/**
* Transition function. Currently, we only get the list of probabilities of each item.
*
* @param current - The current state
* @param action - The action
* @param possible - The possible state
* @return The list of probabilities where each index refers to the probability of that item
*/
private List<Double> transition(State current, Action action, State possible) {
List<Double> probs = new ArrayList<Double>(); // Probabilities
Map<Integer, Integer> currentStock, purchase, possibleStock; // Maps
double currentProb; // The current probability
Matrix currentMatrix; // The current probability matrix
int row, column; // The row and column
for (int i = 0; i < current.getState().size(); ++i) {
currentProb = 0.0;
currentStock = current.getState();
purchase = action.getPurchases();
possibleStock = possible.getState();
row = currentStock.get(i) + purchase.get(i);
column = row - possible.getState().get(i);
currentMatrix = this.probabilities.get(i);
if (column < 0
|| column >= currentMatrix.getNumCols()
|| row >= currentMatrix.getNumRows()) { // Invalid state
probs.add(0.0);
continue;
}
if (possibleStock.get(i) > 0
|| (possibleStock.get(i) == 0 && column == 0)) { // Sufficiently provided
currentProb = currentMatrix.get(row, column);
} else if (possibleStock.get(i) == 0 && column > 0) {
// Range of probabilities because user could have eaten plenty
for (int j = column; j < currentMatrix.getNumCols(); ++j) {
currentProb += currentMatrix.get(row, j);
}
}
probs.add(currentProb);
}
return probs;
}
/**
* Find out the maximum value for a given map of actions
*
* @param actions - The map of actions to their values
* @return Return the action with the largest value out of all the actions
*/
private Action maxArg(Map<Action, Double> actions) {
Action largestValue = null; // The largest value
for (Action current : actions.keySet()) {
if (largestValue == null || (actions.get(largestValue) < actions.get(current))) {
largestValue = current;
}
}
return largestValue;
}
/**
* Return the sum of the integers in the list
*
* @param integerMap - The map of integers
* @return - the sum of integers in the list
*/
private int sumOf(Map<Integer, Integer> integerList) {
int sum = 0; // The sum
for (Integer current : integerList.keySet()) {
sum += integerList.get(current);
}
return sum;
}
/** Generate optimal policy */
private void generateOptimalPolicy() {
double startTime, currentTime; // Timer things
boolean alreadyRunThrough = false; // Record if we've already done one run
Set<State> toLookup; // The states to lookup
Map<Action, Double> differentValues =
new HashMap<Action, Double>(); // The different values for the current state
Map<State, Action> newPolicy = new HashMap<State, Action>(); // New policy
State newState; // New state for new policy
Action best; // The best action
startTime = Global.currentTime();
currentTime = Global.currentTime();
while ((currentTime - startTime) <= this.timeRemaining) {
if (alreadyRunThrough) {
toLookup = this.policy.keySet();
} else {
toLookup = this.possibleStates;
}
for (State currentState : toLookup) {
differentValues.clear(); // Reset different values
for (Action currentAction : this.possibleActions) {
if (!validAction(currentAction, currentState)) {
continue;
}
differentValues.put(
currentAction, valueGeneration(currentState, currentAction, toLookup));
}
best = maxArg(differentValues);
if (isBetterPolicy(currentState, differentValues.get(best))) {
newState = new State(currentState.getState());
newState.setTemporaryCost(differentValues.get(best));
newPolicy.put(newState, best);
}
}
if (CheckDifference(0.1, newPolicy)) {
copyPolicy(newPolicy);
newPolicy.clear();
alreadyRunThrough = true;
} else {
currentTime = Global.currentTime();
break;
}
}
}
/**
* Make a copy of a policy
*
* @param Map<State, Action> newPolicy - The new policy
*/
private void copyPolicy(Map<State, Action> newPolicy) {
/* Containers for clones */
State current;
Action currentAction;
this.policy = new HashMap<State, Action>();
for (State newState : newPolicy.keySet()) {
current = new State(newState.getState());
current.setCost(newState.getTemporaryCost());
currentAction = new Action(newPolicy.get(newState).getPurchases());
this.policy.put(current, currentAction);
}
}
public List<Integer> generateShoppingList(List<Integer> inventory, int numWeeksLeft) {
State current; // current state
Map<Integer, Integer> purchases = new HashMap<Integer, Integer>(); // Purchases
Map<Integer, Integer> map = new HashMap<Integer, Integer>(); // Mapping of integers
List<Integer> shopping = new ArrayList<Integer>(); // Shopping items
for (int i = 0; i < inventory.size(); ++i) {
map.put(i, inventory.get(i));
}
current = new State(map);
purchases = this.policy.get(current).getPurchases();
for (int i = 0; i < inventory.size(); ++i) {
shopping.add(purchases.get(i));
}
return shopping;
}
