Class EpisodeRunner.EpisodeResult
java.lang.Object
neqsim.process.ml.EpisodeRunner.EpisodeResult
- All Implemented Interfaces:
Serializable
- Enclosing class:
EpisodeRunner
Result of running an episode.
- Version:
- 1.0
- Author:
- ESOL
- See Also:
-
Field Summary
FieldsModifier and TypeFieldDescriptionfinal List<double[]> History of actions.final double[]Final observation.final List<double[]> History of observations.History of rewards.private static final longfinal intNumber of steps taken.final booleanWhether episode terminated (vs truncated).final doubleTotal reward accumulated. -
Constructor Summary
ConstructorsConstructorDescriptionEpisodeResult(double totalReward, int steps, boolean terminated, List<double[]> observations, List<double[]> actions, List<Double> rewards, double[] finalObservation) Constructor. -
Method Summary
Modifier and TypeMethodDescriptiondouble[]getFeatureTrajectory(int featureIndex) Get feature trajectory as array.doubleGet mean reward per step.doublegetObservation(int index, int featureIndex) Get observation at specific index.
-
Field Details
-
serialVersionUID
private static final long serialVersionUID- See Also:
-
totalReward
public final double totalRewardTotal reward accumulated. -
steps
public final int stepsNumber of steps taken. -
terminated
public final boolean terminatedWhether episode terminated (vs truncated). -
observations
History of observations. -
actions
History of actions. -
rewards
-
finalObservation
public final double[] finalObservationFinal observation.
-
-
Constructor Details
-
EpisodeResult
public EpisodeResult(double totalReward, int steps, boolean terminated, List<double[]> observations, List<double[]> actions, List<Double> rewards, double[] finalObservation) Constructor.- Parameters:
totalReward- total rewardsteps- steps takenterminated- episode terminatedobservations- observation historyactions- action historyrewards- reward historyfinalObservation- final observation
-
-
Method Details
-
getMeanReward
public double getMeanReward()Get mean reward per step.- Returns:
- mean reward
-
getObservation
public double getObservation(int index, int featureIndex) Get observation at specific index.- Parameters:
index- observation indexfeatureIndex- feature index within observation- Returns:
- feature value
-
getFeatureTrajectory
public double[] getFeatureTrajectory(int featureIndex) Get feature trajectory as array.- Parameters:
featureIndex- feature index- Returns:
- array of feature values over time
-