CG:SHOP - Official pyutils24 for the 2024 Challenge on Packing Polygons

Utilities for verifying solutions of the CG:SHOP 2024 Competition. Feel free to use the code, but it is optimized for exact verification not for sampling or other purposes.

Installation

You can install this package via PyPI:

pip install -U -v cgshop2024-pyutils

During the installation, CGAL and other dependencies will be downloaded and compiled. This can take a while but should happen mostly automatic. You need to have a C++ compiler installed.

Please check for updates of the utils frequently as we are still working on them.

Usage

Reading instances

from cgshop_pyutils24 import InstanceDatabase

db = InstanceDatabase("../cgshop2024_benchmark.zip")

instance = db["atris1240"]
assert isinstance(instance, dict)

for instance in db:
    print("Found", instance["instance_name"])

The instances are dicts of the following format:

{
  "type": "cgshop2024_instance",
  "instance_name": "test",
  "num_items": 9,
  "container": {
    "x": [26, 2, 0, 12, 91, 158, 180, 194, 225, 232, 225, 210, 200],
    "y": [250, 202, 149, 107, 0, 30, 42, 56, 131, 151, 177, 225, 243]
  },
  "items": [
    {
      "value": 601,
      "quantity": 1,
      "x": [3, 40, 113, 134, 85, 143, 152, 157, 157, 92, 18, 12, 2, 0],
      "y": [31, 59, 59, 27, 6, 0, 51, 95, 139, 141, 77, 145, 152, 65]
    },
    {
      "value": 90,
      "quantity": 1,
      "x": [1, 6, 25, 54, 49, 39, 25, 31, 14, 0, 13],
      "y": [11, 10, 8, 0, 23, 28, 37, 56, 34, 39, 26]
    },
    { "value": 1, "quantity": 1, "x": [12, 0, 28], "y": [16, 26, 0] },
    {
      "value": 9,
      "quantity": 1,
      "x": [8, 5, 0, 12, 13],
      "y": [14, 19, 12, 0, 5]
    },
    { "value": 133, "quantity": 1, "x": [101, 61, 0], "y": [0, 104, 95] },
    { "value": 51, "quantity": 2, "x": [68, 48, 0, 71], "y": [45, 38, 17, 0] },
    { "value": 25, "quantity": 1, "x": [21, 28, 0, 1], "y": [0, 47, 17, 0] },
    {
      "value": 37,
      "quantity": 1,
      "x": [11, 1, 17, 26, 19, 0, 9],
      "y": [16, 0, 16, 12, 34, 34, 26]
    },
    { "value": 32, "quantity": 1, "x": [58, 73, 79, 0], "y": [5, 10, 39, 0] }
  ]
}

Verifying solutions

from cgshop2024_pyutils import (
    verify,
    InstanceDatabase,
    InvalidSolution,
    BadSolutionFile,
)

db = InstanceDatabase("../cgshop2024_benchmark.zip")
solution = parse_solution(SOLUTION_DATA)
instance = db[solution["instance_name"]]

try:
    value = verify(instance, solution)
except InvalidSolution as inv_sol:
    print("Solution is invalid:", inv_sol)
except BadSolutionFile as bad_sol:
    print("Solution file is invalid:", bad_sol)
else:
    print("Solution is valid and has value", value)

Verifying a batch of solutions

from cgshop2024_pyutils import (
    verify_batch,
    InstanceDatabase,
    InvalidSolution,
    BadSolutionFile,
    ZipSolutionIterator,
    ZipReaderError,
)

db = InstanceDatabase("../cgshop2024_benchmark.zip")
sol_it = ZipSolutionIterator()
try:
    for solution in sol_it(ZIP_FILE):
        instance = db[solution["instance_name"]]
        try:
            value = verify(instance, solution)
        except InvalidSolution as inv_sol:
            print("Solution is invalid:", inv_sol)
            break
        except BadSolutionFile as bad_sol:
            print("Solution file is invalid:", bad_sol)
            break
        print("Solution is valid and has value", value)
except ZipReaderError as err:
    print("Error while reading zip file:", err)

Trouble Shooting

The native parts are known to cause trouble on some systems. Check out this page for solutions of various problems.

License

The code in this repository can be used under the terms of the MIT license. However, the package binds against CGAL which is licensed under the GPLv3. This should still allow you to modify and use the code for the competition.

Changelog

• 1.2.2: Previous fix did not work sufficiently. Fixed it now (hopefully).
• 1.2.1: Fixing Issue #2 and a crash when the type field is missing.
• 1.2.0: Fixing catching of exception of bad files at the wrong place. Improved error messages.
• 1.1.1: A long on Windows is only 32bit, so changed to using the explicit int64_t type.
• 1.1.0: The previous workaround could lead to excessive memory usage. Changed quadtree implementation to keep large elements on higher levels even if the node is full. Gives a warning in this case.
• 1.0.1: Only giving a warning if a node cannot be split instead of throwing an exception. Issue #1
• 1.0.0: Initial release