S - State classsA - Actions class.public class ProbabilitySolver<S extends State,A extends Action>
extends java.lang.Object
| Constructor and Description |
|---|
ProbabilitySolver(CTMDP<S,A> problem)
Initializes a new solver for continuous chains and solves the
probabilities for the optimal policy.
|
ProbabilitySolver(CTMDP<S,A> problem,
DecisionRule<S,A> dr)
Initializes a new solver for continuous chains and solves the
probabilities for a particular decision rule.
|
ProbabilitySolver(DTMDP<S,A> problem)
Initializes a new solver for discrete chains and solves the probabilities
for the optimal policy.
|
ProbabilitySolver(DTMDP<S,A> problem,
DecisionRule<S,A> dr)
Initializes a new solver for discrete chains
|
| Modifier and Type | Method and Description |
|---|---|
ValueFunction<S> |
getProbability() |
boolean |
isSolved() |
void |
setGaussSeidel(boolean val)
The GaussSeidel modification of the ValueIteration method is a change
that is garanteed to have a performance at least as good as the methods
without the modifications.
|
void |
setJacobi(boolean val) |
void |
solve()
Solves the probabilities
|
public ProbabilitySolver(DTMDP<S,A> problem, DecisionRule<S,A> dr)
problem - discrete time , infinite horizon problemdr - decision rule to be evaluatedpublic ProbabilitySolver(DTMDP<S,A> problem) throws SolverException
problem - discrete time , infinite horizon problemSolverExceptionpublic ProbabilitySolver(CTMDP<S,A> problem, DecisionRule<S,A> dr)
problem - continuous time , infinite horizon problemdr - public ProbabilitySolver(CTMDP<S,A> problem) throws SolverException
problem - continuous time , infinite horizon problemSolverExceptionpublic boolean isSolved()
public void solve()
public void setGaussSeidel(boolean val)
val - sets whether or not the GaussSeidel modification will be used.public void setJacobi(boolean val)
val - true to use jacobi methodspublic ValueFunction<S> getProbability()