func (this *BFS3Agent) getAction() (index discrete.Action) {
if this.fs3.Dump {
println("getAction")
}
if this.Cfg.ReplanEachStep {
this.ResetPlanner()
}
if this.fs3 == nil {
index = discrete.Action(stat.NextRange(int64(this.task.Act.Ints.Count())))
return
}
node := this.fs3.GetNode(this.stepsWithPlanner, this.belief)
var expanded uint64
for i := uint64(0); i < this.Cfg.MaxTrajectories; i++ {
expandedThisTime := this.fs3.RunTrajectory(node, this.Cfg.Depth)
expanded += expandedThisTime
if this.Cfg.Budget != 0 && expanded > this.Cfg.Budget {
break
}
}
if this.Dump {
this.fs3.Dump = true
fmt.Printf("root:\n%v\n\n", node)
this.fs3.RunTrajectory(node, this.Cfg.Depth)
}
index = discrete.Action(this.fs3.GetAction(node))
fmt.Fprintf(os.Stderr, "%v\n", this.fs3.GetQs(node))
if !this.Cfg.FS3.Shallow {
this.fs3.ClearLevel(this.stepsWithPlanner)
this.stepsWithPlanner++
}
time.Sleep(1e9)
return
}
func (ra *RmaxFSSSAgent) GetAction() (action discrete.Action) {
if ra.s == nil {
action = discrete.Action(stat.NextRange(int64(ra.task.Act.Ints.Count())))
return
}
node := ra.s.GetNode(ra.stepsWithPlanner, ra.mdpo)
action = discrete.Action(ra.s.GetAction(node))
ra.s.ClearLevel(ra.stepsWithPlanner)
return
}
func (this *Agent) GetAction() (action uint64) {
if this.s == nil {
action = uint64(stat.NextRange(int64(this.mdp.GetTask().Act.Ints.Count())))
return
}
node := this.s.GetNode(this.stepsWithPlanner, this.mdpo)
action = uint64(this.s.GetAction(node))
this.s.ClearLevel(this.stepsWithPlanner)
return
}
func (this MapBelief) SampleWorld() (truth MapBelief, rejects int) {
truth = make(MapBelief, 16)
var wumpusLocs []int
smellCount := 0
anySmell := false
for k, v := range this {
if (v & Stinky) != 0 {
anySmell = true
smellCount++
}
if v&Wumpus != 0 {
wumpusLocs = []int{k}
}
}
if len(wumpusLocs) == 0 {
for x := 0; x < 4; x++ {
for y := 0; y < 4; y++ {
if x == 0 && y == 0 {
continue
}
known, c := this.AdjacentStinky(x, y)
if (anySmell && known && c == smellCount) || (!anySmell && !known) {
wumpusLocs = append(wumpusLocs, x+4*y)
}
}
}
}
rejects = 0
for {
for i := range truth {
truth[i] = 0
}
for x := 0; x < 4; x++ {
for y := 0; y < 4; y++ {
k := x + y*4
if k == 0 {
continue
}
if this[k] != 0 {
truth[k] = this[k]
} else {
if this.IsPit(x, y) || (this.CanBePit(x, y) && flip(.2)) {
truth.SetFlag(x, y, Pit)
truth.SetAdjacent(x, y, Breezy)
}
}
}
}
if this.IsPitConsistent(truth) {
break
}
rejects++
}
wk := wumpusLocs[stat.NextRange(int64(len(wumpusLocs)))]
x := wk % 4
y := wk / 4
truth.SetFlag(x, y, Wumpus)
truth.SetAdjacent(x, y, Stinky)
for k, v := range this {
if v != 0 {
truth[k] = v
}
}
for k, v := range truth {
if v == 0 {
truth[k] = Empty
}
}
return
}
func (this *Belief) ConsiderRandomFlip() {
child := stat.NextRange(int64(len(this.learners)))
this.learners[child].ConsiderRandomFlip()
}
func (this *DepLearner) ConsiderRandomFlip() {
this.ConsiderFlip(uint32(stat.NextRange(int64(len(this.bg.ranges)))))
}
func (this *Env) EnvStart() (obs rlglue.Observation) {
startState := discrete.State(stat.NextRange(int64(this.mdp.GetTask().Obs.Ints.Count())))
obs = rlglue.NewObservation(this.mdp.GetTask().Obs.Ints.Values(startState.Hashcode()), []float64{}, []byte{})
this.LastState = startState
return
}
func (this *Posterior) SampleRandomState() {
s := discrete.State(stat.NextRange(int64(this.bg.NumStates)))
this.ResampleState(s)
}
