Roadmap

Purpose of this document

This document list general directions that core contributors are interested to see developed in opt-sugar. The fact that an item is listed here is in no way a promise that it will happen, as resources are limited. Rather, it is an indication that help is welcomed on this topic.

Backlog

1. [In Progress] Define a low_sugar submodule, a lighter interface for gurobi models, not requiring that much sweet.

2. Add hierarchical optimization capabilities.

3. Check if the ModelBuilder and OptModel in extra_sugar need to be defined.

4. Include logger results as part of the OptModel object attributes. This can be done using grblogtools.

5. Consider developing mlflow autolog for optimization model autolog.

6. Add an experiment tracking example gallery.

7. Modify the ModelBuilder class to have an optimize method that plays well with mlflow.

Examples related

1. For Supply Chain (superheros factory) examples:

   • Single objective: Mini toy example (No negative inventories).

   • Single objective: Mini toy example (Negative inventories required).

   • Single objective: Establishing best objective balance using experiment tracking.

   • Multiple Objectives: Priority customers.

   • Add output visualization using pandas.

2. Add examples using the model_builder_factory method.

3. Add examples for the low_sugar submodule.

   • Add a sudoku example using low_sugar.

   • Add a coloring example illustrating how to log parameters and metrics.

   • Add a coloring example illustrating how to run experiments with multiple runs.

   • Add (refactor) the superheros factory examples.