hex.tree.ScoreBuildHistogram Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of h2o-algos Show documentation
H2O Algorithms
There is a newer version: 3.46.0.6
package hex.tree;

import hex.genmodel.utils.DistributionFamily;
import water.H2O.H2OCountedCompleter;
import water.MRTask;

/**  Score and Build Histogram
 *
 * Fuse 2 conceptual passes into one:
 *
 * 

 *
 * Pass 1:
Score a prior partially-built tree model, and make new Node assignments to
 * every row.  This involves pulling out the current assigned DecidedNode,
 * "scoring" the row against that Node's decision criteria, and assigning the
 * row to a new child UndecidedNode (and giving it an improved prediction).
 *
 * Pass 2:
Build new summary DHistograms on the new child UndecidedNodes
 * every row got assigned into.  Collect counts, mean, variance, min,
 * max per bin, per column.
 * 
 *
 * The result is a set of DHistogram arrays; one DHistogram array for each
 * unique 'leaf' in the tree being histogramed in parallel.  These have node
 * ID's (nids) from 'leaf' to 'tree._len'.  Each DHistogram array is for all
 * the columns in that 'leaf'.
 *
 * 
The other result is a prediction "score" for the whole dataset, based on
 * the previous passes' DHistograms.
 */
public class ScoreBuildHistogram extends MRTask {
  final int   _k;    // Which tree
  final int   _ncols;// Active feature columns
  final int   _nbins;// Numerical columns: Number of bins in each histogram
  final DTree _tree; // Read-only, shared (except at the histograms in the Nodes)
  final int   _leaf; // Number of active leaves (per tree)
  // Histograms for every tree, split & active column
  DHistogram[/*tree-relative node-id*/][/*column*/] _hcs;
  final DistributionFamily _family;
  final int _weightIdx;
  final int _workIdx;
  final int _nidIdx;
  final int _treatmentIdx;

  public ScoreBuildHistogram(H2OCountedCompleter cc, int k, int ncols, int nbins, DTree tree, int leaf, DHistogram[][] hcs, DistributionFamily family, int weightIdx, int workIdx, int nidIdx, int treatmentIdx) {
    super(cc);
    _k    = k;
    _ncols= ncols;
    _nbins= nbins;
    _tree = tree;
    _leaf = leaf;
    _hcs  = hcs;
    _family = family;
    _weightIdx = weightIdx;
    _workIdx = workIdx;
    _nidIdx = nidIdx;
    _treatmentIdx = treatmentIdx;
  }

  /** Marker for already decided row. */
  static public final int DECIDED_ROW = -1;
  /** Marker for sampled out rows */
  static public final int OUT_OF_BAG = -2;
  /** Marker for a fresh tree */
  static public final int UNDECIDED_CHILD_NODE_ID = -1;

  static public final int FRESH = 0;

  static public boolean isOOBRow(int nid)     { return nid <= OUT_OF_BAG; }
  static public boolean isDecidedRow(int nid) { return nid == DECIDED_ROW; }
  static public int     oob2Nid(int oobNid)   { return -oobNid + OUT_OF_BAG; }
  static public int     nid2Oob(int nid)      { return -nid + OUT_OF_BAG; }

  @Override public void reduce( ScoreBuildHistogram sbh ) {
    // Merge histograms
    if( sbh._hcs == _hcs )
      return; // Local histograms all shared; free to merge
    // Distributed histograms need a little work
    for( int i=0; i<_hcs.length; i++ ) {
      DHistogram[] hs1 = _hcs[i], hs2 = sbh._hcs[i];
      if( hs1 == null ) _hcs[i] = hs2;
      else if( hs2 != null )
        for( int j=0; j