This repo contains setup for three different docker based environments.
- Jupyter notebooks for python
- Jupyter notebooks for R
- R Studio
Run jupyter notebooks in docker.
Notebooks are saved in the host file system.
- clone the repo
- create a dir for notebooks. This needs to be under the docker host user's home dir (can be deep). e.g.
~/notebooks
- Run
NOTEBOOK_DIR=notebooks ./run.sh
- note that NOTEBOOK_DIR is specified as a relative path to home.
When the notebook is ready, open http://localhost:8888 and use the token seen in the docker run output for login.
Container mounts the host home dir at /hdir
NOTEBOOK_DIR=<full path to notebooks dir> <this repo>/R/run.sh
Run RStudio in docker with web browser as UI.
NOTEBOOK_DIR=<full path to notebooks dir> <this repo>/R/studio/run
The script builds the image (takes a long time to install all the libs - first time only), runs the container, opens Rstudio in web browser.
The notebook dir will appear as <HONE>/Documents
in the RStduio
More libraries can be added by modifying R/studio.Dockerfile
Inconsistent ways (relative vs absolute) of specifying notebook dir path (will fix later).
Not yet publishing any of the images to any docker repository.
See the corresponding Dockerfile.