[Bug 1043283] New: Review Request: shogun - Large Scale Machine Learning Toolbox

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



https://bugzilla.redhat.com/show_bug.cgi?id=1043283

            Bug ID: 1043283
           Summary: Review Request: shogun - Large Scale Machine Learning
                    Toolbox
           Product: Fedora
           Version: rawhide
         Component: Package Review
          Severity: medium
          Priority: medium
          Assignee: nobody@xxxxxxxxxxxxxxxxx
          Reporter: bjoern.esser@xxxxxxxxx
        QA Contact: extras-qa@xxxxxxxxxxxxxxxxx
                CC: package-review@xxxxxxxxxxxxxxxxxxxxxxx



Description:

  The SHOGUN machine learning toolbox's focus is on large scale kernel
  methods and especially on Support Vector Machines (SVM).  It provides
  a generic SVM object interfacing to several different SVM implementations,
  among them the state of the art LibSVM.  Each of the SVMs can be combined
  with a variety of kernels.  The toolbox not only provides efficient
  implementations of the most common kernels, like the Linear, Polynomial,
  Gaussian and Sigmoid Kernel but also comes with a number of recent string
  kernels as e.g. the Locality Improved, Fischer, TOP, Spectrum, Weighted
  Degree Kernel (with shifts).  For the latter the efficient LINADD
  optimizations are implemented.  Also SHOGUN offers the freedom of working
  with custom pre-computed kernels.  One of its key features is the "combined
  kernel" which can be constructed by a weighted linear combination of a
  number of sub-kernels, each of which not necessarily working on the same
  domain.  An optimal sub-kernel weighting can be learned using Multiple
  Kernel Learning.  Currently SVM 2-class classification and regression
  problems can be dealt with.  However SHOGUN also implements a number of
  linear methods like Linear Discriminant Analysis (LDA), Linear Programming
  Machine (LPM), (Kernel) Perceptrons and features algorithms to train hidden
  Markov-models.  The input feature-objects can be dense, sparse or strings
  and of type int/short/double/char and can be converted into different
  feature types.  Chains of "pre-processors" (e.g. subtracting the mean) can
  be attached to each feature object allowing for on-the-fly pre-processing.

  SHOGUN is implemented in C++ and offers interfaces for

    * CLI
    * Lua
    * Mono (C#)
    * Octave
    * Python
    * Ruby


Koji Builds:

  el5:  no build ---> missing dependencies
  el6:  no build ---> missing dependencies
  F18:  no build ---> will be soon EOL
  F19:  no build ---> only scratch-build for rawhide, because srpm is huge
  F20:  no build ---> only scratch-build for rawhide, because srpm is huge
  Frh:  currently uploading, will post url during today


Issues:

  fedora-review shows no obvious issues.  AFAIK there might be  some false
  positives from rpmlint.


FAS-User:

  besser82


Urls:

  Spec URL: http://besser82.fedorapeople.org/review/shogun.spec
  SRPM URL:
http://besser82.fedorapeople.org/review/shogun-3.0.0-0.1.fc20.src.rpm


Additional Information:

  SRPM is ~250 MBytes.  Build takes ~15 mins on i7-2860QM (4-core with HT =
  8 Threads).  BuildRequires are ~1 GByte.  Parallel-make will use ~1.5
  GBytes of RAM per thread.  BuildDir needs ~15 Gbytes of disk-space.


Thanks for review in advance!

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
_______________________________________________
package-review mailing list
package-review@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/package-review





[Index of Archives]     [Fedora Legacy]     [Fedora Desktop]     [Fedora SELinux]     [Yosemite News]     [KDE Users]     [Fedora Tools]