Skip to content

Latest commit

 

History

History
167 lines (118 loc) · 7.04 KB

README.md

File metadata and controls

167 lines (118 loc) · 7.04 KB
Raw

nlpsandbox.io

NLP Sandbox Notebooks

GitHub Release GitHub CI GitHub License Docker Pulls Discord

Introduction

This repository provides a portable development environment that includes RStudio and Jupyter. This environment aims to support NLP Sandbox users who develop or apply tools that intereact with the NLP Sandbox ecosystem.

For example, try the example notebooks included with this repository to perform analysis using data stored in a local or remote instance of the NLP Sandbox Data Node. See the Section Notebooks for the list of example notebooks included with this repository.

The Docker image nlpsandbox/notebooks that comes with this project is based on the image sagebionetworks/rstudio.

Requirements

Specification

  • Notebooks version: 0.2.0
  • NLP Sandbox schemas version: 1.2.0
  • RStudio version: 4.1.0
  • JupyterLab version: 3.2.1

Notebooks

Example notebooks for RStudio:

Notebook Description
notebook.Rmd Default RStudio notebook.
r-and-python.Rmd Shows how to use R and Python together.
data-node.Rmd Interact with an NLP Sandbox Data Node instance.

Example notebooks for Jupiter:

Notebook Description
hello-world.ipynb Hello World notebook.

Important: Please make sure when you write your own notebooks that no sensitive information ends up being publicly available. Please check with the information security officer of your organization to confirm that the approach described here can be applied to your use case.

Usage

Create the configuration file.

cp .env.example .env

Specify in the config file what IDE should be used.

# Notebook IDE ("rstudio" or "jupyter")
IDE=rstudio

You can either pull the Docker image nlpsandbox/notebooks or build it using the information included in docker-compose.yml. To pull the image and start the IDE:

docker compose pull
docker compose up

If you want to build the image before starting the IDE, for instance after customizing Dockerfile:

docker compose up --build

To access the IDE, open your browser and follow the instructions below.

Stop the IDE

To stop the IDE, enter Ctrl+C followed by docker compose down. If running in detached mode, you will only need to enter docker compose down.

Tips:

  • Add -d or --detach as in docker compose up -d to run in the background.
  • If the command docker compose is missing, try docker-compose.

Open the RStudio project

In RStudio, open the RStudio project provided by this repository by clicking on the button Project: (None) > Open Project... on the top-right corner, then select the file nlpsandbox/nlpsandbox.Rproj. You should now be able to open and run the notebook r-and-python.Rmd. This will confirm that you can successfully use R and Python togehter.

Configuring the CI/CD workflow

The CI/CD workflow of this repository builds the Docker image nlpsandbox/notebooks and pushes it to Docker Hub.

If you decide to use this repository as a GitHub template, you will need to update the environment variables defined at the top of the CI/CD workflow. You also need to create the following GitHub Secrets.

  • DOCKERHUB_USERNAME: Your Docker Hub username.
  • DOCKERHUB_TOKEN: A personal access token (PAT) that has the permissions to push the image.

Versioning

GitHub tags

This repository uses semantic versioning to track the releases of this project. This repository uses "non-moving" GitHub tags, that is, a tag will always point to the same git commit once it has been created.

Docker image and GitHub Pages

The artifact published by this repository are the Docker image nlpsandbox/notebooks and HTML notebooks published to GitHub Pages.

The table below describes the GH Pages tags available.

Tag name Moving Description
latest Yes Latest stable release.
edge Yes Latest commit made to the default branch.
edge-<sha> No Same as above with the reference to the git commit.
<major>.<minor>.<patch> No Stable release.

You should avoid using a moving tag like latest when deploying containers in production, because this makes it hard to track which version of the image is running and hard to roll back.

License

Apache License 2.0