Re: [PATCH] gnu: Add r-glmnet.

[Ricardo Wurmus <ricardo.wurmus@mdc-berlin.de>]

Ben Woodcroft <donttrustben@gmail.com> writes:

> * gnu/packages/statistics.scm (r-glmnet): New variable.
> ---
>  gnu/packages/statistics.scm | 26 ++++++++++++++++++++++++++
>  1 file changed, 26 insertions(+)
>
> diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
> index f037392..732d874 100644
> --- a/gnu/packages/statistics.scm
> +++ b/gnu/packages/statistics.scm
> @@ -2244,6 +2244,32 @@ singular and eigenvalue decompositions, as well as for principal component
>  analysis of large sparse or dense matrices.")
>      (license (list license:gpl2+ license:gpl3+))))
>  
> +(define-public r-glmnet
> +  (package
> +   (name "r-glmnet")
> +   (version "2.0-5")
> +   (source
> +    (origin
> +     (method url-fetch)
> +     (uri (cran-uri "glmnet" version))
> +     (sha256
> +      (base32
> +       "1cbpzmbv837fvq88rgn6mgzgr9f1wqp9fg8gh2kkmngvr1957a9c"))))
> +   (build-system r-build-system)
> +    (inputs
> +     `(("gfortran" ,gfortran)))
> +   (propagated-inputs
> +    `(("r-foreach" ,r-foreach)))
> +   (home-page "http://www.jstatsoft.org/v33/i01")
> +   (synopsis "Lasso and Elastic-Net Regularized Generalized Linear Models")

Does this need to be all caps?

> +   (description
> +    "The glmnet package provides efficient procedures for fitting the entire
> +lasso or elastic-net regularization path for linear and Poisson regression, as
> +well as logistic, multinomial, Cox, multiple-response Gaussian and grouped
> +multinomial models.  The algorithm uses cyclical coordinate descent in a
> +path-wise fashion.")
> +   (license license:gpl2+)))

The rest looks good to me.  Thanks!

~~ Ricardo