Personal configuration to use a consistent Hydrogen -> Atom environment with a Docker container
Only works with Python3 kernels; not yet able to use kernels for Julia, R, Javascript.
- Install Atom
$ brew cask install atom
- Install hydrogen
$ apm install hydrogen
- Install Docker
$ brew cask install docker
- Clone this repository
$ git clone git@github.com:riordan/my-hydrogen.git
- Set a new token for this machine. Copy
.env.sample
to.env
and randomly generate a new token - Run the docker-compose
$ docker-compose up -d
- In Atom go to Settings->Hydrogen. Add this as a Gateway (but update the
token
): ``` [{ "name": "Local jupyter-datascience-notebook", "options": { "baseUrl": "http://127.0.0.1:8855", "token": "my_secret_token" } }]
1. In Atom go to Packages->Hydrogen->Connect to Remote Kernels and connect to this kernel.
To shut down, after the session's over, you can just destroy the kernel by switching to this folder and running `$ docker-compose down` or stop it by `$ docker-compose stop`.
# Background
This is just a Dockerfile for use with [Hydrogen's Remote Kernels](https://nteract.gitbooks.io/hydrogen/docs/Usage/RemoteKernelConnection.html) feature.