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