In D-Se/ML: Machine Leanring targets-bookdown pipeline
```{css, echo=FALSE}
body .main-container {
max-width: 1280px !important;
width: 1280px !important;
}
body {
max-width: 1280px !important;
}
h1 {
text-align: center;
}
```r
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>"
)
Welcome!
Hi there! Glad to have you on board. This is a guide to help you help us. If anything is unclear, questions can be asked by posting on Github Discussions, or by sending an email to r paste0(gsub("[<>]", "", strsplit(maintainer("ML"), "<")[[1]] |> trimws()), collapse = " at: ").
Getting started
-
Install project tools and dependencies by running devtools::install_github("D-Se/ML") in the R console. This might require you to install tensorflow separately.
-
Connect RStudio to github (guide).
-
Hop to the dev branch of the github repository.
-
Find issues to solve (picked by you or assigned to you).
About the project
Tools
Languages
- R for most of its computations, C++ where speed is important.
- HTML, CSS & JavaScript for making stylish documents.
- YAML is used in rmarkdown for metadata description.
R packages
This is a (partial) list of packages used in the project at various stages:
Reporting packages
+--------------+---------------------------+-----------------------------------------------+
| Package | Description | Resources |
+==============+===========================+===============================================+
| ggplot2 | Data visualization | |
+--------------+---------------------------+-----------------------------------------------+
| plotly | Interactive plots | |
+--------------+---------------------------+-----------------------------------------------+
| visNetwork | Ntwork graphs | |
+--------------+---------------------------+-----------------------------------------------+
| rmarkdown | Markdown with an R accent | Book |
+--------------+---------------------------+-----------------------------------------------+
| bookdown | | |
+--------------+---------------------------+-----------------------------------------------+
| shiny | Web application framework | |
+--------------+---------------------------+-----------------------------------------------+
Modelling packages
+-------------+------------------------------------------+------------------------------------------------------------------------+
| Package | Description | Resources |
+=============+==========================================+========================================================================+
| keras | APi to tensorflow neural nets | keras ; TensorFlow |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| rpart | Regression trees | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| glmnet | Penalized maximum likelihood models | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| earth | Multivariate Adaptive Regression Splines | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| rsample | Data sampling | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| recipes | Pre-processor specification | Book |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| parsnip | Model specification | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| tune | Hyper parameter tuning | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| workflows | | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
Package development
+---------------+--------------------------------+----------------------------------------------------------------------+
| Package | Description | Resources |
+===============+================================+======================================================================+
| targets | Make-like pipeline toolkit | Book |
+---------------+--------------------------------+----------------------------------------------------------------------+
| tarchetypes | Function-oriented targets | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| rlang | Tidyverse programming toolbox | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| testthat | Unit testing | Book |
+---------------+--------------------------------+----------------------------------------------------------------------+
| lifecycle | function life cycle management | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| future | Parallel processing framework | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| promises | Asynchronous programming | website |
+---------------+--------------------------------+----------------------------------------------------------------------+
| crayon | Pretty console printing | |
+---------------+--------------------------------+----------------------------------------------------------------------+
Github repo
The Github repo is where the project lives. In it, you can find the following files. This may change in future releases.
General
├─_targets {all objects and metadata of the targets pipeline.}
├─_targets.R {Project workflow file.}
├─.github {Folder for Github workflows, code that is run everytime a new commit is made to the repo.}
├── check-standard.yml {runs R CMD check to see if R package can be installed on Windows, OS and Linux.}
├── lint-project.yml {Static code analysis, checks language in code. See the lintr package}
├─.gitignore {Ignored files. Files stored on your computer that will not be sent to Github repository.}
├─ ML.Rproj {File that can be opened with RStudio, tells the setup.}
Book-related
├─ _book
├── ... {Book output files created by tar_make(). It includes copies of other files. Click $\color{green}{\text{ index.html}}$ to view it.}
├─_bookdown.yml {specifies the order of files in the book.}
├─_chapters {The rmarkdown files, book chapters. report_xx is dynamically generated, its contents may change.}
├─_common.R {R functions that are run before each chapter of the book.}
├─_output.yml {Options for creating different output book and formats.}
├─ js {JavaScript files to make the book look good.}
├─ assets {Folder for images. and CSS styling used in markdown files.}
├─ inst {Extra book materials. Insert any references into the xx.bib files.}
├─ index.Rmd {The first file in the book, }
Package-related
An R package requires many different files in specific locations, for historical reasons.
├─ DESCRIPTION {ML package metadata.}
├─ NAMESPACE {Control which functions are available to users who use library("ML")}
├─ NEWS.md {Historical news of version releases}
├─ README.Rmd {R markdown-enabled package introduction. Can run R code within.}
├─ README.md {Package introduction for Github first page introduction.}
├─ R {All R code used in the ML package. Cannot contain subfolders.}
├── data.R {Data sets available, created from the _targets directory. Edit with usethis::use_data(object, overwrite = F).}
├── ML-package.R {Package introduction used for help files.}
├── modelling_x.R {Code to make models of a certain type.}
├── reporting_x.R {Helper functions to make good looking reports and book.}
├── utils.R {Helper functions that are useful in multiple other .R files. Never exported.}
├── visualization.R {Functions specifically for plotting and graphs.}
├─ man {Directly for help files, automatically generated by devtools::document() based on the roxygen2 comments that look like #' @something blabla.}
├─ src {C++ functions.}
├─ tests {Folder for unit tests to check if functions in ML are working properly.}
├── vignettes {Long-form documentation available to users who run library(ML).}
The project is split up into segments, each ending with finishing a deliverable. These steps can be found in the milestones section of the repo. The milestones are made up of collections of issues. These are tasks assigned to project members, and can be added by any person with access rights to the repo. labels are placed on issues to make it easy to find and assign tasks. Each issue is allocated a difficulty and priority label, and further labels were appropriate.
External Resources - learning & finding help
R resources
Visualization
Modelling
Reporting
Productivity tips
A list of useful version controlled R project commands:
-
devtools::document() to update (locally) the package using roxygen2.
-
lintr::lint_dir() to use static code analysis for code formatting checks.
lintr is useful to check code style line-by-line and make adjustments on the fly. It will open an interface in the RStudio Markers pane where individual lines can be clicked. On a click, the relevant code snippet will be opened in the editor. This command is also run automatically on commits to the master branch of Github repo, viewable as comments in the Github Actions pane.
-
covr::report() for visual inspection of unit test coverage.
-
usethis::edit_r_profile() to run functions on startup.
An example of Donald's ML-project specific R profile line: runs library(devtools) and imports specific targets functions as an alias for brevity.
if (tryCatch(endsWith(readLines("DESCRIPTION", 1), "ML"), warning = function(x) F)) {
#library(devtools, quietly = T)
cat("loading devtools & select targets functions\n")
box::use(
devtools[...],
targets[make = tar_make,
make_future = tar_make_future,
grab = tar_read]
)
}
D-Se/ML documentation built on April 1, 2022, 10:53 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
```{css, echo=FALSE} body .main-container { max-width: 1280px !important; width: 1280px !important; } body { max-width: 1280px !important; }
h1 { text-align: center; }
```r knitr::opts_chunk$set( collapse = TRUE, comment = "#>" )
Welcome!
Hi there! Glad to have you on board. This is a guide to help you help us. If anything is unclear, questions can be asked by posting on Github Discussions, or by sending an email to r paste0(gsub("[<>]", "", strsplit(maintainer("ML"), "<")[[1]] |> trimws()), collapse = " at: ").
Getting started
-
Install project tools and dependencies by running
devtools::install_github("D-Se/ML")in the R console. This might require you to installtensorflowseparately. -
Connect RStudio to github (guide).
-
Hop to the dev branch of the github repository.
-
Find issues to solve (picked by you or assigned to you).
About the project
Tools
Languages
- R for most of its computations, C++ where speed is important.
- HTML, CSS & JavaScript for making stylish documents.
- YAML is used in rmarkdown for metadata description.
R packages
This is a (partial) list of packages used in the project at various stages:
Reporting packages
+--------------+---------------------------+-----------------------------------------------+
| Package | Description | Resources |
+==============+===========================+===============================================+
| ggplot2 | Data visualization | |
+--------------+---------------------------+-----------------------------------------------+
| plotly | Interactive plots | |
+--------------+---------------------------+-----------------------------------------------+
| visNetwork | Ntwork graphs | |
+--------------+---------------------------+-----------------------------------------------+
| rmarkdown | Markdown with an R accent | Book |
+--------------+---------------------------+-----------------------------------------------+
| bookdown | | |
+--------------+---------------------------+-----------------------------------------------+
| shiny | Web application framework | |
+--------------+---------------------------+-----------------------------------------------+
Modelling packages
+-------------+------------------------------------------+------------------------------------------------------------------------+
| Package | Description | Resources |
+=============+==========================================+========================================================================+
| keras | APi to tensorflow neural nets | keras ; TensorFlow |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| rpart | Regression trees | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| glmnet | Penalized maximum likelihood models | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| earth | Multivariate Adaptive Regression Splines | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| rsample | Data sampling | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| recipes | Pre-processor specification | Book |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| parsnip | Model specification | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| tune | Hyper parameter tuning | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
| workflows | | |
+-------------+------------------------------------------+------------------------------------------------------------------------+
Package development
+---------------+--------------------------------+----------------------------------------------------------------------+
| Package | Description | Resources |
+===============+================================+======================================================================+
| targets | Make-like pipeline toolkit | Book |
+---------------+--------------------------------+----------------------------------------------------------------------+
| tarchetypes | Function-oriented targets | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| rlang | Tidyverse programming toolbox | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| testthat | Unit testing | Book |
+---------------+--------------------------------+----------------------------------------------------------------------+
| lifecycle | function life cycle management | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| future | Parallel processing framework | |
+---------------+--------------------------------+----------------------------------------------------------------------+
| promises | Asynchronous programming | website |
+---------------+--------------------------------+----------------------------------------------------------------------+
| crayon | Pretty console printing | |
+---------------+--------------------------------+----------------------------------------------------------------------+
Github repo
The Github repo is where the project lives. In it, you can find the following files. This may change in future releases.
General
├─_targets {all objects and metadata of the targets pipeline.}
├─_targets.R {Project workflow file.}
├─.github {Folder for Github workflows, code that is run everytime a new commit is made to the repo.}
├── check-standard.yml {runs R CMD check to see if R package can be installed on Windows, OS and Linux.}
├── lint-project.yml {Static code analysis, checks language in code. See the lintr package}
├─.gitignore {Ignored files. Files stored on your computer that will not be sent to Github repository.}
├─ ML.Rproj {File that can be opened with RStudio, tells the setup.}
Book-related
├─ _book
├── ... {Book output files created by tar_make(). It includes copies of other files. Click $\color{green}{\text{ index.html}}$ to view it.}
├─_bookdown.yml {specifies the order of files in the book.}
├─_chapters {The rmarkdown files, book chapters. report_xx is dynamically generated, its contents may change.}
├─_common.R {R functions that are run before each chapter of the book.}
├─_output.yml {Options for creating different output book and formats.}
├─ js {JavaScript files to make the book look good.}
├─ assets {Folder for images. and CSS styling used in markdown files.}
├─ inst {Extra book materials. Insert any references into the xx.bib files.}
├─ index.Rmd {The first file in the book, }
Package-related
An R package requires many different files in specific locations, for historical reasons.
├─ DESCRIPTION {ML package metadata.}
├─ NAMESPACE {Control which functions are available to users who use library("ML")}
├─ NEWS.md {Historical news of version releases}
├─ README.Rmd {R markdown-enabled package introduction. Can run R code within.}
├─ README.md {Package introduction for Github first page introduction.}
├─ R {All R code used in the ML package. Cannot contain subfolders.}
├── data.R {Data sets available, created from the _targets directory. Edit with usethis::use_data(object, overwrite = F).}
├── ML-package.R {Package introduction used for help files.}
├── modelling_x.R {Code to make models of a certain type.}
├── reporting_x.R {Helper functions to make good looking reports and book.}
├── utils.R {Helper functions that are useful in multiple other .R files. Never exported.}
├── visualization.R {Functions specifically for plotting and graphs.}
├─ man {Directly for help files, automatically generated by devtools::document() based on the roxygen2 comments that look like #' @something blabla.}
├─ src {C++ functions.}
├─ tests {Folder for unit tests to check if functions in ML are working properly.}
├── vignettes {Long-form documentation available to users who run library(ML).}
The project is split up into segments, each ending with finishing a deliverable. These steps can be found in the milestones section of the repo. The milestones are made up of collections of issues. These are tasks assigned to project members, and can be added by any person with access rights to the repo. labels are placed on issues to make it easy to find and assign tasks. Each issue is allocated a difficulty and priority label, and further labels were appropriate.
External Resources - learning & finding help
R resources
Visualization
Modelling
Reporting
Productivity tips
A list of useful version controlled R project commands:
-
devtools::document()to update (locally) the package usingroxygen2. -
lintr::lint_dir()to use static code analysis for code formatting checks.
lintr is useful to check code style line-by-line and make adjustments on the fly. It will open an interface in the RStudio Markers pane where individual lines can be clicked. On a click, the relevant code snippet will be opened in the editor. This command is also run automatically on commits to the master branch of Github repo, viewable as comments in the Github Actions pane.
-
covr::report()for visual inspection of unit test coverage. -
usethis::edit_r_profile()to run functions on startup.
An example of Donald's ML-project specific R profile line: runs library(devtools) and imports specific targets functions as an alias for brevity.
if (tryCatch(endsWith(readLines("DESCRIPTION", 1), "ML"), warning = function(x) F)) { #library(devtools, quietly = T) cat("loading devtools & select targets functions\n") box::use( devtools[...], targets[make = tar_make, make_future = tar_make_future, grab = tar_read] ) }
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.