private OnPolicyRewardMonitor createRewardMonitor(
String prefix, int nbBins, Parameters parameters) {
int nbEpisode = parameters.nbEpisode();
int maxEpisodeTimeSteps = parameters.maxEpisodeTimeSteps();
if (nbEpisode == 1 || parameters.hasFlag(RLParameters.OnPolicyTimeStepsEvaluationFlag))
return new RewardMonitorAverage(prefix, nbBins, maxEpisodeTimeSteps);
return new RewardMonitorEpisode(prefix, nbBins, nbEpisode);
}
public static List<Parameters> combine(List<Parameters> existing, String label, double[] values) {
assert existing.size() > 0;
List<Parameters> combination = new ArrayList<Parameters>();
for (Parameters parameters : existing) {
for (double value : values) {
Parameters combinedParameters = new Parameters(parameters);
combinedParameters.putSweepParam(label, value);
combination.add(combinedParameters);
}
}
return combination;
}
public static List<Parameters> filter(List<Parameters> parameters, String... filters) {
Map<String, Double> filterMap = new LinkedHashMap<String, Double>();
for (String filterString : filters) {
int equalIndex = filterString.indexOf('=');
filterMap.put(
filterString.substring(0, equalIndex),
Double.parseDouble(filterString.substring(equalIndex + 1)));
}
List<Parameters> result = new ArrayList<Parameters>();
for (Parameters parameter : parameters) {
boolean satisfy = true;
for (Map.Entry<String, Double> entry : filterMap.entrySet()) {
if (!parameter.hasKey(entry.getKey())) continue;
double parameterValue = parameter.get(entry.getKey());
if (parameterValue != entry.getValue()) {
satisfy = false;
break;
}
}
if (satisfy) result.add(parameter);
}
return result;
}
