commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sébastien Brisard (Commented) (JIRA) <j...@apache.org>
Subject [jira] [Commented] (MATH-732) Major Speed Improvement to 1D Discrete Fourier Transform (approximately 5x-9x improvement). Preserved public API 100%. Removed unnecessary use of instance variables and instance state.
Date Thu, 09 Feb 2012 07:39:00 GMT

    [ https://issues.apache.org/jira/browse/MATH-732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13204338#comment-13204338
] 

Sébastien Brisard commented on MATH-732:
----------------------------------------

Thanks Kurt, I took item #2 and #3 into account in {{r1242230}}. As for item #1, here is what
I've done
# Created {{enum FastFourierTransformer.Normalization}} which can be {{STANDARD}} or {{UNITARY}}.
# The type of normalization is then specified in the constructor of {{FastFourierTransformer}}
as well as the static method {{transformInPlace}}.
# Removed the factory methods {{create()}} and {{createUnitary()}}.

I would like to suggest one last change: I do not like boolean parameters (in public methods),
as they are difficult to interpret. I see two options
# Define an enum {FORWARD, INVERSE}, and replace the boolean parameter in {{transformInPlace}}
with this enum,
# Create two methods {{transformInPlace}} and {{inverseTransformInPlace}}. This would be consistent
with class methods.

The first option would be my preferred option, but then I suppose we should also change the
class methods, and replace {{transform}} and {{inverseTransform}} with one single method taking
FORWARD or INVERSE as an argument.
                
> Major Speed Improvement to 1D Discrete Fourier Transform (approximately 5x-9x improvement).
Preserved public API 100%. Removed unnecessary use of instance variables and instance state.
> ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MATH-732
>                 URL: https://issues.apache.org/jira/browse/MATH-732
>             Project: Commons Math
>          Issue Type: Improvement
>    Affects Versions: 3.0
>            Reporter: Kurt Ostfeld
>            Assignee: Sébastien Brisard
>              Labels: api, perfomance
>             Fix For: 3.0
>
>         Attachments: DFTPerformanceWithPatch.png, DFTPerformanceWithoutPatch.png, FastFourierTransformer.java,
FastFourierTransformerTest.java, Main.java
>
>
> I wrote my own Discrete Fourier Transform function in Java and ran some benchmarks and
found that it ran dramatically faster than the Apache library implementation. This is a pretty
straight forward implementation of the standard Cooley Tukey algorithm that I read out of
a textbook. This passes all the Apache library test cases plus several I had written on my
own. I created a source code library patch that preserves the public API completely while
changing the internal implementation to achieve the performance improvement.
> In addition to the performance issue, I suggest that Discrete Fourier Transform functionality
be provided as stateless pure functions (in Java this would be implemented with static methods)
just like most other math functions. As-is, the library requires the user to instantiate a
Transformer instance which maintains instance state, which is an unecessary complexity for
a pure math function. I held off on this change since it would change the public API and affect
existing code. However, I see similar backward compatability breaking API changes are already
in the FastFourierTransformer class in the 3.0 code base.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

       

Mime
View raw message