ConfigsGenAlg
ConfigsGenAlg is an abstract base class for the group organization algorithms. These algorithms are usually called coalition structure generation algorithms. GIMME currently implements a few of them, such as: a purely random generator; an exploratory algorithm which does not save any information between iterations; a genetic driven approach (still to be fully implemented); etc...
ConfigsGenAlg(numberOfConfigChoices, minNumberOfPlayersPerGroup, maxNumberOfPlayersPerGroup, preferredNumberOfPlayersPerGroup, fitnessWeights, regAlg)| Name: expected type | Default value | Description |
|---|---|---|
| playerModelBridge: PlayerModelBridge | - | The connector for the players data model |
| interactionsProfileTemplate: InteractionsProfile | - | A template for creating new profiles. The template acts as a facilitator to not only the number of interaction dimensions to be considered, but also their designations. |
| preferredNumberOfPlayersPerGroup: int | None | A preferred group size to consider. When filled (=/= None), the configurations will try following a fixed group size (with the few smaller groups containing the remaining players) |
| minNumberOfPlayersPerGroup: int | 2 | The minimum group size to consider |
| maxNumberOfPlayersPerGroup: int | 5 | The maximum group size to consider |
init(): voidAbstract method to be used for initializing a configuration algorithm if data is stored between iterations.
reset(): voidAbstract method to be used for resetting the configuration algorithm if data is stored between iterations.
randomConfigGenerator(playerIds, minNumGroups, maxNumGroups): voidAuxiliary method which generates a random group configuration.
organize(): int[][]Actually performs the group organization and returns a group configuration (or coalition structure).
updateMetrics( generatedConfig: int[][] ) : voidUpdates groupSizeFreqs and configSizeFreqs.
| Name: expected type | Default value | Description |
|---|---|---|
| generatedConfig: int[][] | - | The generated group organization |
configsAlg = SimpleConfigsGen(playerBridge, regAlg = KNNRegression(playerBridge, 5), numberOfConfigChoices=100, preferredNumberOfPlayersPerGroup = 4, qualityWeights = PlayerCharacteristics(ability=0.5, engagement=0.5))
print(configsAlg.organize())
# Example return for a 10 player class: {'groups': [[6, 0, 2, 7], [1, 5, 4, 3], [9, 8]], 'profiles':
#[<InteractionsProfile.InteractionsProfile object at 0x7f426050bad0>, <InteractionsProfile.InteractionsProfile object
#at 0x7f426050bb10>, <InteractionsProfile.InteractionsProfile object at 0x7f426050bcd0>], 'avgStates':
#[<PlayerStructs.PlayerState object at 0x7f426050bb50>, <PlayerStructs.PlayerState object at 0x7f426050bc10>,
#<PlayerStructs.PlayerState object at 0x7f426050bbd0>]}