class
RandomForestClassification[V, F, K] extends RandomForest[V, F, K]
Instance Constructors
-
new
RandomForestClassification()(implicit V: CoordinateSpace[V, F], order: Order[F], vectorClassTag: ClassTag[V])
Type Members
-
case class
FixedOptions(numAxesSample: Int, numPointsSample: Int, numTrees: Int, minSplitSize: Int, parallel: Boolean) extends Product with Serializable
-
case class
Forest(trees: List[DecisionTree[V, F, K]]) extends Product with Serializable
-
-
-
-
trait
RegionLike extends AnyRef
Value Members
-
final
def
!=(arg0: AnyRef): Boolean
-
final
def
!=(arg0: Any): Boolean
-
final
def
##(): Int
-
final
def
==(arg0: AnyRef): Boolean
-
final
def
==(arg0: Any): Boolean
-
implicit
def
F: Field[F]
-
-
implicit
val
V: CoordinateSpace[V, F]
-
def
apply(data: Array[V], out: Array[K], options: RandomForestOptions): (V) ⇒ K
-
final
def
asInstanceOf[T0]: T0
-
def
clone(): AnyRef
-
def
defaultOptions(size: Int): FixedOptions
-
final
def
eq(arg0: AnyRef): Boolean
-
def
equals(arg0: Any): Boolean
-
def
finalize(): Unit
-
def
fromForest(forest: Forest): (V) ⇒ K
-
final
def
getClass(): Class[_]
-
def
hashCode(): Int
-
final
def
isInstanceOf[T0]: Boolean
-
final
def
ne(arg0: AnyRef): Boolean
-
final
def
notify(): Unit
-
final
def
notifyAll(): Unit
-
implicit
val
order: Order[F]
-
def
randomForest(data: Array[V], outputs: Array[K], opts: FixedOptions): Forest
-
final
def
synchronized[T0](arg0: ⇒ T0): T0
-
def
toString(): String
-
implicit
val
vectorClassTag: ClassTag[V]
-
final
def
wait(): Unit
-
final
def
wait(arg0: Long, arg1: Int): Unit
-
final
def
wait(arg0: Long): Unit
Inherited from AnyRef
Inherited from Any
A
RandomForest
implementation for classification. In this case, the outputs (dependent variable) belongs to some typeK
. This type needs to be a well behaved Java object as itsequals
andhashCode
will be used to determine equality of classes. This implementation uses a majority vote method to determine classification. Each region in a tree is associated with the most popular class in that region. Ties are broken randomly (not really). Within a forest, each tree casts its vote for classification of a point and the majority wins. Again, ties are broken randomly (again, not really).