Dependency management: environment.yml requires old versions

[rossbar]

According to the conda environment docs the environment.yaml requires the use of old versions of many packages (1.12 for numpy, 2.0 for matplotlib, etc.)

This also means that the environment.yaml is actually out-of-sync with the requirements.txt, which takes the latest version for most libraries. It seems there have been issues with this in the past (#217).

I would propose the following:

1. Create an environment using the current release of the dependencies
2. Verify that the building/running of the jupyter notebooks is successful
3. Modify both environment.yaml and requirements.txt so that they are consistent.
4. Verify that all changes result in consistent, working environments with conda and venv independently

I'll take a look at this. Further suggestions (setting up a circleci?) are welcome!

[stefanv]

@rossbar Making those consistent should definitely be done. We have a CI Wercker, so let me know if you have any problems.