Nothing
knitr::opts_chunk$set( collapse = TRUE, comment = "#>" )
multitool
uses a set of argument naming conventions to help users.
Whereas multitool function names are meant to describe what they do, the first argument of each function is meant to help users understand which multitool objects should be passed to a particular function.
I designed first argument names to follow the steps for setting up, building, running, unpacking, and summarizing a multiverse-style analysis. They are as follows:
.df
= the original data set .pipeline
= a data analysis pipeline blueprint produced by calling a series of add_* functions .grid
= a data analysis pipeline fully expanded into all possible combinations of decisions outlined by the .pipeline
blueprint.multi
= a set of results produced by running an analysis across all rows of a .grid
.unpacked
= a specific set of results that have been unnested for plotting or summarizing.condensed
= a specific set of results that have been summarizedThese conventions make more sense when visualized across a typical multitool
workflow:
DiagrammeR::grViz(" digraph boxes_and_circles { # a 'graph' statement graph [overlap = true, fontsize = 10] # several 'node' statements node [shape = box, fill = gray, fontname = Helvetica] 1 [label = 'base data'] 2 [label = 'add_*() functions'] 3 [label = 'expand_decisions()'] 4 [label = 'create_blueprint_graph()'] 5 [label = 'run_descriptives()'] 6 [label = 'show_code_*()'] 7 [label = 'run_multiverse()'] 8 [label = 'reveal(), reveal_model_*()'] 9 [label = 'condense()'] 10 [label = 'reveal_corrs(), reveal_reliabilities(), reveal_summary_stats()'] 11 [label = 'column_plot(), custom plotting/tables'] 12 [label = 'detect_*(), summarize_filter_ns()'] # several 'edge' statements 1->2 [label = '.df'] 2->3 [label = '.pipeline'] 2->4 [label = '.pipeline'] 2->5 [label = '.pipeline'] 3->6 [label = '.grid'] 3->7 [label = '.grid'] 7->8 [label = '.multi'] 8->9 [label = '.unpacked'] 5->10 [label = '.descriptives'] 10->9 [label = '.unpacked'] 9->11 [label = '.condensed'] 2->12 [label = '.pipeline'] } ")
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.