object Chapter3
- Source
- Chapter3.scala
Linear Supertypes
Ordering
- Alphabetic
- By Inheritance
Inherited
- Chapter3
- AnyRef
- Any
- Hide All
- Show All
Visibility
- Public
- All
Value Members
- val allowedIterations: Long
- val emptyFn: StateValueFn[Position, DecayState[Double]]
- val epsilon: Double
- val gamma: Double
- val gridConf: Config
-
def
main(items: Array[String]): Unit
This currently works, and displays rough tables for each of the required bits.
- def notConverging(iterations: Long, allowed: Long): Boolean
- def printFigure[T](conf: Config, pair: (StateValueFn[Position, T], Long), title: String)(implicit arg0: ToDouble[T]): Unit
- def shouldStop[Obs, T](l: StateValueFn[Obs, T], r: StateValueFn[Obs, T], iterations: Long)(implicit arg0: ToDouble[T]): Boolean
-
def
threeFive: (StateValueFn[Position, DecayState[Double]], Long)
This is Figure 3.5.
This is Figure 3.5. This is currently working!
-
def
threeTwo: (StateValueFn[Position, DecayState[Double]], Long)
This is Figure 3.2, with proper stopping conditions and everything.
This is Figure 3.2, with proper stopping conditions and everything. Lots of work to go.
- def toTable(conf: Config, f: (Position) ⇒ Double): Iterable[Iterable[Double]]
-
def
valueFunctionConverged[Obs, T](l: StateValueFn[Obs, T], r: StateValueFn[Obs, T])(implicit arg0: ToDouble[T]): Boolean
Note...
Note... this version, following the python code, checks that the sum of all differences is less than epsilon. In the next chapter we use the max function instead here to get this working, to check that the maximum delta is less than epsilon.
edit this text on github
ScalaRL
This is the API documentation for the ScalaRL functional reinforcement learning library.
Further documentation for ScalaRL can be found at the documentation site.
Check out the ScalaRL package list for all the goods.