uptr<Action> chooseAction(State* state) {
if (!haveSeenState(state)) {
handleNewState(state);
}
return chooseActionForState(state);
}
void GameClient::connectToServer(QString host, int port)
{
QString ipAddress;
QList<QHostAddress> ipAddressesList = QNetworkInterface::allAddresses();
// use the first non-localhost IPv4 address
for(int i = 0; i < ipAddressesList.size(); ++i) {
if(ipAddressesList.at(i) != QHostAddress::LocalHost &&
ipAddressesList.at(i).toIPv4Address()) {
ipAddress = ipAddressesList.at(i).toString();
break;
}
}
m_connection = new Connection(Connection::Client);
// bind the connection to a certain port
int i = 0;
bool ok;
do {
i ++;
ok = m_connection->bind(QHostAddress(ipAddress), StartBindPort+1000*i);
} while(!ok);
m_connBindPort = StartBindPort + 1000 * i;
m_connection->connectToHost(host, port);
// each connection use the same port as first time
connect(m_connection, SIGNAL(newGreeting(QByteArray)), this, SLOT(handleNewGreeting(QByteArray)));
connect(m_connection, SIGNAL(newState(QByteArray)), this, SLOT(handleNewState(QByteArray)));
connect(m_connection, SIGNAL(newBackupServer(QByteArray)), this, SLOT(handleNewBackupServer(QByteArray)));
connect(this, SIGNAL(haveMessageToSend(Connection::DataType,QByteArray)), m_connection, SLOT(sendMessage(Connection::DataType,QByteArray)));
emit newPrimaryServerInfo(host+','+QString::number(port));
}
/**
* @brief SCGraphicsView::createGraph
* add all states and transitions to the graphics view
*/
void SCGraphicsView::createGraph()
{
QList<SCState*> states;
_dm->getAllStates(states);
// connect the top state
connectState(_dm->getTopState());
// first load all states
for(int s = 0; s < states.count(); s++)
{
handleNewState(states[s]);
}
// load all transitions
for(int i = 0; i< states.count(); i++)
{
QList<SCTransition*> transitions;
states[i]->getTransitions(transitions);
for(int t = 0; t < transitions.count(); t ++)
{
handleNewTransition(transitions[t]);
}
}
}
void ActionOutcome(pair<State *, Action *> performed, pair<State *, double> outcome) {
State *startState = performed.first;
Action *actionPerformed = performed.second;
State *resultState = outcome.first;
double reward = outcome.second;
if (!haveSeenState(startState)) {
handleNewState(startState);
}
if (!haveSeenState(resultState)) {
handleNewState(resultState);
}
ActionValue *curValue = findActionValue(startState, actionPerformed);
assert(curValue != nullptr);
double newQ = reward + futureDiscount * maxQ(resultState);
// cout << "r: " << reward << " fd: " << futureDiscount << " newQ: " << newQ << endl;
curValue->value += learnRate * (newQ - curValue->value);
// cout << "v : " << curValue->value << endl;
// getchar();
}
void actionOutcome(pair<State*, Action*> performed, pair<State*, double> outcome) {
State *startState = performed.first;
Action *actionPerformed = performed.second;
State *resultState = outcome.first;
double reward = outcome.second;
assert(haveSeenState(startState));
if (!haveSeenState(resultState)) {
handleNewState(resultState);
}
ActionValue *curValue = findActionValue(startState, actionPerformed);
assert(curValue != nullptr);
double newQ = reward + futureDiscount * maxQ(resultState);
curValue->value += learnRate * (newQ - curValue->value);
}
