Usage guide#

Without any installation#

Visit mybinder.org/v2/gh/colomoto/colomoto-docker/mybinder/latest to launch the most recent CoLoMoTo Notebook environment without any installation thanks to Binder services. You can replace latest with a specific image tag.

Note that the computing resources are limited and the storage is not persistent.

With Python Helper Script#

To ease interaction with Docker, we provide a Python helper script colomoto-docker. You need Docker and Python. We support GNU/Linux, macOS, and Windows.

sudo pip install -U colomoto-docker   # only once; you may have to use pip3
colomoto-docker

The container can be stopped by pressing Ctrl+C keys.

Selecting image version#

By default, the script will fetch the most recent colomoto/colomoto-docker tag. A specific tag can be specified using the -V option; or use -V same to use the most recently fetched image. For example:

colomoto-docker                 # uses the most recently fetched image
colomoto-docker -V latest       # fetches the latest published image
colomoto-docker -V 2018-05-29   # fetches a specific image

Writing notebooks#

Warning: by default, the files within the Docker container are isolated from the running host computer, therefore files are deleted after stopping the container, except the files within the persistent directory.

To have access to the files of your current directory you can use the --bind option:

colomoto-docker --bind .

If you want to have the tutorial notebooks alongside your local files, you can do the following:

mkdir notebooks
colomoto-docker -v notebooks:local-notebooks

in the Jupyter browser, you will see a local-notebooks directory which is bound to your notebooks directory.

Selecting interface#

colomoto-docker --lab          # laucnh Jupyter Lab interface (default)
colomoto-docker --notebook     # launch Jupyter Notebook interface
colomoto-docker --shell        # launch shell
colomoto-docker command line   # execute command line in place of launching the interface

Running old images#

On some systems, older images may require changing default security options.

colomoto-docker --ulimit nofile=8096 -V 2018-05-29

Other options#

See

colomoto-docker --help

for other options.

Issues#

Having issues? have a look at our Troubleshooting page, or open an issue.

Python-less usage#

You need Docker.

First, pick an image version among colomoto/colomoto-docker tags. Fetch the image with

docker pull colomoto/colomoto-docker:TAG

The image can be ran using

docker run -it --rm -p 8888:8888 colomoto/colomoto-docker:TAG

then, open your browser and go to http://localhost:8888 for the Jupyter notebook web interface (note: when using Docker Toolbox, replace localhost with the result of docker-machine ip default command).