R Package for simultaneously modelling multiple facets of biodiversity (α- and β-diversity, taxonomic, functional, and phylogenetic).
Before installing MBM, it is necessary to have installations of R, Python and GPy. Please install these packages and verify they are working before installing MBM. New users and those unfamiliar with Python should see the "Recommended Installation Procedure" below.
Additionally, some functionality (namely, svgp
for fitting large models) requires the climin
package in Python. Thus, it is highly recommended to install this package when instally GPy. Note that climin pre-0.1
is the only supported version. Thus it is necessary to follow these instructions to install climin from Github.
Within R, MBM can be installed directly from github using the devtools
package.
# uncomment the line below if devtools is not yet installed
# install.packages('devtools')
library('devtools')
install_github('mtalluto/mbm')
library('mbm')
?mbm
Alternatively, if you would like additional features (principally, vignettes), you can clone or download the project and build and install it (from the command line):
git clone https://github.com/mtalluto/mbm.git
Rscript -e "devtools::build('mbm')"
R CMD INSTALL mbm_i.j.k.tar.gz
Replacing the i.j.k
with the version number you have downloaded; this file will have been created by devtools in the second step.
pip install GPy
. If you have errors, you can check the linked page for installation instructions.devtools
and mbm
within R; follow the instructions above under Installing MBM.library("mbm")
then run check_python()
. If all works, you can proceed with the examples.The following procedure has been tested on a clean mac running macos 10.13 using the system Python (version 2.7). From the command line:
sudo easy_install pip
sudo pip install -U --ignore-installed numpy scipy
sudo pip install GPy tornado climin
From there, assuming everything proceeds without errors, you can install and test MBM.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.