You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
FYI, after I forked, fixed, tried to improve pyschedule, I decided to write another scheduling python library from scratch. Code is hosted at https://github.com/tpaviot/ProcessScheduler (GPLv3 or higher), it's a first draft that currently lacks many useful pyschedule features. From my experience with pyschedule, design choices are:
use one single solver to perform computations. It's too much work to maintain different solvers. The choice for the solver is the Z3Prover
don't use list comprehension. Although it is an elegant way to define and create lists, nested lists comprehension result in code I can't fully understand. I prefer procedures, where the algorithm clearly appears even if the code size is larger.
The idea is not to compete with pyschedule (that would make no sense at all), but rather to explore what can be achieved using a different technology than MILP.
The text was updated successfully, but these errors were encountered:
FYI, after I forked, fixed, tried to improve pyschedule, I decided to write another scheduling python library from scratch. Code is hosted at https://github.com/tpaviot/ProcessScheduler (GPLv3 or higher), it's a first draft that currently lacks many useful pyschedule features. From my experience with pyschedule, design choices are:
The idea is not to compete with pyschedule (that would make no sense at all), but rather to explore what can be achieved using a different technology than MILP.
The text was updated successfully, but these errors were encountered: