In pearsonplc/pearsonverse: A package which stores all pearson-related packages
options(htmltools.dir.version = FALSE)
library(ggplot2)
library(pearsonverse)
knitr::opts_chunk$set(comment = "#>")
Agenda
--
- What is Pearsonverse?
--
- The world before Pearsonverse
--
- Project phases using Pearsonverse
--
- Useful functions in Pearsonverse
--
- Final thoughts
class: center, middle
What is Pearsonverse?
class: center, middle
Pearsonverse is a collection of in-house packages that make it easier to conduct an exploratory data science project in R from start to end.
background-image: url('intro_img/vispear_ex1.png')
background-size: 70%
background-image: url('intro_img/vispear_ex2.png')
background-size: 70%
background-image: url('intro_img/index_ex.png')
background-size: 70%
background-image: url('intro_img/report_ex.png')
background-size: 70%
class: middle, center
Final report example
class: center, middle
The world before Pearsonverse
The world before Pearsonverse
--
Everybody in our team ...
--
- creates plots in a different layout.
--
- uses different colour palettes.
--
- creates different project structures.
--
- uses different report templates.
class: center, middle
Pearsonverse solves all those problems!
Pearsonverse
--
Collection of packages:
--
- skelpear a standardised project environment that facilitates collaboration.
--
- reppear a standardised report layout that facilitates reproducibility.
--
- vispear a standardised visualisation template.
.bottom-text[
See pearsonverse documentation on GitHub.
]
Pearsonverse
All you have to do is:
# Install package
devtools::install_github("pearsonplc/pearsonverse")
# Attach package
library(pearsonverse)
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Project Phases
Project phases
- Build a project environment (skelpear)
--
- Conduct the analysis (vispear)
--
- Produce a report (reppear, vispear)
--
- Publish it (reppear)
--
Throughout the entire project: Maintain reproducibility (skelpear)
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Building a project
environment
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
--
Two arguments
--
name - Project name. Must be a valid directory name for your file system.
--
path - A path for storing your new project.
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
- After executing
project_create(), a new RStudio session will open automatically.
--
- The function initialises a
git environment.
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
--
.bottom-text[
More info about each directory in the documentation.
]
class: middle, center
Showtime
Project environment + git
To push your project to bitbucket, you need to do 2 things:
--
- Create a repository on bitbucket.
--
2. In terminal, use git remote add origin <remote_URL> to link your local project with the bitbucket repository.
--
After that, you're ready to push your commit/s.
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Conduct
the analysis
Conduct the analysis
Basic layout
.pull-three[
.small-code[
ggplot(mtcars,
aes(x = qsec,
y = mpg,
color = factor(cyl)
)
) +
geom_point(size = 5)
]]
.pull-seven[
ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) +
geom_point(size = 5)
]
Conduct the analysis
theme_enr layout
.pull-three[
.small-code[
ggplot(mtcars,
aes(x = qsec,
y = mpg,
color = factor(cyl)
)
) +
geom_point(size = 5) +
theme_enr(color = p_colors)
]]
.pull-seven[
ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) +
geom_point(size = 5) +
theme_enr(color = p_colors)
]
Conduct the analysis
Save plot
There are two functions:
save_plot() - by default, it creates .png file,save_plot2() - by default, it creates two files at once, in png and pdf format.
--
.small-code[
plot <- ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) +
geom_point(size = 5) +
theme_enr(color = p_colors)
# .png and .pdf files are created
save_plot2(name = plot,
file = "graphs/01_mtcars")
]
.bottom-text[
Both functions are compatible with the theme_enr layout.
]
class: middle, center
Showtime
Conduct the analysis
Show palettes
The function show_enr_palettes() lists all palletes available in the vispear package.
--
.small-code[
vispear::show_enr_palettes()
palette_list <- lapply(vispear::show_enr_palettes(), attributes)
data.frame(name = names(palette_list), description = unlist(unname(palette_list)))
]
Conduct the analysis
.small-code[
datatable_enr(mtcars, tooltip = "This is the description of table.")
]
.bottom-text[
Read more about the function in documentation (Epic or VPN required).
]
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Produce
the report
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
- The function creates a blank report based on the template.
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
Two arguments
--
file - Report name. Must be a valid directory name for your file system.
--
subdir - Subdirectory where you want to store your new report.
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
01_report - a directory which stores a report and all its dependendies. The report itself has a consistent layout.
--
- index.Rmd - a file which acts as a home page for all reports within the project*.
.bottom-text[
*More info about the file in documentation.
]
Produce the report
To render index.Rmd,
reppear::render_index()
--
It re/builds index.html, tooltip_content.html and header.html files.
class: middle, center
Showtime
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Publish it
Publish it
reppear::publish(file = "reports/")
.bottom-text[
For now only kvm1-e01.ioki.pl server is defined.
]
class: center middle no-page-number
background-image: url('img/divider_blue.jpg')
background-size: 100%
Maintain
reproducibility
Maintain reproducibility
--
# Save your package environment
snapshot_pkg()
# Compare package environment
compare_snapshot()
--
These two functions allow you to save and compare the packages used during the project. Use this if you want to make sure that others with whom you share the codebase have the same setup*.
.bottom-text[
*More info about how the functions work in documentation.
]
class: center, middle
Final thoughts
Goals
--
- present Pearsonverse to the broader audience
--
- propose a shared github repository for storing and publishing useful R code produced by ACDSL, so that the R users in ACDSL can work faster using standardised tools…
--
- … and so that we start giving back to the open-source community
Testers wanted
--
- Pearsonverse was only tested on macOS. We plan to test it on Linux and Windows soon.
--
- Testers wanted!
class: middle, center
Ideas, bugs?
--
Please write an issue on https://github.com/pearsonplc/pearsonverse/issues
class: middle, center
Questions?
class: no-page-number
background-image: url('img/final.jpg')
background-size: 100%
pearsonplc/pearsonverse documentation built on May 30, 2019, 3:45 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.
options(htmltools.dir.version = FALSE) library(ggplot2) library(pearsonverse) knitr::opts_chunk$set(comment = "#>")
Agenda
--
- What is Pearsonverse?
--
- The world before Pearsonverse
--
- Project phases using Pearsonverse
--
- Useful functions in Pearsonverse
--
- Final thoughts
class: center, middle
What is Pearsonverse?
class: center, middle
Pearsonverse is a collection of in-house packages that make it easier to conduct an exploratory data science project in R from start to end.
background-image: url('intro_img/vispear_ex1.png') background-size: 70%
background-image: url('intro_img/vispear_ex2.png') background-size: 70%
background-image: url('intro_img/index_ex.png') background-size: 70%
background-image: url('intro_img/report_ex.png') background-size: 70%
class: middle, center
Final report example
class: center, middle
The world before Pearsonverse
The world before Pearsonverse
--
Everybody in our team ...
--
- creates plots in a different layout.
--
- uses different colour palettes.
--
- creates different project structures.
--
- uses different report templates.
class: center, middle
Pearsonverse solves all those problems!
Pearsonverse
--
Collection of packages:
--
- skelpear a standardised project environment that facilitates collaboration.
--
- reppear a standardised report layout that facilitates reproducibility.
--
- vispear a standardised visualisation template.
.bottom-text[ See pearsonverse documentation on GitHub. ]
Pearsonverse
All you have to do is:
# Install package devtools::install_github("pearsonplc/pearsonverse") # Attach package library(pearsonverse)
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Project Phases
Project phases
- Build a project environment (skelpear)
--
- Conduct the analysis (vispear)
--
- Produce a report (reppear, vispear)
--
- Publish it (reppear)
--
Throughout the entire project: Maintain reproducibility (skelpear)
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Building a project
environment
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
--
Two arguments
--
name - Project name. Must be a valid directory name for your file system.
--
path - A path for storing your new project.
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
- After executing
project_create(), a new RStudio session will open automatically.
--
- The function initialises a
gitenvironment.
Project environment
.small-code[
skelpear::project_create(name = "project", path = ".")
]
--
.bottom-text[ More info about each directory in the documentation. ]
class: middle, center
Showtime
Project environment + git
To push your project to bitbucket, you need to do 2 things:
--
- Create a repository on bitbucket.
--
2. In terminal, use git remote add origin <remote_URL> to link your local project with the bitbucket repository.
--
After that, you're ready to push your commit/s.
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Conduct
the analysis
Conduct the analysis
Basic layout .pull-three[ .small-code[
ggplot(mtcars,
aes(x = qsec,
y = mpg,
color = factor(cyl)
)
) +
geom_point(size = 5)
]]
.pull-seven[
ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) + geom_point(size = 5)
]
Conduct the analysis
theme_enr layout .pull-three[ .small-code[
ggplot(mtcars,
aes(x = qsec,
y = mpg,
color = factor(cyl)
)
) +
geom_point(size = 5) +
theme_enr(color = p_colors)
]]
.pull-seven[
ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) + geom_point(size = 5) + theme_enr(color = p_colors)
]
Conduct the analysis
Save plot
There are two functions:
save_plot()- by default, it creates.pngfile,save_plot2()- by default, it creates two files at once, inpngandpdfformat.
--
.small-code[
plot <- ggplot(mtcars, aes(qsec, mpg, color = factor(cyl))) + geom_point(size = 5) + theme_enr(color = p_colors) # .png and .pdf files are created save_plot2(name = plot, file = "graphs/01_mtcars")
]
.bottom-text[
Both functions are compatible with the theme_enr layout.
]
class: middle, center
Showtime
Conduct the analysis
Show palettes
The function show_enr_palettes() lists all palletes available in the vispear package.
--
.small-code[
vispear::show_enr_palettes()
palette_list <- lapply(vispear::show_enr_palettes(), attributes) data.frame(name = names(palette_list), description = unlist(unname(palette_list)))
]
Conduct the analysis
.small-code[
datatable_enr(mtcars, tooltip = "This is the description of table.")
]
.bottom-text[ Read more about the function in documentation (Epic or VPN required). ]
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Produce
the report
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
- The function creates a blank report based on the template.
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
Two arguments
--
file - Report name. Must be a valid directory name for your file system.
--
subdir - Subdirectory where you want to store your new report.
Produce the report
reppear::report_create(file = "01_report", subdir = "reports")
--
01_report- a directory which stores a report and all its dependendies. The report itself has a consistent layout.
--
- index.Rmd - a file which acts as a home page for all reports within the project*.
.bottom-text[ *More info about the file in documentation. ]
Produce the report
To render index.Rmd,
reppear::render_index()
--
It re/builds index.html, tooltip_content.html and header.html files.
class: middle, center
Showtime
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Publish it
Publish it
reppear::publish(file = "reports/")
.bottom-text[ For now only kvm1-e01.ioki.pl server is defined. ]
class: center middle no-page-number background-image: url('img/divider_blue.jpg') background-size: 100%
Maintain
reproducibility
Maintain reproducibility
--
# Save your package environment snapshot_pkg() # Compare package environment compare_snapshot()
--
These two functions allow you to save and compare the packages used during the project. Use this if you want to make sure that others with whom you share the codebase have the same setup*.
.bottom-text[ *More info about how the functions work in documentation. ]
class: center, middle
Final thoughts
Goals
--
- present Pearsonverse to the broader audience
--
- propose a shared github repository for storing and publishing useful R code produced by ACDSL, so that the R users in ACDSL can work faster using standardised tools…
--
- … and so that we start giving back to the open-source community
Testers wanted
--
- Pearsonverse was only tested on macOS. We plan to test it on Linux and Windows soon.
--
- Testers wanted!
class: middle, center
Ideas, bugs?
--
Please write an issue on https://github.com/pearsonplc/pearsonverse/issues
class: middle, center
Questions?
class: no-page-number background-image: url('img/final.jpg') background-size: 100%
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.