Ricardo Wurmus <[email protected]> skribis: > From 6a17d83d4bc2512f8089477e70c485e9ca93377a Mon Sep 17 00:00:00 2001 > From: Ricardo Wurmus <[email protected]> > Date: Thu, 28 May 2015 15:53:14 +0200 > Subject: [PATCH 1/2] gnu: Add libsvm. > > * gnu/packages/machine-learning.scm: New file. > * gnu-system.am (GNU_SYSTEM_MODULES): Add it.
[...] > + (synopsis "Library for Support Vector Machines") > + (description > + "LIBSVM is an integrated software for support vector s/software/piece of software/ or “library”. > +classification, (C-SVC, nu-SVC), regression (epsilon-SVR, nu-SVR) and > +distribution estimation (one-class SVM). It supports multi-class > +classification.") I think it should mention machine learning because, as ignorant of ν-SVC and ε-SVR as I am, I find it hard from the description to understand what it’s about. :-) > From 1103750f0c38c48f69f040b6a2e701f511907e26 Mon Sep 17 00:00:00 2001 > From: Ricardo Wurmus <[email protected]> > Date: Tue, 2 Jun 2015 15:10:37 +0200 > Subject: [PATCH 2/2] gnu: Add python-libsvm. > > * gnu/packages/machine-learning.scm (python-libsvm): New variable. [...] > +(define-public python-libsvm > + (package (inherit libsvm) > + (name "python-libsvm") Also add a specific synopsis like “Python bindings of libSVM”. OK with these changes. Thank you! Ludo’.
