|tidyviz||Adds reporting & interactive charts packages|
base is jvera/tidyviz-base
- report specific (rmdformats, tufte, prettydoc, rticles, printR...)
- visualization (ggvis, leaflet, ggThemeAssist, ggthemes...)
- imputation and assertions (mice, validate, VIM, assertive, tester...)
- interactive graphics (highcharter, plotly, DiagrammeR...)
- text analytics (tidytext, textclean, wordcloud, topicmodels...)
- machine learning (randomForest, caret, party, e1071, gbm ...)
Suitable for sharing ·even· prettier analysis and reproducibility.
Pull the image from repo:
docker pull jvera/tidyviz:latest
Then build a container:
docker run -d -p 8787:8787 jvera/tidyviz
Connect to port 8787 with your browser (rstudio as user and password) to check if your Rserver is up and running.
By default, the RStudio user does not have access to root, such that users cannot install binary libraries with apt-get without first entering the container. To enable root from within RStudio, launch the container with the flag -e ROOT=TRUE, e.g.
docker run -d -p 8787:8787 -e ROOT=TRUE jvera/tidyviz
You can now open a shell from RStudio (see the "Tools" menu), or directly from the R console using system(), e.g.
system("sudo apt-get install -y vim")
When container is built you can start it with:
docker start mycontainer --interactive /bin/bash
Sharing data with host:
docker start mycontainer --interactive -v ~/dockerdata:/data /bin/bash
Any file you place there, will be available for the container to use, and vice versa. Maybe you need a Shiny Server, so run a Dockerized Shiny and share the same folder so you can develop your viz in Rstudio and serve with Shiny.