Custom lightweight Jupyter images for your taste!
I use Jupyter images for almost everything now, but one reason is learning languages or experimenting.
This repo contains the dockerfiles for custom learning base images for certain jupyter kernels:
- Jupyter Python3, ~615MB (cprieto/jupyter-python3), with Python3 kernel, pandas, matplotlib, Pillow, plotnine, numpy, scipy and drivers for postgresql and mysql.
- Jupyter OCaml, ~700MB (cprieto/jupyter-ocaml), with OCaml kernel https://github.com/akabe/ocaml-jupyter
- Jupyter Elixir, ~415MB (cprieto/jupyter-elixir), with the IElixir kernel https://github.com/pprzetacznik/IElixir
- Jupyter F#, ~580MB (cprieto/jupyter-fsharp) with the IFSharp kernel https://github.com/fsprojects/IfSharp
- Jupyter Javascript, ~360M (cprieto/jupyter-javascript) with the IJavascript kernel https://github.com/n-riesco/ijavascript
- Jupyter TypeScript, ~405MB (cprieto/jupyter-typescript) with the ITypeScript kernel https://github.com/nearbydelta/itypescript
- Jupyter Kotlin, ~750MB (cprieto/jupyter-kotlin) with the Kotlin kernel https://github.com/ligee/kotlin-jupyter
- Jupyter Erlang, ~350MB (cprieto/jupyter-erlang) with Erlang kernel https://github.com/filmor/ierl
- Jupyter Go, ~980MB (cprieto/jupyter-go) with LGo kernel https://github.com/yunabe/lgo, the biggest in the group thanks to a bug in the LGO kernel when using MUSL instead of GLib
- Jupyter Scala, ~400MB (cprieto/jupyter-scala) with the Almond Scala kernel https://github.com/almond-sh/almond
- Jupyter Ruby, ~380MB (cprieto/jupyter-ruby) with the SciRuby/iruby kernel
To get any of these images just docker pull <image name> and run it with:
docker run -p 8888:8888 <image name>
If you want to delete the container after done (because why should you keep it?)
docker run -p 8888:8888 --rm <image name>
And finally, the notebooks are saved in a local volume under /notebooks and you can mount that directory to a local directory, for example, if you want mount and save your notebooks in the experiments directory in your local path
docker run 8888:8888 --rm -v $(pwd)/experiments:/notebooks <image name>
Enjoy!