Dear Colleagues, Despite my hesitation with what name to choose (I would like it to be "Statistical Learning", but that would have excluded some other "learning" strategies I guess), I've decided to initiate "Machine Learning" task within science blend. I wanted to check with you first, either it looks reasonable/desired before I commit it. Please find a tentative task file attached. -- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ (yoh@|www.)onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555]
Task: Machine Learning Description: Debian Science Machine Learning packages This metapackage will install Debian packages which might be useful for scientists interested in machine learning. Included packages range from knowledge-based (expert) inference systems to software implementing dominant nowadays statistical methods. Depends: python-pyke, gprolog, yap Comments: Prolog (and alike) systems for inductive reasoning Depends: libtorch3-dev Depends: libshogun-dev, libfann-dev, libsvm-dev, libcomplearn-dev Comment: above libraries have also variety of interfaces to high-level scripting languages (e.g. Python) and even possibly some interactive GUI Depends: python-scikits-learn, python-mdp, python-mlpy Comment: Native Python toolkits Depends: weka Depends: vowpal-wabbit Depends: r-cran-mass, r-cran-bayesm, r-cran-class, r-cran-cluster, r-cran-msm Comment: R packages Depends: python-mvpa Why: Framework for statistical learning analysis of large datasets. Published-Title: PyMVPA: a unifying approach to the analysis of neuroscientific data Published-Authors: Michael Hanke, Yaroslav O. Halchenko, Per B. Sederberg, Emanuele Olivetti, Ingo Fründ, Jochem W. Rieger, Christoph S. Herrmann, James V. Haxby, Stephen José Hanson, Stefan Pollmann Published-In: Frontiers in Neuroinformatics, 3:3 Published-Year: 2009 Published-URL: http://www.frontiersin.org/neuroinformatics/paper/10.3389/neuro.11/003.2009/ Published-DOI: 10.3389/neuro.11.003.2009 Depends: python-scikits-statsmodels Homepage: http://statsmodels.sourceforge.net/ Language: Python License: BSD Responsible: Yaroslav Halchenko <debian@onerussian.com> Pkg-URL: http://neuro.debian.net/pkgs/python-scikits-statsmodels.html WNPP: 570604 Why: Statistical models Description: classes and functions for the estimation of statistical models scikits.statsmodels is a pure Python package that provides classes and functions for the estimation of several categories of statistical models. These currently include linear regression models, OLS, GLS, WLS and GLS with AR(p) errors, generalized linear models for six distribution families and M-estimators for robust linear models. An extensive list of result statistics are available for each estimation problem. Suggests: science-statistics Suggests: science-typesetting Meta-Suggests: svn://svn.debian.org/blends/projects/science/trunk/debian-science/tasks/typesetting
Attachment:
signature.asc
Description: Digital signature