19 rewards(std::vector<std::vector<double>>(S, std::vector<double>(A, 0.0)))
25 return rewards[state][action];
30 rewards[state][action] = reward;
36 stream <<
"Printing rewards/costs\n";
39 for (
action_t j = 0; j < nbOfActions; j++)
41 stream << rewards[
i][j] <<
" ";
void print(std::ostream &stream)
void setReward(state_t state, action_t action, double reward)
double getReward(state_t, action_t)
Rewards(size_t nbOfStates, size_t nbOfActions)