Idego Pytech Challenge

We have prepared a contest especially for the attendees of the Pytech Summit 2022 conference! Recall our speech on the biggest event in Poland dedicated to the Python language:

Details of the challenge

You have a set of tasks that needs to be performed by a machine. Each task has the following format:

{
  'name': string,
  'resources': list of strings,
  'payoff': float
}

Each task provides some payoff when completed, and requires a set of resources to be performed.

If two tasks require the same resource, those are incompatible.
If two tasks are incompatible, then the machine can’t do both at the same time.

Example of tasks:

[
  {"name": "picture 1", "payoff": 2.1, "resources": ["resource 5", "resource 2"]},
  {"name": "picture 2", "payoff": 4.5, "resources": ["resource 5", "resource 4", "resource 1"]},
  {"name": "fsck", "payoff": 0.48, "resources": ["resource 4", "resource 2", "resource 3"]},
  {"name": "clean", "payoff": 3.64, "resources": ["resource 5", "resource 6"]}, 
  {"name": "picture 3", "payoff": 5.92, "resources": ["resource 6", "resource 3", "resource 4", "resource 1", "resource 2"]},
  {"name": "backup", "payoff": 1.85, "resources": ["resource 4", "resource 6"]}, 
  {"name": "clean", "payoff": 0.38, "resources": ["resource 3", "resource 2"]}, 
  {"name": "clean camera", "payoff": 4.97, "resources": ["resource 6", "resource 3", "resource 4"]}, 
  {"name": "clean motor", "payoff": 2.95, "resources": ["resource 4", "resource 6", "resource 2"]}, 
  {"name": "picture 4", "payoff": 3.78, "resources": ["resource 5", "resource 4", "resource 2", "resource 1"]}
]

The best subtask for the example tasks is:

[
  {'name': 'clean camera', 'payoff': 4.97, 'resources': ['resource 6', 'resource 4', 'resource 3']},
  {'name': 'picture 1', 'payoff': 2.1, 'resources': ['resource 2', 'resource 5']}
]

Goal

Develop a system that resolves which tasks the machine has to perform to maximize the payoff. This is, find the subset of tasks in a way that the payoff is maximum and all the tasks in the resulting set are compatible.

Results

We received many very well structured answers, so we decided to reward a bit more participants. And so, using a rating system checking for correctness, efficiency, readability, complexity, stability and time, we have selected six laureates: