func NewEnv(belief *Belief) (this *Env) {
this = new(Env)
this.numActions = len(belief.counts)
this.belief = belief
this.obs = rlglue.NewObservation([]int32{0}, nil, nil)
return
}
func (this *Env) MakeObs() (obs rlglue.Observation) {
k := this.hunter.x + this.hunter.y*4
this.observed[k] = this.hunter.belief[k]
intarray := []int32{this.hunter.x, this.hunter.y, this.hunter.dir}
indices := make([]int32, 16)
for i, v := range this.observed {
indices[i] = GetIndex(v)
}
intarray = append(intarray, indices...)
obs = rlglue.NewObservation(intarray, []float64{}, []byte{})
return
}
func (this *Environment) ConstructObs() (obs rlglue.Observation) {
ints := make([]int32, this.cfg.NumSystems)
for i := range ints {
if this.status[i] {
ints[i] = 1
}
}
obs = rlglue.NewObservation(ints, []float64{}, []byte{})
this.hash = this.task.Obs.Ints.Index(ints)
count++
return
}
func (this *Env) makeObs() (obs rlglue.Observation) {
ints := make([]int32, 4*len(this.Cans))
for i, can := range this.Cans {
if can.Painted {
ints[i*4] = 1
}
if can.Polished {
ints[i*4+1] = 1
}
if can.Scratched {
ints[i*4+2] = 1
}
if can.Done {
ints[i*4+3] = 1
}
}
obs = rlglue.NewObservation(ints, []float64{}, []byte{})
return
}
func (ge *Env) EnvStep(action rlglue.Action) (obs rlglue.Observation, r float64, t bool) {
t = ge.Pos.Equals(ge.Goal)
r = -1
if t {
r = 0
}
dir := action.Ints()[0]
u := rand.Float64()
if u < .1 {
dir += 1
} else if u < .2 {
dir += 3
}
dir %= 4
newPos := ge.Pos
switch dir {
case 0:
newPos.Y++
case 1:
newPos.X++
case 2:
newPos.Y--
case 3:
newPos.X--
}
if newPos.X < 0 {
newPos.X = 0
}
if newPos.Y < 0 {
newPos.Y = 0
}
if newPos.X >= ge.Width {
newPos.X = ge.Width - 1
}
if newPos.Y >= ge.Height {
newPos.Y = ge.Height - 1
}
ge.Pos = newPos
obs = rlglue.NewObservation([]int32{ge.Pos.X, ge.Pos.Y}, []float64{}, []byte{})
return
}
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 (ge *Env) EnvStart() (obs rlglue.Observation) {
ge.Pos = ge.Start
obs = rlglue.NewObservation([]int32{ge.Pos.X, ge.Pos.Y}, []float64{}, []byte{})
return
}