There's ORLite::PDL
http://search.cpan.org/~adamk/ORLite-PDL-0.02/lib/ORLite/PDL.pm

Not DBI, but along the same line of thought.

Yes diag sounds like a great tool here. Thanks for pointing it out!

Best,
Maggie

On Thu, Dec 22, 2011 at 10:51 PM, chm <[email protected]> wrote:

> On 12/22/2011 10:43 PM, Puneet Kishor wrote:
>
>>
>> On Dec 22, 2011, at 9:29 PM, Maggie X wrote:
>>
>>  Thank you for trying it out. Glad it worked! But I'm not sure why
>>> just printing out the numbers would make it work. Was there
>>> anything else that you have tried to make it work?
>>>
>>>
>>
>> No... just added the `print` statements. That was it. Maybe the
>> `print` output resolved to `true` and that was it. You might want to
>> alter that test to account for such things.
>>
>> Many thanks for your help. Now I can have a lot of fun wasting time
>> trying to redo stuff in PDL::Stats that I can already do in
>> Postgres/R ;-)
>>
>>
>> On a tangential note -- seriously, it would be a great boon if PDL
>> could ingest results of DBI queries as piddles, kinda like
>> (theoretical code ahead)
>>
>> $sth->execute; my $piddle = $sth->fetchrow_arrayref; my $result =
>> $piddle->do_magic; return $result->to_json;
>>
>
> Two things: it appears that your email app has problems with
> correctly encoding line endings (it seems to be a mac only
> issue), and I seem to recall a module for reading database
> info as piddles.  Try checking the ones in the list of
> "other" PDL modules I posted in the past week.  Sorry, I
> can't recall and I'm not using DBI stuff and the moment
> myself....
>
> Cheers,
> Chris
>
>
>  piano, piano...
>>
>>
>>
>>
>>  Best, Maggie
>>>
>>> On Thu, Dec 22, 2011 at 10:13 PM, Puneet
>>> Kishor<[email protected]>  wrote:
>>>
>>>  Here ya go... I did as you said, and now it says all tests
>>>> passed. My guess is that the darn thing would work well if I just
>>>> installed it without testing it, but `cpanm` stops as soon as the
>>>> first test croaks.
>>>>
>>>> punkish@mumbai ~/Projects/PDL-Stats-0.5.5$**make test
>>>> PERL_DL_NONLAZY=1 /opt/local/bin/perl "-MExtUtils::Command::MM"
>>>> "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
>>>> t/stats_basic.t ..... ok t/stats_glm.t ....... ok
>>>> t/stats_kmeans.t .... 1/18 R2   0.786191536748329 centroid [ [
>>>> 1.5  5.5] [ 8.5   12] [15.5   19] ]
>>>>
>>>> cluster [ [1 1 1 1 0 0 0] [0 0 0 0 1 1 1] ]
>>>>
>>>> ms [ [      1.25       0.25] [      1.25 0.66666667] [      1.25
>>>> 0.66666667] ]
>>>>
>>>> n       [4 3] -6.66666570836583e-08R2 0.786191536748329 centroid
>>>> [ [ 1.5  5.5] [ 8.5   12] [15.5   19] ]
>>>>
>>>> cluster [ [1 1 1 1 0 0 0] [0 0 0 0 1 1 1] ]
>>>>
>>>> ms [ [      1.25       0.25] [      1.25 0.66666667] [      1.25
>>>> 0.66666667] ]
>>>>
>>>> n       [4 3] t/stats_kmeans.t .... ok t/stats_ols_rptd.t .. ok
>>>> t/stats_ts.t ........ ok All tests successful. Files=5,
>>>> Tests=147,  1 wallclock secs ( 0.03 usr  0.01 sys +  0.93 cusr
>>>> 0.03 csys =  1.00 CPU) Result: PASS No tests defined for
>>>> PDL::Stats::Basic extension. PERL_DL_NONLAZY=1
>>>> /opt/local/bin/perl "-MExtUtils::Command::MM"
>>>>
>>>
>>  "-e" "test_harness(0, '../blib/lib', '../blib/arch')" t/*.t
>>>> t/stats_distr.t .. ok All tests successful. Files=1, Tests=43,  0
>>>> wallclock secs ( 0.02 usr  0.00 sys +  0.13 cusr  0.00 csys =
>>>> 0.15 CPU) Result: PASS No tests defined for PDL::Stats::GLM
>>>> extension. No tests defined for PDL::Stats::Kmeans extension.
>>>> PERL_DL_NONLAZY=1 /opt/local/bin/perl "-MExtUtils::Command::MM"
>>>>
>>>
>>  "-e" "test_harness(0, '../blib/lib', '../blib/arch')" t/*.t
>>>> t/cdf.t .. ok All tests successful. Files=1, Tests=4,  0
>>>> wallclock secs ( 0.02 usr  0.00 sys +  0.06 cusr  0.00 csys =
>>>> 0.08 CPU) Result: PASS No tests defined for PDL::Stats::TS
>>>> extension. punkish@mumbai ~/Projects/PDL-Stats-0.5.5$
>>>>
>>>> On Dec 22, 2011, at 9:08 PM, Maggie X wrote:
>>>>
>>>>  I've seen cases where the test fails simply because of the
>>>>> precision
>>>>>
>>>> differ
>>>>
>>>>> on different platforms. That's why I wanted to see what you get
>>>>> when you
>>>>>
>>>> do
>>>>
>>>>> print $m{ms}->sumover(). But I guess make test doesn't actually
>>>>> print out the values to stdout... How about this?
>>>>>
>>>>> Add
>>>>>
>>>>> print STDERR t_kmeans_bad();
>>>>>
>>>>> 168>   is(tapprox( t_kmeans_bad(), 0 ), 1); 169>   sub
>>>>> t_kmeans_bad { 170>     my $data = sequence 7, 3; 171>
>>>>> $data = $data->setbadat(4,0); 172>     my %m =
>>>>> $data->kmeans({NCLUS=>2, NTRY=>10, V=>0});
>>>>>
>>>>
>>
>>>>> print STDERR "$_\t$m{$_}\n" for (sort keys %m);
>>>>>
>>>>> 173>     return sum( $m{ms}->sumover - pdl qw( 1.5  1.9166667
>>>>> 1.9166667
>>>>>
>>>> ) );
>>>>
>>>>> 174>   }
>>>>>
>>>>>
>>>>> Best, Maggie
>>>>>
>>>>>
>>>>> On Thu, Dec 22, 2011 at 9:06 PM, Puneet
>>>>> Kishor<[email protected]>
>>>>>
>>>> wrote:
>>>>
>>>>>
>>>>>  I should have added my congratulations and gratitude for
>>>>>> taking 2.4.9
>>>>>>
>>>>>
>>  to a
>>>>
>>>>> point that it builds and installs beautifully, without any
>>>>>> heartburn.
>>>>>>
>>>>>
>>  Perhaps because most of the dependencies got installed when I
>>>>>> installed 2.4.6 via MacPorts.
>>>>>>
>>>>>>
>>>>>> I am doing a lot of stuff using R, so am curious to try out
>>>>>> PDL::Stats. For now, I am using R compiled inside Postgres
>>>>>> using PL/R... what a wonderful experience to get my data out
>>>>>> of Postgres and analyze it right within the database.
>>>>>>
>>>>>> Wrt PDL::Stats, my guess is the test failure has nothing to
>>>>>> do with it being on a Mac/10.7. Perl 5.14.1 should be Perl
>>>>>> 5.14.1 no matter what
>>>>>>
>>>>>
>>  platform, no?
>>>>>>
>>>>>> On Dec 22, 2011, at 7:50 PM, chm wrote:
>>>>>>
>>>>>>  On 12/22/2011 7:48 PM, Puneet Kishor wrote:
>>>>>>>
>>>>>>>> Inspired by the lovely blurb below, I tried to install
>>>>>>>> PDL::Stats but
>>>>>>>>
>>>>>>> failed. Here is the relevant bit from the log of my failed
>>>>>> attempt --
>>>>>>
>>>>>
>>
>>>>>>>> ---- PERL_DL_NONLAZY=1 /opt/local/bin/perl
>>>>>>>> "-MExtUtils::Command::MM" "-e"
>>>>>>>>
>>>>>>> "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
>>>>>>
>>>>>>> t/stats_basic.t ..... ok t/stats_glm.t ....... ok
>>>>>>>>
>>>>>>>> #   Failed test at t/stats_kmeans.t line 168. #
>>>>>>>> got: '' #     expected: '1' # Looks like you failed 1
>>>>>>>> test of 18. t/stats_kmeans.t .... Dubious, test returned
>>>>>>>> 1 (wstat 256, 0x100) Failed 1/18 subtests
>>>>>>>> t/stats_ols_rptd.t .. ok t/stats_ts.t ........ ok
>>>>>>>>
>>>>>>>> Test Summary Report ------------------- t/stats_kmeans.t
>>>>>>>> (Wstat: 256 Tests: 18 Failed: 1) Failed test:  13
>>>>>>>> Non-zero exit status: 1 Files=5, Tests=147,  1 wallclock
>>>>>>>> secs ( 0.03 usr  0.01 sys +  0.91
>>>>>>>>
>>>>>>> cusr
>>>>
>>>>> 0.04 csys =  0.99 CPU)
>>>>>>
>>>>>>> Result: FAIL Failed 1/5 test programs. 1/147 subtests
>>>>>>>> failed. make: *** [test_dynamic] Error 255 ->   FAIL
>>>>>>>> Installing PDL::Stats failed. See
>>>>>>>>
>>>>>>> /Volumes/roller/Users/punkish/**.cpanm/build.log for details.
>>>>>>
>>>>>>> ----
>>>>>>>>
>>>>>>>>
>>>>>>>> Here is line 168 (test #13) from t/stats_kmeans.t
>>>>>>>>
>>>>>>>> 168>   is(tapprox( t_kmeans_bad(), 0 ), 1); 169>   sub
>>>>>>>> t_kmeans_bad { 170>     my $data = sequence 7, 3; 171>
>>>>>>>> $data = $data->setbadat(4,0); 172>     my %m =
>>>>>>>> $data->kmeans({NCLUS=>2, NTRY=>10, V=>0}); 173>
>>>>>>>> return sum( $m{ms}->sumover - pdl qw( 1.5  1.9166667
>>>>>>>>
>>>>>>> 1.9166667
>>>>
>>>>> ) );
>>>>>>
>>>>>>> 174>   }
>>>>>>>>
>>>>>>>>
>>>>>>>> I am on Mac OS X Lion with Perl 5.14.1 and PDL 2.4.6
>>>>>>>> installed via
>>>>>>>>
>>>>>>> MacPorts
>>>>>>
>>>>>>> ^^^^^ |||||
>>>>>>>
>>>>>>> My guess it the problem is you are using PDL-2.4.6 and not
>>>>>>> the current PDL-2.4.9 (or even better the current CPAN
>>>>>>> Developers release version).
>>>>>>>
>>>>>>> I know there were some fixes the are in PDL-2.4.9 that
>>>>>>> helped Maggie X's stuff work.  Maybe she can recall any
>>>>>>> tricks to build her modules with a fairly old PDL (i.e.,
>>>>>>> one not having the many fixes and features added in the
>>>>>>> past couple of years).
>>>>>>>
>>>>>>> --Chris
>>>>>>>
>>>>>>>  On Dec 22, 2011, at 3:26 PM, Maggie X wrote:
>>>>>>>>
>>>>>>>>  Hi Chris,
>>>>>>>>>
>>>>>>>>> Thanks for including the info! Here's the blurb for
>>>>>>>>> PDL::Stats.
>>>>>>>>>
>>>>>>>>> -------------- This is a collection of statistics
>>>>>>>>> modules in Perl Data Language,
>>>>>>>>>
>>>>>>>> with
>>>>
>>>>> a
>>>>>>
>>>>>>>
>>>>>>>>  quick-start guide for non-PDL people.
>>>>>>>>>
>>>>>>>>> They make perldl--the simple shell for PDL--work like a
>>>>>>>>> teenie weenie
>>>>>>>>>
>>>>>>>> R,
>>>>>>
>>>>>>>
>>>>>>>>  but with PDL threading--"the fast (and automagic)
>>>>>>>>> vectorised
>>>>>>>>>
>>>>>>>> iteration
>>>>
>>>>> of
>>>>>>
>>>>>>>  'elementary operations' over arbitrary slices of
>>>>>>>>> multidimensional
>>>>>>>>>
>>>>>>>> data"--of
>>>>>>
>>>>>>>  procedures including t-test, ordinary least squares
>>>>>>>>> regression, and
>>>>>>>>>
>>>>>>>> k-means
>>>>>>
>>>>>>>  clustering. ---------------
>>>>>>>>>
>>>>>>>>> Best, Maggie
>>>>>>>>>
>>>>>>>>
_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to