sofia_taf_tutorial.md

SOFIA-TAF Tutorial

How to run

First install the main packages (sraplus, TAF, SOFIA):

library(remotes)
install_github("DanOvando/sraplus")
install.packages("TAF")
install_github("sofia-taf/SOFIA")

The sraplus package has many underlying package dependencies that should install automatically.

Then clone/download a SOFIA analysis, for example:

https://github.com/sofia-taf/2022Area37Demo

Open R in the 2022Area37Demo directory and run:

library(TAF)
taf.boot()
source.all()

The source.all function runs the TAF scripts sequentially in alphabetical order:

data.R
model.R
output.R
report.R

Explore results

The results from each script appear in the corresponding working subdirectory. For example, the data.R script writes data tables and plots into a folder called data. The results from the analysis are found in the model folder (sraplus results object), output folder (detailed tables) and report folder (plots and summary tables).

Back to basics

For a minimal example of TAF, a simple linear regression (LM) demo is available at:

https://github.com/PacificCommunity/ofp-sam-taf-demo

This analysis has no software dependencies and can be useful for studying the mechanics and features of TAF analyses, without the additional complexity of the sraplus model.

Learn more

The sofia-taf/doc page gives an overview of SOFIA-TAF.

https://github.com/ices-taf/doc gives an overview of TAF.

In an R session, the package help page can be brought up with the command ?TAF.

TAF discussion points

See taf_discussion.md.