CPMpy tutorial video and roadmap

[Tias Guns]

Dear constraints mailinglist,

The video tutorial of CPMpy, a Constraint Programming and Modeling
library in Python, based on numpy, with direct solver access, was first
presented during (virtual) CP in October. It is also generally available
on YouTube:

https://www.youtube.com/watch?v=A4mmmDAdusQ


We announced earlier that we would aim for a first stable release by the
tutorial. Based on the experience of implementing the many meta-examples
(incl MUS, explanation sequences, neural network integrations and more),
and based on the feedback at the Q&As at CP, we decided to post-pone this.

Our current plans for a stable release (Feb 2022) is:
  - the 'expressions' part is considered stable; how to write models
and constraints will not or barely change
  - the SolverInterfaces will be refactored in a uniform way, allowing
the following:
  - add a MIP solver (currently: CP [ortools,minizinc] and SAT [pysat])
  - document how to add more solvers
  - make a generic solveAll() function, so that we can use efficient
solver-specific methods behind it
  - document how to use solver-specific constraints (much more easily)


In the mean time I encourage you to try it out, or have a look at the
(advanced) examples or the docs:
https://github.com/CPMpy/cpmpy/tree/master/examples/advanced
https://cpmpy.readthedocs.io/

If you have feedback or things you wish were possible, reach out on the
Issues page or by mail.

https://github.com/CPMpy/cpmpy


Kind regards,
Tias Guns
https://people.cs.kuleuven.be/~tias.guns/