fastFMM: Fast Functional Mixed Models using Fast Univariate Inference (FUI)

Repository Description

Repository for the development version of the R Package fastFMM. For more information, see the official fastFMM \(\texttt{CRAN}\) site.

fastFMM R Package

Installation

Download the \(\texttt{R}\) Package fastFMM by running the following command within \(\texttt{R}\) or \(\texttt{RStudio}\):

install.packages("fastFMM", dependencies = TRUE)

Alternatively, the development version of the \(\texttt{R}\) Package fastFMM can be downloaded as follows:

library(devtools)
install_github("gloewing/fastFMM")

Package Usage

Repository Folders

  1. The ‘R’ folder contains the code of the package, including fui.R and plot_fui.R. The plot_fui.R is still under development and has not been widely tested.

  2. The ‘vignettes’ folder contains a vignette which shows how to use different arguments of the fui function. This vignette can also be viewed in the link above (under Package Usage).


The example data set is available in the ‘vignettes’ folder under the name ‘time_series.csv’.

Calling fastFMM from Python

See ‘python_fastFMM_vignette.py’ in the Github repo for a brief example of using fastFMM on Python through the Python package rpy2. We are working on more documentation. The tutorial assumes the fastFMM R package (and all its dependenices), and the rpy2 Python package have already been installed. Even if you intend to use the package purely within Python, it may be helpful to first install fastFMM in RStudio to ensure all package dependenices are installed automatically.