随机森林 java_Spark随机森林RandomForest

位于ml/tree/impl/目录下。mllib目录下的随机森林算法也是调用的ml下的RandomForest。ml是mllib的最新实现，将来是要替换掉mllib库的。

RandomForest核心代码

train方法

每次迭代将要计算的node推入堆栈，选择参与计算的抽样数据，计算该节点，循环该过程。

while(nodeStack.nonEmpty) {

// Collect some nodes to split, and choose features for each node (if subsampling).

// Each group of nodes may come from one or multiple trees, and at multiple levels.

val(nodesForGroup,treeToNodeToIndexInfo) =

RandomForest.selectNodesToSplit(nodeStack,maxMemoryUsage,metadata,rng)

// Sanity check (should never occur):

assert(nodesForGroup.nonEmpty,

s"RandomForest selected empty nodesForGroup.  Error for unknown reason.")

// Only send trees to worker if they contain nodes being split this iteration.

val topNodesForGroup: Map[Int, LearningNode] =

nodesForGroup.keys.map(treeIdx => treeIdx ->