-
Notifications
You must be signed in to change notification settings - Fork 1
ProximitySortPlayerDataTrimAlg

This class is a child of PlayerDataTrimAlg, which trims a list of states based on the age and proximity of the elements. The older states are discarded, and if there are close points they are also discarded to avoid redundancy and increase exploration. It also supports that the residue between two phases is discarded first (this is informed by the stateType in each PlayerState). GIMME can benefit from the latter when bootstrapping the system because simulated data needs to be discarded as real data is added.
+ ProximitySortPlayerDataTrimAlg(maxNumModelElements: int, epsilon: decimal = 0.01, accStateResidue: bool = False): void| Name: expected type | Default value | Description |
|---|---|---|
| maxNumModelElements: int | - | The maximum number of states in the trimmed list. |
| epsilon: decimal | 0.01 | A limit to the distance considered to be close. |
| accStateResidue: bool | False | Informs the algorithm if it should consider stateType (see above). |
+ considerStateResidue(accStateResidue: bool): voidSets the accStateResidue flag.
- proximitySort(elem: PlayerState): decimalComparer to sort a list by proximity.
- creationTimeSort(elem: PlayerState): decimalComparer to sort a list by age.
Overrides trimmedList(pastModelIncs: PlayerState[]): PlayerState[] (see PlayerDataTrimAlg).
Adaptation
Group Configuration Generation
- ConfigsGenAlg
- RandomConfigsGenAlg
- PureRandomSearchConfigsGenAlg
- EvolutionaryConfigsGenAlg
- ODPIPConfigsGenAlg (exact)
- CLinkConfigsGenAlg (legacy)
Preferences Estimation
Quality Evaluation Algorithms
- QualityEvalAlg
- Group-Based Quality Evaluation:
- Regression-Based Quality Evaluation:
- Tabular Quality Evaluation:
Auxiliary Structures
- InteractionsProfile
- PlayerCharacteristics
- PlayerState
- Personality (Inherent Preference):
- PlayerStatesDataFrame
Model Bridges
Player Data Trim