terms in a more reliable way. Added explanatory comments describing what's going on. 
[SVN r3151]
 2006-08-16 11:40:38 +00:00
Paul A. Bristow
d66b96d4d6 Added #include <boost/cstdint.hpp> for uintmax_t 
[SVN r3150]
 2006-08-16 11:40:11 +00:00
John Maddock
2befb07120 Adjusted limits to avoid asserts. 
[SVN r3149]
 2006-08-16 11:39:11 +00:00
John Maddock
a7c3186b5a Added fix for cdf when degrees of freedom parameter is large. 
[SVN r3148]
 2006-08-16 11:38:07 +00:00
John Maddock
4ff84d85ab Added sample programs that generate test data for incomplete beta and gamma inverses. 
Added asin to ntl.hpp: needed for incomplete beta inverse.
Fixed limits on table size in the incomplete beta.


[SVN r3146]
 2006-08-14 17:29:59 +00:00
John Maddock
7e902a680e Added root-2 constant. 
Fixed complement code so that constructors are used now.
Adjusted pre-formatted parts of tutorial.
Added normal distribution plus tests.


[SVN r3145]
 2006-08-14 12:33:49 +00:00
Paul A. Bristow
eacc0ceb7c quieted warnings on unreachable code.. 
[SVN r3142]
 2006-08-14 11:36:38 +00:00
Paul A. Bristow
c6e4a1aabc Quieted warning unreachable code. 
[SVN r3141]
 2006-08-14 11:35:20 +00:00
Paul A. Bristow
ee24bfd3ee comments and quieted of warnings unreachable code. 
[SVN r3140]
 2006-08-14 11:34:15 +00:00
John Maddock
4e98273785 Added upper cap on the number of iterations permitted in series evaluation. 
Fixed beta function so it's better behaved with very large arguments.
Updated tests.


[SVN r3139]
 2006-08-13 17:42:17 +00:00
Paul A. Bristow
8cc9b6d50d commented out distributions now obselete. Need to redo all these with the distribution classes. 
[SVN r3138]
 2006-08-13 11:39:58 +00:00
John Maddock
3ce59eaed5 Fixed gcc errors (duplicate definitions of default parameters). 
[SVN r3135]
 2006-08-12 18:00:54 +00:00
Paul A. Bristow
fec4b853a1 *** empty log message *** 
[SVN r3131]
 2006-08-12 13:28:12 +00:00
Paul A. Bristow
50ad897d31 Removed - to be replaced by version in \distributions 
[SVN r3130]
 2006-08-12 11:13:59 +00:00
Paul A. Bristow
d34cdb4311 Replaced by version in \distributions. 
[SVN r3129]
 2006-08-12 11:11:08 +00:00
Paul A. Bristow
9895ce48e1 Use distributions instead of dist. 
[SVN r3125]
 2006-08-11 10:36:18 +00:00
Paul A. Bristow
59a3e0b48f Moved to distributions folder. 
[SVN r3124]
 2006-08-11 10:33:53 +00:00
John Maddock
5c645698c3 Added new style Student's t code and examples. 
Added tutorial for the new stats code.


[SVN r3121]
 2006-08-10 12:53:07 +00:00
John Maddock
2dcc9167e8 Updated inverse gamma tests. 
Updated inverse gamma code to prevent premature termination of Halley Iteration.


[SVN r3120]
 2006-08-10 09:27:19 +00:00
John Maddock
46d9b33645 Tightened up root finding tests. 
Added docs for roots without derivatives.
Added derivatives of incomplete gamma and beta.


[SVN r3112]
 2006-08-01 17:23:10 +00:00
John Maddock
cd90a9f0d4 Updated and documented the new root finding code. 
[SVN r3111]
 2006-07-31 18:05:41 +00:00
John Maddock
5e67622a7d Fixed gamma_inva. 
Added floor and ceil termination conditions to toms748_solve.
Added ibeta_inva and ibeta_invb.


[SVN r3109]
 2006-07-27 15:45:17 +00:00
John Maddock
79a8199f9c Added new root finding algorithm toms748. 
Tidied up inverse code and moved files to detail directory.
Updated docs and Jamfile.


[SVN r3107]
 2006-07-26 13:06:56 +00:00
John Maddock
4ecb968fc7 Begun to refactor and tighten up the tests so that results are actually regression tested as well as printed to screen. 
Added code for extracting the sign of a number.
Fixed a few bugs and tidied up the code a little in places.
Changed test_result to properly encapsulate it's contents.


[SVN r3106]
 2006-07-25 09:17:33 +00:00
Gennaro Prota
2875cd33b7 same as for acos.hpp: fixed license reference text; slight changes to comments to avoid spurious inspect tool warnings 
[SVN r34711]
 2006-07-24 21:37:11 +00:00
Gennaro Prota
f31a98ad95 fixed license reference text; slight changes to comments to avoid spurious inspect tool warnings 
[SVN r34709]
 2006-07-24 21:29:43 +00:00
John Maddock
94ebae476a Updated generic erf version so it doesn't call the incomplete gamma directly. 
Refactored the incomplete gamma, improved the changeover between P and Q, added two new finite-sum expansions for integer and half integer arguments.
Improved error handing in the inverse gamma.


[SVN r3099]
 2006-07-20 12:58:29 +00:00
John Maddock
74b41709bf Borland C++ fix. 
[SVN r34532]
 2006-07-13 09:47:18 +00:00
John Maddock
607e6990cd Added needed include. 
[SVN r34514]
 2006-07-12 17:52:28 +00:00
Paul A. Bristow
162f416732 Looks nearer to correct now with test and example1. 
[SVN r3091]
 2006-07-12 16:13:11 +00:00
Paul A. Bristow
e06f1fa83b Corrected missing k+1 
[SVN r3089]
 2006-07-11 16:09:04 +00:00
Paul A. Bristow
305ca1f705 comments etc PAB 
[SVN r3088]
 2006-07-11 14:17:49 +00:00
Paul A. Bristow
8417d24efd comment lexical cast fqult added 
[SVN r3087]
 2006-07-11 14:16:50 +00:00
Paul A. Bristow
58d4299525 Another unused parameter conveted to /* l */ 
[SVN r3086]
 2006-07-11 14:15:42 +00:00
John Maddock
45fce41aa2 Warning suppression. 
[SVN r3085]
 2006-07-11 12:30:40 +00:00
John Maddock
cda4c449a9 Crush a couple more warnings. 
[SVN r3084]
 2006-07-11 12:27:48 +00:00
Paul A. Bristow
11bb475338 More comments, still not tested much. 
[SVN r3082]
 2006-07-10 17:41:53 +00:00
John Maddock
86c65d8c9f Add using declaration so that std::fabs is actually found. 
[SVN r3081]
 2006-07-10 16:25:06 +00:00
John Maddock
31fbe1eaeb Added needed math_fwd.hpp includes were needed, cleaned up code, and generally got things compiling with gcc. 
[SVN r3080]
 2006-07-10 15:30:36 +00:00
Paul A. Bristow
6307ffc6d3 tested by PAB at 8 July - links and runs OK 
[SVN r3078]
 2006-07-09 18:33:33 +00:00
John Maddock
bc87a3747d Added #pragma's where warnings couldn't be suppressed any other way. 
Refactored code to suppress warnings where possible.


[SVN r3077]
 2006-07-09 17:56:55 +00:00
John Maddock
f7961cdc6d Added missing inline specifiers. 
[SVN r3076]
 2006-07-09 17:54:06 +00:00
Paul A. Bristow
ea8e3cf385 1st commit - some tests OK? 
[SVN r3072]
 2006-07-06 15:19:30 +00:00
Paul A. Bristow
a5603dc093 not sure why changed - comments? 
[SVN r3071]
 2006-07-06 15:18:19 +00:00
Paul A. Bristow
c9afe467ac Seems sane? 
[SVN r3069]
 2006-07-06 10:00:35 +00:00
Paul A. Bristow
a78a218dd6 REMOVED two argument domain_error, etc functions. Several calls changed but May still be 2 arg calls in code that will fail to compile now. 
[SVN r3059]
 2006-07-04 17:43:22 +00:00
Paul A. Bristow
866fb8fb17 argument promotion (needed by math function forward declarations math_fwd.hpp) 
See also test_promotions.cpp


[SVN r3057]
 2006-07-04 17:39:30 +00:00
Paul A. Bristow
da4f485ceb Update to match math_fwd.hpp OK with test_function_declarations.cpp 
[SVN r3055]
 2006-07-04 17:17:40 +00:00
Paul A. Bristow
1d0576a788 cosmetic only. TODO add #include math_fwd.hpp 
[SVN r3050]
 2006-07-03 15:50:58 +00:00
John Maddock
9b6fd61a8f Fixed VC warnings. 
Added needed includes.


[SVN r3036]
 2006-06-27 16:34:16 +00:00
Paul A. Bristow
f6b8971322 updated to use JM promoted arguments 
[SVN r3034]
 2006-06-27 09:45:59 +00:00
Paul A. Bristow
a2d122db66 Updated with JM promoted arguments 
[SVN r3033]
 2006-06-27 09:45:59 +00:00
John Maddock
6b0825841d Updated code to use the new 3-arg domain_error and pole_error functions. 
[SVN r3030]
 2006-06-24 17:41:23 +00:00
John Maddock
b2c1ef342f Added abs to real_concept. 
Added logic_error to error handling code.


[SVN r3029]
 2006-06-24 15:50:52 +00:00
John Maddock
6910a75524 Reorganised and refactored limits and constants code. 
[SVN r3028]
 2006-06-24 12:00:53 +00:00
Paul A. Bristow
818cfd0bbf Improved error messages and try at warning and constraining probability outside 0 to 1 
[SVN r3027]
 2006-06-23 18:30:40 +00:00
John Maddock
fa6e01c6ee Fix typos. 
[SVN r3024]
 2006-06-23 09:41:26 +00:00
John Maddock
3519ff8b03 Update digits10 formula used. 
[SVN r3023]
 2006-06-23 09:36:09 +00:00
John Maddock
3fc272db77 Fixed factorial code, so it is now fully tested. 
Added static asserts to ensure data tables in BGRAT don't exceed the number of factorials available.


[SVN r3021]
 2006-06-22 18:19:42 +00:00
John Maddock
0de45f85d1 Cleaned up BGRAT table size. 
Fixed typos in factorials.hpp.
Added needed using declarations in gamma ratios.
Cygwin fix for test.hpp.


[SVN r3017]
 2006-06-22 15:57:23 +00:00
Paul A. Bristow
57dafe1629 1st commit PAB 
[SVN r3010]
 2006-06-21 16:58:28 +00:00
John Maddock
e53697d039 Changed precision handling to set to full precision before formatting error message. 
[SVN r3009]
 2006-06-21 16:30:34 +00:00
John Maddock
80b35e3a3a Added additional overloads for domain and pole errors, so that formatted output including the actual value is now possible. 
[SVN r3008]
 2006-06-21 12:53:49 +00:00
Paul A. Bristow
5cb47fed11 1st try. 
[SVN r3006]
 2006-06-21 11:52:18 +00:00
John Maddock
a3157d5e12 Updated incomplete beta to use DiDonato and Morris's BGRAT routine in more cases. 
[SVN r3004]
 2006-06-20 17:21:29 +00:00
Paul A. Bristow
49b94b80a6 1st try. 
[SVN r3002]
 2006-06-20 14:21:46 +00:00
Paul A. Bristow
a33eceb484 *** empty log message *** 
[SVN r2999]
 2006-06-19 17:36:33 +00:00
Paul A. Bristow
ad83ab3ac2 1st release 
[SVN r2998]
 2006-06-19 17:08:19 +00:00
John Maddock
bbc25de1da Updated comments to reflect the state of the code. 
[SVN r2989]
 2006-06-14 17:42:01 +00:00
John Maddock
0cdad086c5 Updated beta inverse with slighly less costly code. 
[SVN r2988]
 2006-06-14 16:30:56 +00:00
John Maddock
9c044882bb More or less finished beta inverse and gamma inverse plus misc warning suppression and added missing docs. 
[SVN r2987]
 2006-06-14 13:59:25 +00:00
John Maddock
97e50116bb Fix warning 
[SVN r2975]
 2006-06-10 12:42:55 +00:00
John Maddock
cefe1428ea Chained series evaluation (optimisation). 
[SVN r2973]
 2006-06-09 11:42:11 +00:00
John Maddock
c3611d0c6b Added factorials and gamma function ratios. 
Updated incomplete beta with BGRAT routine from Didonato and Morris.
Added even and odd polynomials.
Lots of small tweeks.


[SVN r2972]
 2006-06-08 16:08:05 +00:00
John Maddock
5645943c9c Temme inverses are now working. 
[SVN r2963]
 2006-05-30 17:58:39 +00:00
John Maddock
586cb50312 Added some proper error handling. 
Got the incomplete beta inverse working better (Temme's expansions).


[SVN r2962]
 2006-05-26 16:42:50 +00:00
nobody
49e8efd352 This commit was manufactured by cvs2svn to create branch 'math_toolkit'. 
[SVN r2957]
 2006-05-21 15:53:01 +00:00
John Maddock
3d116fefb8 Initial commit. 
[SVN r2956]
 2006-05-21 15:53:01 +00:00
Alisdair Meredith
aae578b90a Patch for Borland compilers. 
[SVN r32424]
 2006-01-26 21:14:27 +00:00
John Maddock
4b41c341d6 Borland workarounds for a type deduction bug. 
[SVN r32399]
 2006-01-24 17:26:10 +00:00
John Maddock
604ec8f8ae Added needed #include 
[SVN r32316]
 2006-01-13 19:03:50 +00:00
John Maddock
2c309c55e4 Added missing math.h include. 
[SVN r32285]
 2006-01-11 10:31:58 +00:00
John Maddock
eeebc1031a Fix comment. 
[SVN r32279]
 2006-01-10 16:02:10 +00:00
Daryle Walker
ab2e8055f7 Merged brief-description comments to thier parent comment when they are adjacent. 
[SVN r2790]
 2005-12-27 13:07:13 +00:00
Daryle Walker
9c11319890 Optimized expressions now that boost::rational<> supports Boolean conversion. 
[SVN r2789]
 2005-12-27 12:42:35 +00:00
John Maddock
4c25a20654 Updated code to work even when numeric_limits<>::infinity() is unavailable. 
[SVN r32073]
 2005-12-16 14:10:05 +00:00
Daryle Walker
8b0b64986e Added Doxygen comments for non-member operator functions; did formatting changes to some code-based formula comments 
[SVN r2764]
 2005-12-03 07:42:13 +00:00
John Maddock
0cf71150fb Changed calculation of upper and lower safe bounds to avoid problems with std::sqrt(long double) returning infinities. 
[SVN r31885]
 2005-12-02 19:23:28 +00:00
John Maddock
fc99ec6f69 Tentative fix for pi having wrong value on one platform. 
Fix non-conforming call to atan2(0,0).


[SVN r31587]
 2005-11-07 18:48:47 +00:00
John Maddock
974a55463e Changed config logic to not use log1pf and expm1f unless C99 compatibility is specified. 
[SVN r31450]
 2005-10-24 17:07:46 +00:00
John Maddock
816a5d51bd Updated Nan test, in case signed-nan comparisons are buggy. 
[SVN r31435]
 2005-10-23 17:30:23 +00:00
John Maddock
f9d4a1a9f7 Tentative fix for some Sunpro failures: don't negate NAN's it doesn't work! 
[SVN r31430]
 2005-10-23 16:19:58 +00:00
John Maddock
67f638cbef Fixed up config to include <utility> via a wrapper that turns off TR1 inclusion in case the Boost.TR1 lib is in the compilers search path. 
Likewise for <complex> in the Math/TR1 components.


[SVN r31387]
 2005-10-19 16:38:40 +00:00
John Maddock
7308af23cc Added TR1 math lib headers. 
[SVN r31334]
 2005-10-15 11:59:11 +00:00
Daryle Walker
45ec720e9f Corrected 'll' function names to 'l1' 
[SVN r2682]
 2005-08-16 07:46:48 +00:00
Daryle Walker
b1b3339876 Added Doxygen comments for non-operator non-member functions 
[SVN r2679]
 2005-08-06 22:44:23 +00:00
Daryle Walker
ef97fe7be6 Preliminary code for Cayley-Dickson hypercomplex elements 
[SVN r2669]
 2005-07-27 23:57:04 +00:00
Douglas Gregor
76d9993b29 Need workaround header 
[SVN r28653]
 2005-05-04 13:40:34 +00:00
Hubert Holin
33e3aa09d0 Greater use of BOOST_WORKAROUND. 
[SVN r28214]
 2005-04-13 14:41:00 +00:00
Daryle Walker
8eb25c451e Added missing 'typename' 
[SVN r27219]
 2005-02-07 14:34:32 +00:00
John Maddock
a84e01fcc6 Changed files to BSL based on permissions in blanket-permission.txt 
[SVN r25451]
 2004-09-28 12:55:45 +00:00
Hubert Holin
de88dc5e38 Fixed BOOST_OCTONION_MEMBER_DIV_GENERATOR_4. 
[SVN r24750]
 2004-08-26 12:22:51 +00:00
Douglas Gregor
4471d90c95 License updates 
[SVN r24597]
 2004-08-19 15:19:17 +00:00
John Maddock
b1b4e377fc Added Daryle Walker's copyrights as requested. 
[SVN r24481]
 2004-08-15 10:13:49 +00:00
John Maddock
668e01c050 Removed boost.org copyright assignments, and reverted to orginal author (as based on cvs history). 
[SVN r24402]
 2004-08-11 10:59:33 +00:00
Douglas Gregor
2c8db4a10c License update 
[SVN r24180]
 2004-07-30 04:46:56 +00:00
Eric Niebler
a607b80188 remove minmax hack from win32.hpp and fix all places that could be affected by the minmax macros 
[SVN r22394]
 2004-02-26 18:27:02 +00:00
Daryle Walker
6792aa6963 Added bit-searching, bit-intersection, and scaling functions 
[SVN r1984]
 2004-02-17 03:43:40 +00:00
Daryle Walker
ae74e09deb Added addition and subtraction operators 
[SVN r1975]
 2004-02-14 09:20:48 +00:00
Daryle Walker
37a37daa30 Added absolute-value and sign functions 
[SVN r1974]
 2004-02-13 22:50:54 +00:00
Daryle Walker
2439924bff Added increment and decrement operators 
[SVN r1973]
 2004-02-13 22:06:26 +00:00
Daryle Walker
06bdded4f5 Fixed-up, hopefully, the implementation and exception policy of the bit-shift operators 
[SVN r1968]
 2004-02-13 10:34:00 +00:00
Daryle Walker
692dc90c11 Added unary + and - operators 
[SVN r1967]
 2004-02-13 01:05:16 +00:00
Daryle Walker
2ffe57e17c Added bitwise and-not function; did minor documentation tweak 
[SVN r1966]
 2004-02-12 13:29:21 +00:00
Daryle Walker
bbe2679ca9 Refactored implementations of bitwise operators 
[SVN r1965]
 2004-02-12 07:23:26 +00:00
Daryle Walker
11de9cab52 Added bit-shift operators; improved test for bit-reversal 
[SVN r1961]
 2004-02-11 08:20:28 +00:00
Daryle Walker
dba06e2c25 Fixed potential bug; did minor tweaks 
[SVN r1954]
 2004-02-10 09:59:38 +00:00
Daryle Walker
4488e0ff33 Added bit-string converter; did minor tweaks 
[SVN r1953]
 2004-02-10 03:08:15 +00:00
Daryle Walker
7b2bf3e95c Added bitwise-and, -or, and -xor operators, including combined assignment versions 
[SVN r1952]
 2004-02-10 01:11:29 +00:00
Daryle Walker
54484f15df Added even/odd test, comparison method and operators, and minor tweaks 
[SVN r1947]
 2004-02-09 05:42:24 +00:00
Daryle Walker
8067b24e14 Added single-bit-specific implementation of reset/set/flip 
[SVN r1939]
 2004-02-07 05:49:08 +00:00
Daryle Walker
dbb3463a51 Did more minor formatting changes 
[SVN r1938]
 2004-02-06 08:31:20 +00:00
Daryle Walker
008f2f77cc Did minor formatting changes 
[SVN r1937]
 2004-02-06 08:18:14 +00:00
Daryle Walker
bcf7ac8a38 Added method to assign an arbitrary bit pattern within a bit range of another big_whole object 
[SVN r1936]
 2004-02-06 08:00:19 +00:00
Daryle Walker
8ee4efc5f7 Added files for the multi-precision non-negative integer type based on std::valarray 
[SVN r1935]
 2004-02-05 10:36:38 +00:00
John Maddock
e8ddc9e36f Metrowerks appears to require static_cast workaround. 
[SVN r21445]
 2004-01-02 11:33:28 +00:00
John Maddock
2fbd58f0db Fixed gcc 3.2 and 3.3 build issues, tests still fail on cygwin, but pass on linux and mingw32 (seems to be a numeric_limits issue). 
[SVN r21433]
 2003-12-31 12:04:45 +00:00
Daryle Walker
bf81a3172e Moved my stuff to the Boost Software License, Version 1.0 
[SVN r1735]
 2003-11-05 07:06:17 +00:00
Hubert Holin
359857bd48 Gcc 3.3 compatibility 
[SVN r18677]
 2003-06-05 09:07:33 +00:00
Beman Dawes
a450e29e48 spurious semicolon removed to quiet warnings 
[SVN r18573]
 2003-05-28 00:51:53 +00:00
Beman Dawes
67408fbb0d replace cr with nl 
[SVN r17689]
 2003-03-01 18:10:27 +00:00
Hubert Holin
bccbe62dc8 now includes the config header (via quaternion.hpp) 
[SVN r17637]
 2003-02-25 13:40:16 +00:00
Hubert Holin
97edca44d8 now includes config header 
[SVN r17617]
 2003-02-24 15:54:16 +00:00
Hubert Holin
19e88761f5 Now includes the configuration header. 
[SVN r17546]
 2003-02-20 11:44:24 +00:00
Beman Dawes
02fdcc02cf add or update See www.boost.org comments 
[SVN r16708]
 2002-12-27 16:51:53 +00:00
Björn Karlsson
c30e40787d Update from Hubert Holin 
[SVN r15963]
 2002-10-22 06:24:39 +00:00
Markus Schöpflin
b518e7ac2c Merged fix from release branch to main trunk. 
[SVN r15607]
 2002-10-01 13:31:42 +00:00
Daryle Walker
314daf2a52 Added modulo arithmetic library 
[SVN r522]
 2002-09-20 17:04:13 +00:00
Daryle Walker
5255f7d989 Adjusted for moving the Compile-Time and Run-Time GCD and LCM components to new, separate headers 
[SVN r14273]
 2002-07-02 08:31:15 +00:00
Daryle Walker
93af018235 Moved the Compile-Time and Run-Time components to separate headers 
[SVN r14272]
 2002-07-02 07:51:48 +00:00
John Maddock
5346f498c3 Added fixes for the case that BOOST_NO_LIMITS_COMPILE_TIME_CONSTANTS is defined. 
[SVN r13684]
 2002-05-06 10:35:57 +00:00
Douglas Gregor
a937b274dc Fixes for GCC 2.95.x 
[SVN r12639]
 2002-02-02 15:39:51 +00:00
Douglas Gregor
413c15ad7f Don't use std::ends for ostringstream 
[SVN r12287]
 2002-01-12 13:29:30 +00:00
Dave Abrahams
9ccffd661f Integrated MSVC fixes from Daryle 
[SVN r12012]
 2001-12-11 04:59:26 +00:00
Jens Maurer
bc5d3ed695 fix errors (from Daryle Walker) 
[SVN r11935]
 2001-12-05 20:56:26 +00:00
Dave Abrahams
d199e37ac9 Daryle Walker's GCD library submission 
[SVN r11814]
 2001-11-29 21:42:58 +00:00
Darin Adler
28650c0720 Math library fixes for BOOST_NO_STDC_NAMESPACE. 
[SVN r11346]
 2001-10-05 23:37:12 +00:00
Jens Maurer
3c8349e7ec update from Hubert Holin 
now works with significantly more compilers


[SVN r11284]
 2001-09-26 21:29:40 +00:00
Darin Adler
f3b38b93c4 Use <boost/limits.hpp> to help out platforms that don't have <limits>. 
[SVN r11246]
 2001-09-25 06:52:03 +00:00
Jens Maurer
d77e86b74c moved to namespace boost::math 
formatting changes


[SVN r10568]
 2001-07-08 21:34:21 +00:00
Jens Maurer
f483d1189b Special functions, octonions, quaternions by Hubert Holin 
[SVN r10404]
 2001-06-23 08:02:37 +00:00