packages/random.git
2 years agobasic polar method rejection sampler for normal distribution
Carter Tazio Schonwald [Sat, 27 May 2017 21:07:30 +0000 (17:07 -0400)] 
basic polar method rejection sampler for normal distribution

2 years agoprovisional fast and really sloppy unit interval generator. For basic uses should...
Carter Tazio Schonwald [Fri, 26 May 2017 15:50:44 +0000 (11:50 -0400)] 
provisional fast and really sloppy unit interval generator. For basic uses should be good enough.

2 years agofix case on CHANGELOG.md
Carter Tazio Schonwald [Sun, 21 May 2017 20:08:11 +0000 (16:08 -0400)] 
fix case on CHANGELOG.md

2 years agofixed up file names
Carter Tazio Schonwald [Sun, 21 May 2017 19:02:31 +0000 (15:02 -0400)] 
fixed  up file names

2 years agocorrect the PCG module name to reflect that its the 32bit word sampler.
Carter Tazio Schonwald [Sun, 21 May 2017 18:35:08 +0000 (14:35 -0400)] 
correct the PCG module name to reflect that its the 32bit word sampler.
have a candidate travis file,
added entropy to the dependencies for now (spoke with TomMD and we agreed that they should be combined sometime)
started work on basic distribution generation tools

2 years agoformmatting tweaks to splitmix code plus enabling wall
Carter Tazio Schonwald [Mon, 17 Apr 2017 19:16:33 +0000 (15:16 -0400)] 
formmatting tweaks to splitmix code plus enabling wall

2 years agoAdd PCG internal module
Christopher Chalmers [Sat, 18 Feb 2017 22:13:25 +0000 (22:13 +0000)] 
Add PCG internal module

2 years agobasic sampler values
Carter Tazio Schonwald [Fri, 10 Feb 2017 19:57:43 +0000 (14:57 -0500)] 
basic sampler values

2 years agodrop comments
Carter Tazio Schonwald [Fri, 10 Feb 2017 19:48:28 +0000 (14:48 -0500)] 
drop comments

2 years agoRandomT in the style of s -> m (a,s) is working
Carter Tazio Schonwald [Fri, 10 Feb 2017 19:46:57 +0000 (14:46 -0500)] 
RandomT in the style of s -> m (a,s) is working

2 years agoinitial attempt to see how far i could get with s -> (# m a , s #) monad flavor
Carter Tazio Schonwald [Fri, 10 Feb 2017 19:15:22 +0000 (14:15 -0500)] 
initial attempt to see how far i could get with  s -> (# m a , s #) monad flavor
thats impossible, so switchign to s -> m (a,s) flavor

2 years agopure Splitmix implementation v2
Carter Tazio Schonwald [Sun, 15 Jan 2017 15:30:12 +0000 (10:30 -0500)] 
pure Splitmix implementation

2 years agoversion 2.0 of random begins
Carter Tazio Schonwald [Sun, 15 Jan 2017 19:43:11 +0000 (14:43 -0500)] 
version 2.0 of random begins

3 years agoMerge pull request #33 from thomie/master
Carter Tazio Schonwald [Wed, 20 Jan 2016 16:36:29 +0000 (11:36 -0500)] 
Merge pull request #33 from thomie/master

Fix for GHC testsuite

3 years agoGHC testsuite: require random to be installed
Thomas Miedema [Wed, 20 Jan 2016 15:35:06 +0000 (16:35 +0100)] 
GHC testsuite: require random to be installed

3 years agoMerge pull request #24 from frankcash/patch-1
Edward Kmett [Fri, 31 Jul 2015 15:22:54 +0000 (11:22 -0400)] 
Merge pull request #24 from frankcash/patch-1

Update README.md

3 years agoMerge pull request #22 from hvr/pr-detab
Edward Kmett [Fri, 31 Jul 2015 15:22:37 +0000 (11:22 -0400)] 
Merge pull request #22 from hvr/pr-detab

detab

4 years agoMake benchmark build and add benchmark for tf-random.
Dominic Steinitz [Sat, 4 Apr 2015 17:25:33 +0000 (18:25 +0100)] 
Make benchmark build and add benchmark for tf-random.

4 years agoAdd some tests demonstrating https://github.com/haskell/random/issues/25.
Dominic Steinitz [Sat, 4 Apr 2015 16:29:03 +0000 (17:29 +0100)] 
Add some tests demonstrating https://github.com/haskell/random/issues/25.

4 years agoUpdate README.md
Frank Cash [Fri, 13 Mar 2015 03:04:47 +0000 (23:04 -0400)] 
Update README.md

Updates Haskell 98 link

4 years agoSet GitHub's Git repo as primary repo
Herbert Valerio Riedel [Sun, 21 Dec 2014 09:45:03 +0000 (10:45 +0100)] 
Set GitHub's Git repo as primary repo

4 years agoM-x delete-trailing-whitespace & M-x untabify
Herbert Valerio Riedel [Sun, 21 Dec 2014 09:43:15 +0000 (10:43 +0100)] 
M-x delete-trailing-whitespace & M-x untabify

4 years agoset maintainer v1.1
Edward Kmett [Tue, 16 Sep 2014 21:34:25 +0000 (17:34 -0400)] 
set maintainer

4 years agoMerge pull request #19 from mainland/master
Edward Kmett [Tue, 16 Sep 2014 21:03:24 +0000 (17:03 -0400)] 
Merge pull request #19 from mainland/master

Fix compilation with GHC HEAD.

4 years agoMerge pull request #18 from cartazio/master
Edward Kmett [Tue, 16 Sep 2014 20:25:35 +0000 (16:25 -0400)] 
Merge pull request #18 from cartazio/master

naive extra-files bits

4 years agoFix unused import warning.
Geoffrey Mainland [Tue, 16 Sep 2014 15:57:35 +0000 (11:57 -0400)] 
Fix unused import warning.

4 years agoHide empty when importing Control.Monad.
Geoffrey Mainland [Tue, 16 Sep 2014 15:56:31 +0000 (11:56 -0400)] 
Hide empty when importing Control.Monad.

This is need for compatibility with AMP since Control.Monad now exports
empty.

4 years agonaive extra files bits
Carter Tazio Schonwald [Tue, 16 Sep 2014 01:53:10 +0000 (21:53 -0400)] 
naive extra files bits

4 years agoMerge pull request #16 from cartazio/master
Edward Kmett [Mon, 15 Sep 2014 22:02:41 +0000 (18:02 -0400)] 
Merge pull request #16 from cartazio/master

adding more changes to the change log

4 years agoadding more changes to the change log
Carter Tazio Schonwald [Mon, 15 Sep 2014 21:46:04 +0000 (17:46 -0400)] 
adding more changes to the change log

4 years agoMerge pull request #15 from cartazio/master
Edward Kmett [Mon, 15 Sep 2014 19:50:07 +0000 (15:50 -0400)] 
Merge pull request #15 from cartazio/master

prepping for the 1.1 fixup release

4 years agoMerge pull request #9 from bgamari/master
Edward Kmett [Mon, 15 Sep 2014 18:53:25 +0000 (14:53 -0400)] 
Merge pull request #9 from bgamari/master

rangeTest: Fix type for "CSigAtomic R" test

4 years agoprepping for the 1.1 fixup release
Carter Tazio Schonwald [Mon, 15 Sep 2014 18:42:59 +0000 (14:42 -0400)] 
prepping for the 1.1 fixup release

1) still need to merge in  https://github.com/haskell/random/pull/9
2) should we force -O2 in the ghcoptions? (its currently set that way, but is that a good idea/needed here?)

4 years agoMerge pull request #12 from 23Skidoo/unbreak-ghc-7.4
Edward Kmett [Mon, 25 Aug 2014 15:25:49 +0000 (11:25 -0400)] 
Merge pull request #12 from 23Skidoo/unbreak-ghc-7.4

Support base < 4.6 / GHC 7.4.

4 years agoSupport base < 4.6 / GHC 7.4.
Mikhail Glushenkov [Sat, 23 Aug 2014 18:55:31 +0000 (20:55 +0200)] 
Support base < 4.6 / GHC 7.4.

Fixes #10.

4 years agoAdd Travis icon.
Mikhail Glushenkov [Sat, 23 Aug 2014 19:19:50 +0000 (21:19 +0200)] 
Add Travis icon.

4 years agoUpdate .gitignore.
Mikhail Glushenkov [Sat, 23 Aug 2014 19:16:43 +0000 (21:16 +0200)] 
Update .gitignore.

4 years agoAdd .travis.yml.
Mikhail Glushenkov [Sat, 23 Aug 2014 19:15:48 +0000 (21:15 +0200)] 
Add .travis.yml.

4 years agorangeTest: Fix type for "CSigAtomic R" test
Ben Gamari [Fri, 22 Aug 2014 21:41:42 +0000 (17:41 -0400)] 
rangeTest: Fix type for "CSigAtomic R" test

4 years agoBump version for various bugfixes
Ryan Newton [Fri, 22 Aug 2014 19:52:30 +0000 (15:52 -0400)] 
Bump version for various bugfixes

4 years agoMerge pull request #5 from bgamari/master
Ryan Newton [Fri, 22 Aug 2014 19:50:01 +0000 (15:50 -0400)] 
Merge pull request #5 from bgamari/master

rangeTest: Fix signed-ness of types

4 years agoMerge pull request #6 from thomie/master
Ryan Newton [Fri, 22 Aug 2014 19:49:08 +0000 (15:49 -0400)] 
Merge pull request #6 from thomie/master

Update README + issue tracker link in cabal file

4 years agoMerge pull request #7 from thomie/T8899
Ryan Newton [Fri, 22 Aug 2014 19:49:01 +0000 (15:49 -0400)] 
Merge pull request #7 from thomie/T8899

Fix for ghc ticket #8899: StdGen does not generate 0

4 years agoMerge pull request #8 from thomie/memory-leaks
Ryan Newton [Fri, 22 Aug 2014 19:48:56 +0000 (15:48 -0400)] 
Merge pull request #8 from thomie/memory-leaks

Fix for memory leaks (GHC #7936 and #4218)

4 years agoRevert "rangeTest fails on Windows (#7379)"
Ben Gamari [Tue, 19 Aug 2014 20:02:23 +0000 (16:02 -0400)] 
Revert "rangeTest fails on Windows (#7379)"

This reverts commit 2117e38729adaa4f465f3c5b7a8c5c4d77702d3f.

We no longer assume that C types are unsigned.

4 years agorangeTest: Fix signed-ness of types
Ben Gamari [Sat, 22 Mar 2014 14:29:45 +0000 (15:29 +0100)] 
rangeTest: Fix signed-ness of types

On some architectures types like CChar are signed whereas the use previously
assumed they were unsigned. Fix this by relying on Bounded instances where
possible.

4 years agoUse strict fields for StdGen (GHC #7936)
Thomas Miedema [Thu, 10 Jul 2014 18:56:17 +0000 (20:56 +0200)] 
Use strict fields for StdGen (GHC #7936)

4 years agoUse atomicModifyIORef' (strict) (GHC #4218)
Thomas Miedema [Thu, 10 Jul 2014 19:01:44 +0000 (21:01 +0200)] 
Use atomicModifyIORef' (strict) (GHC #4218)

4 years agoMake TestRandomRs (GHC #4218) fast and add to cabal file
Thomas Miedema [Tue, 1 Jul 2014 14:28:57 +0000 (16:28 +0200)] 
Make TestRandomRs (GHC #4218) fast and add to cabal file

4 years agoCleanup + add comments
Thomas Miedema [Tue, 8 Jul 2014 12:32:26 +0000 (14:32 +0200)] 
Cleanup + add comments

Fixme reverred to formatting of comments, which was done in
24260b389852ab109de6b62822d889d0e66ae723

4 years agoThe lowest int generated by StdGen is 1. Fixes ghc #8899.
Thomas Miedema [Tue, 8 Jul 2014 12:06:40 +0000 (14:06 +0200)] 
The lowest int generated by StdGen is 1. Fixes ghc #8899.

This follows from the following line in the function stdNext
(z' is the next generated Int):

    z' = if z < 1 then z + 2147483562 else z

4 years agoUpdate README + issue tracker link in cabal file
Thomas Miedema [Fri, 27 Jun 2014 13:38:24 +0000 (15:38 +0200)] 
Update README + issue tracker link in cabal file

5 years agoMerge pull request #4 from NovaDenizen/master
Ryan Newton [Fri, 28 Mar 2014 20:01:01 +0000 (16:01 -0400)] 
Merge pull request #4 from NovaDenizen/master

fix for randomIvalInteger, ghc #8898

5 years agoMerge remote-tracking branch 'remotes/downstream/master'
Ryan Newton [Fri, 28 Mar 2014 19:53:29 +0000 (15:53 -0400)] 
Merge remote-tracking branch 'remotes/downstream/master'

Conflicts:
.gitignore

5 years agofix for randomIvalInteger, ghc #8898
Ken Bateman [Sat, 22 Mar 2014 20:42:44 +0000 (20:42 +0000)] 
fix for randomIvalInteger, ghc #8898

5 years agoVersion bump to go with prev
Ryan Newton [Wed, 5 Feb 2014 06:17:13 +0000 (01:17 -0500)] 
Version bump to go with prev

5 years agoMerge branch 'master' of github.com:haskell/random
Ryan Newton [Tue, 4 Feb 2014 21:25:26 +0000 (16:25 -0500)] 
Merge branch 'master' of github.com:haskell/random

5 years agoUse GHC.Exts.build in randoms, randomRs to achieve fusion
Johan Kiviniemi [Sun, 26 Jan 2014 12:59:55 +0000 (14:59 +0200)] 
Use GHC.Exts.build in randoms, randomRs to achieve fusion

5 years agoMerge pull request #3 from thapakrish/master
Ryan Newton [Tue, 4 Feb 2014 21:17:01 +0000 (13:17 -0800)] 
Merge pull request #3 from thapakrish/master

Edited readme

5 years agoMerge pull request #2 from pcapriotti/master
Ryan Newton [Tue, 4 Feb 2014 21:16:24 +0000 (13:16 -0800)] 
Merge pull request #2 from pcapriotti/master

Add .gitignore

5 years agoUpdate Git repo URL in `.cabal` file
Herbert Valerio Riedel [Wed, 11 Sep 2013 13:15:01 +0000 (15:15 +0200)] 
Update Git repo URL in `.cabal` file

6 years agoUpdated link
Krishna Thapa [Mon, 29 Apr 2013 06:12:25 +0000 (01:12 -0500)] 
Updated link

6 years agoAdd .gitignore
Gabor Greif [Tue, 26 Feb 2013 21:10:47 +0000 (22:10 +0100)] 
Add .gitignore

6 years agoFollow changes in the testsuite
Ian Lynagh [Mon, 11 Feb 2013 14:11:34 +0000 (14:11 +0000)] 
Follow changes in the testsuite

6 years agorangeTest fails on Windows (#7379)
Ian Lynagh [Sun, 3 Feb 2013 16:28:55 +0000 (16:28 +0000)] 
rangeTest fails on Windows (#7379)

7 years agoAdd .gitignore.
Paolo Capriotti [Tue, 6 Mar 2012 10:57:35 +0000 (10:57 +0000)] 
Add .gitignore.

7 years agoMinor: comments only random-1.0.1.1-release
Ryan Newton [Tue, 6 Dec 2011 20:07:53 +0000 (15:07 -0500)] 
Minor: comments only

7 years agoMinor, Binsearch documentation update. TODO: Should move to using submodule for...
Ryan Newton [Mon, 14 Nov 2011 15:21:53 +0000 (10:21 -0500)] 
Minor, Binsearch documentation update.  TODO: Should move to using submodule for this.

7 years agoMinor: change the comments to refer to the correct LICENSE file. ghc-7.4
Ryan Newton [Sat, 22 Oct 2011 10:01:00 +0000 (06:01 -0400)] 
Minor: change the comments to refer to the correct LICENSE file.

7 years agoFix test following removal of Num's superclasses
Ian Lynagh [Wed, 12 Oct 2011 17:02:26 +0000 (18:02 +0100)] 
Fix test following removal of Num's superclasses

7 years agoFixed bug in Ticket 5501. But there appears to be a significant performance regressi...
Ryan Newton [Fri, 30 Sep 2011 05:54:44 +0000 (01:54 -0400)] 
Fixed bug in Ticket 5501.  But there appears to be a significant performance regression that warrants further investigation.

7 years agoFixed benchmark script to NOT use splittableGen
Ryan Newton [Fri, 30 Sep 2011 05:53:58 +0000 (01:53 -0400)] 
Fixed benchmark script to NOT use splittableGen

7 years agominor: DEVLOG comments
Ryan Newton [Fri, 30 Sep 2011 05:17:37 +0000 (01:17 -0400)] 
minor: DEVLOG comments

7 years agoBumped version for recent bugfixes but disabled SplittableGen for now.
Ryan Newton [Thu, 25 Aug 2011 15:14:45 +0000 (11:14 -0400)] 
Bumped version for recent bugfixes but disabled SplittableGen for now.

7 years agoMerge pull request #1 from dterei/master
Ryan Newton [Tue, 9 Aug 2011 03:51:50 +0000 (20:51 -0700)] 
Merge pull request #1 from dterei/master

Patch for random to use Safe Haskell

7 years agoUse Safe Haskell for GHC >= 7.2
David Terei [Thu, 4 Aug 2011 21:03:48 +0000 (14:03 -0700)] 
Use Safe Haskell for GHC >= 7.2

7 years agoCleaned up the previous change, removing commented line and all reference to opts.
Ryan Newton [Thu, 28 Jul 2011 21:00:27 +0000 (17:00 -0400)] 
Cleaned up the previous change, removing commented line and all reference to opts.

7 years agoDisabled opts line that causes data races.
Ryan Newton [Mon, 25 Jul 2011 21:42:12 +0000 (17:42 -0400)] 
Disabled opts line that causes data races.

7 years agoAdding the simple speed test from ticket #427.
Ryan Newton [Tue, 28 Jun 2011 18:25:45 +0000 (14:25 -0400)] 
Adding the simple speed test from ticket #427.

7 years agoAdded currently unworking test from ticket 4218.
Ryan Newton [Tue, 28 Jun 2011 17:44:56 +0000 (13:44 -0400)] 
Added currently unworking test from ticket 4218.

7 years agoThe tests are failing with stack overflow on large random Integers (2^5000). This...
Ryan Newton [Tue, 28 Jun 2011 14:43:22 +0000 (10:43 -0400)] 
The tests are failing with stack overflow on large random Integers (2^5000).  This represents a genuine problem, but for now I'm commenting out the relevant tests to get things passing.  This problem and several others are fixed on the new_api branch.

7 years agoAdded tests and benchmarks for BIG Integers.
Ryan Newton [Tue, 28 Jun 2011 14:07:45 +0000 (10:07 -0400)] 
Added tests and benchmarks for BIG Integers.

7 years agoAdded README.md for github.
Ryan Newton [Tue, 28 Jun 2011 05:10:31 +0000 (01:10 -0400)] 
Added README.md for github.

7 years agoExtended rangeTest to handle some more corner cases.
Ryan Newton [Tue, 28 Jun 2011 05:06:21 +0000 (01:06 -0400)] 
Extended rangeTest to handle some more corner cases.

7 years agoUpdated rangeTest to test randomR as well as random.
Ryan Newton [Tue, 28 Jun 2011 04:11:41 +0000 (00:11 -0400)] 
Updated rangeTest to test randomR as well as random.

7 years agoAdded a new test to check the range on each type\'s Random instance.
Ryan Newton [Tue, 28 Jun 2011 03:41:20 +0000 (23:41 -0400)] 
Added a new test to check the range on each type\'s Random instance.

7 years agoConverted Float/Double randomR methods to simply use the existing random methods...
Ryan Newton [Mon, 27 Jun 2011 17:38:38 +0000 (13:38 -0400)] 
Converted Float/Double randomR methods to simply use the existing random methods to generate a coefficient.

7 years agoMinor: using unsafeDupablePerformIO to be more fair to MWC.
Ryan Newton [Mon, 27 Jun 2011 03:51:11 +0000 (23:51 -0400)] 
Minor: using unsafeDupablePerformIO to be more fair to MWC.

7 years agoRearranged module organization for Benchmark. Added Makefile and more comparison...
Ryan Newton [Mon, 27 Jun 2011 03:47:13 +0000 (23:47 -0400)] 
Rearranged module organization for Benchmark.  Added Makefile and more comparison against other RNGs on hackage.

7 years agoBumped the version number to 1.0.0.4 because of the previous.
Ryan Newton [Mon, 27 Jun 2011 03:02:49 +0000 (23:02 -0400)] 
Bumped the version number to 1.0.0.4 because of the previous.

7 years agoMinor: Added a bit of testing related code corresponding to ticket 5278.
Ryan Newton [Mon, 27 Jun 2011 02:52:08 +0000 (22:52 -0400)] 
Minor: Added a bit of testing related code corresponding to ticket 5278.

7 years agoExtended the float fix to Doubles. Fixed a bug introduced by the last checkin. ...
Ryan Newton [Sun, 26 Jun 2011 06:18:52 +0000 (02:18 -0400)] 
Extended the float fix to Doubles.  Fixed a bug introduced by the last checkin.  next is not a sufficient substitute for randomBounded in the Int instance.  It doesn't have sufficient genRange.

7 years agoFixed performance dent for random generating Ints. I can't explain why it was slower...
Ryan Newton [Sun, 26 Jun 2011 04:24:46 +0000 (00:24 -0400)] 
Fixed performance dent for random generating Ints.  I can't explain why it was slower than other types using randomBounded, but this patch circumvents the problem by using next directly.

7 years agoTweaked SimpleRNGBench to be warning-free as well and added some additional tests...
Ryan Newton [Sun, 26 Jun 2011 04:08:40 +0000 (00:08 -0400)] 
Tweaked SimpleRNGBench to be warning-free as well and added some additional tests of randomR alongside random.

7 years agoTweaked BinSearch to issue no warnings.
Ryan Newton [Sun, 26 Jun 2011 03:59:51 +0000 (23:59 -0400)] 
Tweaked BinSearch to issue no warnings.

7 years agoSmall tweaks to fix Integer-defaulting warnings.
Ryan Newton [Sun, 26 Jun 2011 03:54:36 +0000 (23:54 -0400)] 
Small tweaks to fix Integer-defaulting warnings.

7 years agoAdding a DEVLOG file to keep track of development.
Ryan Newton [Sat, 25 Jun 2011 02:44:54 +0000 (22:44 -0400)] 
Adding a DEVLOG file to keep track of development.

7 years agoFix #5133 - also yields this change a 240X speedup in generating random floats in...
Ryan Newton [Sat, 25 Jun 2011 02:39:44 +0000 (22:39 -0400)] 
Fix #5133 - also yields this change a 240X speedup in generating random floats in my tests.

7 years agoCleaned up the simple benchmark and added tests for non-Int datatypes. The results...
Ryan Newton [Fri, 24 Jun 2011 22:22:26 +0000 (18:22 -0400)] 
Cleaned up the simple benchmark and added tests for non-Int datatypes.  The results are a bit surprising.  Float generation is performing badly.

Here are some results from a 3.33ghz Intel Nehalem:

  Cost of rdtsc (ffi call):    75
  Approx getCPUTime calls per second: 206,493
  Approx clock frequency:  3,336,174,789
  First, timing with System.Random interface:
    112,276,629 randoms generated [constant zero gen]         ~ 29.71 cycles/int
     14,289,712 randoms generated [System.Random stdGen]      ~ 233 cycles/int
         82,546 randoms generated [System.Random Floats]      ~ 40,416 cycles/int
         83,138 randoms generated [System.Random CFloats]     ~ 40,128 cycles/int
      2,533,007 randoms generated [System.Random Doubles]     ~ 1,317 cycles/int
        841,737 randoms generated [System.Random Integers]    ~ 3,963 cycles/int
      4,704,318 randoms generated [System.Random Bools]       ~ 709 cycles/int

7 years agoAdded some very minor comments and a simple benchmarking script.
Ryan Newton [Fri, 24 Jun 2011 18:26:22 +0000 (14:26 -0400)] 
Added some very minor comments and a simple benchmarking script.

Using this script I saw a 13X regression in stdGen in recent revisions
vs the 1.0.0.3 version, dropping to only 900K random ints/sec on my
machine.  However, it was some kind of heisenbug because now it's gone
away for unknown reasons.  Watch out for it in the future though.