Diet example rendered in Tidy, Tested, Safe format

Usage:

    python -m tts_diet -i input.xlsx -o solution.xlsx

If you're new to ticdat, then please read this to orient yourself.

• tidy The engine is organized as a package. Free standing .py files are appropriate for proof-of-concept and educational purposes, but industrial code typically requires a package whose __version__ attribute can be cross-referenced against a GitHub release tag. The package is organized into public and private sections (this example is too small to have the latter).
• tested There is a testing section for the engine that uses unittest (or something similar, like pytest) to validate the behavior of the engine. Using a tool like coverage with the testing section will create a report demonstrating that the testing code exercises close to 100% of the production code.
• safe The solve function fully validates the input data for data integrity problems. This is not to say that infeasible models are always pre-diagnosed (analyzing feasibility is mathematically the same as analyzing optimality) but all mathematically trivial data issues are discovered prior to performing true optimization.

This example requires gurobipy to solve, but the schemas can still be examined even if gurobipy is not installed.

We also include a very simple setup.py file, so that the tts_diet package can be distributed as a .whl file.

Contents

• tts_diet The subdirectory that defines the tts_diet package.
• test_tts_diet The subdirectory that contains data and code required for testing tts_diet.
• setup.py Standard file for distributing tts_diet. Note that gurobipy is not specified as a requirement, as tts_diet can still provide partial functionality in its absence. Deliberately creating a very thin setup.py for demonstration purposes. Use python setup.py sdist bdist_wheel to create the .whl file.