From 129fee64f32150547d331430bb4f208018ab468d Mon Sep 17 00:00:00 2001
From: tqchen <tianqi.tchen@gmail.com>
Date: Sun, 16 Nov 2014 11:38:21 -0800
Subject: [PATCH] fix regression

---
 demo/kaggle-higgs/higgs-cv.py      |  2 +-
 src/tree/updater_histmaker-inl.hpp | 22 +++++++++++++++-------
 src/utils/quantile.h               |  7 ++++++-
 3 files changed, 22 insertions(+), 9 deletions(-)

diff --git a/demo/kaggle-higgs/higgs-cv.py b/demo/kaggle-higgs/higgs-cv.py
index 3e36fa66b..1d660aa8f 100755
--- a/demo/kaggle-higgs/higgs-cv.py
+++ b/demo/kaggle-higgs/higgs-cv.py
@@ -10,7 +10,7 @@ label  = train[:,32]
 data   = train[:,1:31]
 weight = train[:,31]
 dtrain = xgb.DMatrix( data, label=label, missing = -999.0, weight=weight )
-param = {'max_depth':6, 'eta':0.1, 'silent':1, 'objective':'binary:logitraw', 'nthread':4}
+param = {'max_depth':6, 'eta':0.1, 'silent':1, 'objective':'binary:logitraw', 'nthread':4, 'updater':'grow_histmaker,prune'}
 num_round = 120
 
 print ('running cross validation, with preprocessing function')
diff --git a/src/tree/updater_histmaker-inl.hpp b/src/tree/updater_histmaker-inl.hpp
index 68ceca371..8cade6313 100644
--- a/src/tree/updater_histmaker-inl.hpp
+++ b/src/tree/updater_histmaker-inl.hpp
@@ -53,7 +53,7 @@ class HistMaker: public IUpdater {
                     const std::vector<bst_gpair> &gpair,
                     const BoosterInfo &info,
                     const bst_uint ridx) {
-      unsigned i = std::lower_bound(cut, cut + size, fv) - cut;
+      unsigned i = std::upper_bound(cut, cut + size, fv) - cut;      
       utils::Assert(i < size, "maximum value must be in cut");
       data[i].Add(gpair, info, ridx);
     }
@@ -155,7 +155,7 @@ class HistMaker: public IUpdater {
                       RegTree *p_tree) {
     this->InitData(gpair, *p_fmat, info.root_index, *p_tree);
     this->UpdateNode2WorkIndex(*p_tree);
-    for (int depth = 0; depth < param.max_depth; ++depth) {      
+    for (int depth = 0; depth < param.max_depth; ++depth) {
       this->FindSplit(depth, gpair, p_fmat, info, p_tree);
       this->UpdateQueueExpand(*p_tree);
       this->UpdateNode2WorkIndex(*p_tree);
@@ -278,6 +278,7 @@ class HistMaker: public IUpdater {
                              SplitEntry *best,
                              TStats *left_sum) {
     if (hist.size == 0) return;
+
     double root_gain = node_sum.CalcGain(param);
     TStats s(param), c(param);
     for (bst_uint i = 0; i < hist.size; ++i) {
@@ -383,7 +384,7 @@ class QuantileHistMaker: public HistMaker<TStats> {
     sketchs.resize(this->qexpand.size() * tree.param.num_feature);
     for (size_t i = 0; i < sketchs.size(); ++i) {
       sketchs[i].Init(info.num_row, this->param.sketch_eps);
-    }    
+    }
     // start accumulating statistics
     utils::IIterator<RowBatch> *iter = p_fmat->RowIterator();
     iter->BeforeFirst();
@@ -453,14 +454,21 @@ class QuantileHistMaker: public HistMaker<TStats> {
     this->wspace.rptr.clear();
     this->wspace.rptr.push_back(0);
     for (size_t wid = 0; wid < this->qexpand.size(); ++wid) {
-      for (size_t fid = 0; fid < tree.param.num_feature; ++fid) {
+      for (int fid = 0; fid < tree.param.num_feature; ++fid) {
         const WXQSketch::Summary a = summary_array[wid * tree.param.num_feature + fid];
-        for (size_t i = 0; i < a.size; ++i) {
-          bst_float cpt = a.data[i].value + rt_eps;
-          if (i == 0 || cpt > this->wspace.cut.back()) {
+        for (size_t i = 1; i < a.size; ++i) {
+          bst_float cpt = a.data[i].value - rt_eps;
+          if (i == 1 || cpt > this->wspace.cut.back()) {
             this->wspace.cut.push_back(cpt);
           }
         }
+        // push a value that is greater than anything
+        if (a.size != 0) {
+          bst_float cpt = a.data[a.size - 1].value;
+          // this must be bigger than last value in a scale
+          bst_float last = cpt + fabs(cpt);
+          this->wspace.cut.push_back(last);
+        }
         this->wspace.rptr.push_back(this->wspace.cut.size());
       }
       // reserve last value for global statistics
diff --git a/src/utils/quantile.h b/src/utils/quantile.h
index a3b8c18dd..53117f28b 100644
--- a/src/utils/quantile.h
+++ b/src/utils/quantile.h
@@ -10,6 +10,7 @@
 #include <cstring>
 #include <algorithm>
 #include <iostream>
+#include "./io.h"
 #include "./utils.h"
 
 namespace xgboost {
@@ -481,7 +482,11 @@ class QuantileSketchTemplate {
   /*! \brief same as summary, but use STL to backup the space */
   struct SummaryContainer : public Summary {
     std::vector<Entry> space;
-    SummaryContainer(void) : Summary(NULL, 0) { 
+    explicit SummaryContainer(void) : Summary(NULL, 0) { 
+    }
+    explicit SummaryContainer(const SummaryContainer &src) : Summary(NULL, src.size) { 
+      this->space = src.space;
+      this->data = BeginPtr(this->space);
     }
     /*! \brief reserve space for summary */
     inline void Reserve(size_t size) {