Distributed optimizations for 'hist' method with CPUs (#5557)

Co-authored-by: SHVETS, KIRILL <kirill.shvets@intel.com>
This commit is contained in:
ShvetsKS
2020-05-20 06:03:03 +03:00
committed by GitHub
parent e21a608552
commit dd01e4ba8d
4 changed files with 451 additions and 94 deletions

View File

@@ -14,7 +14,7 @@
#include "xgboost/json.h"
#include "./param.h"
#include "../common/io.h"
#include "../common/timer.h"
namespace xgboost {
namespace tree {
@@ -25,6 +25,7 @@ class TreePruner: public TreeUpdater {
public:
TreePruner() {
syncher_.reset(TreeUpdater::Create("sync", tparam_));
pruner_monitor_.Init("TreePruner");
}
char const* Name() const override {
return "prune";
@@ -52,6 +53,7 @@ class TreePruner: public TreeUpdater {
void Update(HostDeviceVector<GradientPair> *gpair,
DMatrix *p_fmat,
const std::vector<RegTree*> &trees) override {
pruner_monitor_.Start("PrunerUpdate");
// rescale learning rate according to size of trees
float lr = param_.learning_rate;
param_.learning_rate = lr / trees.size();
@@ -60,6 +62,7 @@ class TreePruner: public TreeUpdater {
}
param_.learning_rate = lr;
syncher_->Update(gpair, p_fmat, trees);
pruner_monitor_.Stop("PrunerUpdate");
}
private:
@@ -105,6 +108,7 @@ class TreePruner: public TreeUpdater {
std::unique_ptr<TreeUpdater> syncher_;
// training parameter
TrainParam param_;
common::Monitor pruner_monitor_;
};
XGBOOST_REGISTER_TREE_UPDATER(TreePruner, "prune")

View File

@@ -55,12 +55,13 @@ void QuantileHistMaker::Update(HostDeviceVector<GradientPair> *gpair,
DMatrix *dmat,
const std::vector<RegTree *> &trees) {
if (dmat != p_last_dmat_ || is_gmat_initialized_ == false) {
updater_monitor_.Start("GmatInitialization");
gmat_.Init(dmat, static_cast<uint32_t>(param_.max_bin));
column_matrix_.Init(gmat_, param_.sparse_threshold);
if (param_.enable_feature_grouping > 0) {
gmatb_.Init(gmat_, column_matrix_, param_);
}
updater_monitor_.Stop("GmatInitialization");
// A proper solution is puting cut matrix in DMatrix, see:
// https://github.com/dmlc/xgboost/issues/5143
is_gmat_initialized_ = true;
@@ -76,10 +77,18 @@ void QuantileHistMaker::Update(HostDeviceVector<GradientPair> *gpair,
std::move(pruner_),
std::unique_ptr<SplitEvaluator>(spliteval_->GetHostClone()),
int_constraint_, dmat));
if (rabit::IsDistributed()) {
builder_->SetHistSynchronizer(new DistributedHistSynchronizer());
builder_->SetHistRowsAdder(new DistributedHistRowsAdder());
} else {
builder_->SetHistSynchronizer(new BatchHistSynchronizer());
builder_->SetHistRowsAdder(new BatchHistRowsAdder());
}
}
for (auto tree : trees) {
builder_->Update(gmat_, gmatb_, column_matrix_, gpair, dmat, tree);
}
param_.learning_rate = lr;
p_last_dmat_ = dmat;
@@ -95,43 +104,151 @@ bool QuantileHistMaker::UpdatePredictionCache(
}
}
void QuantileHistMaker::Builder::SyncHistograms(
int starting_index,
int sync_count,
RegTree *p_tree) {
builder_monitor_.Start("SyncHistograms");
const bool isDistributed = rabit::IsDistributed();
const size_t nbins = hist_builder_.GetNumBins();
common::BlockedSpace2d space(nodes_for_explicit_hist_build_.size(), [&](size_t node) {
void BatchHistSynchronizer::SyncHistograms(QuantileHistMaker::Builder* builder,
int starting_index,
int sync_count,
RegTree *p_tree) {
builder->builder_monitor_.Start("SyncHistograms");
const size_t nbins = builder->hist_builder_.GetNumBins();
common::BlockedSpace2d space(builder->nodes_for_explicit_hist_build_.size(), [&](size_t node) {
return nbins;
}, 1024);
common::ParallelFor2d(space, this->nthread_, [&](size_t node, common::Range1d r) {
const auto entry = nodes_for_explicit_hist_build_[node];
auto this_hist = hist_[entry.nid];
common::ParallelFor2d(space, builder->nthread_, [&](size_t node, common::Range1d r) {
const auto entry = builder->nodes_for_explicit_hist_build_[node];
auto this_hist = builder->hist_[entry.nid];
// Merging histograms from each thread into once
hist_buffer_.ReduceHist(node, r.begin(), r.end());
if (!(*p_tree)[entry.nid].IsRoot() && entry.sibling_nid > -1 && !isDistributed) {
auto parent_hist = hist_[(*p_tree)[entry.nid].Parent()];
auto sibling_hist = hist_[entry.sibling_nid];
builder->hist_buffer_.ReduceHist(node, r.begin(), r.end());
if (!(*p_tree)[entry.nid].IsRoot() && entry.sibling_nid > -1) {
const size_t parent_id = (*p_tree)[entry.nid].Parent();
auto parent_hist = builder->hist_[parent_id];
auto sibling_hist = builder->hist_[entry.sibling_nid];
SubtractionHist(sibling_hist, parent_hist, this_hist, r.begin(), r.end());
}
});
builder->builder_monitor_.Stop("SyncHistograms");
}
if (isDistributed) {
this->histred_.Allreduce(hist_[starting_index].data(), hist_builder_.GetNumBins() * sync_count);
// use Subtraction Trick
for (auto const& node : nodes_for_subtraction_trick_) {
SubtractionTrick(hist_[node.nid], hist_[node.sibling_nid],
hist_[(*p_tree)[node.nid].Parent()]);
void DistributedHistSynchronizer::SyncHistograms(QuantileHistMaker::Builder* builder,
int starting_index,
int sync_count,
RegTree *p_tree) {
builder->builder_monitor_.Start("SyncHistograms");
const size_t nbins = builder->hist_builder_.GetNumBins();
common::BlockedSpace2d space(builder->nodes_for_explicit_hist_build_.size(), [&](size_t node) {
return nbins;
}, 1024);
common::ParallelFor2d(space, builder->nthread_, [&](size_t node, common::Range1d r) {
const auto entry = builder->nodes_for_explicit_hist_build_[node];
auto this_hist = builder->hist_[entry.nid];
// Merging histograms from each thread into once
builder->hist_buffer_.ReduceHist(node, r.begin(), r.end());
// Store posible parent node
auto this_local = builder->hist_local_worker_[entry.nid];
CopyHist(this_local, this_hist, r.begin(), r.end());
if (!(*p_tree)[entry.nid].IsRoot() && entry.sibling_nid > -1) {
const size_t parent_id = (*p_tree)[entry.nid].Parent();
auto parent_hist = builder->hist_local_worker_[parent_id];
auto sibling_hist = builder->hist_[entry.sibling_nid];
SubtractionHist(sibling_hist, parent_hist, this_hist, r.begin(), r.end());
// Store posible parent node
auto sibling_local = builder->hist_local_worker_[entry.sibling_nid];
CopyHist(sibling_local, sibling_hist, r.begin(), r.end());
}
});
builder->builder_monitor_.Start("SyncHistogramsAllreduce");
builder->histred_.Allreduce(builder->hist_[starting_index].data(),
builder->hist_builder_.GetNumBins() * sync_count);
builder->builder_monitor_.Stop("SyncHistogramsAllreduce");
ParallelSubtractionHist(builder, space, builder->nodes_for_explicit_hist_build_, p_tree);
common::BlockedSpace2d space2(builder->nodes_for_subtraction_trick_.size(), [&](size_t node) {
return nbins;
}, 1024);
ParallelSubtractionHist(builder, space2, builder->nodes_for_subtraction_trick_, p_tree);
builder->builder_monitor_.Stop("SyncHistograms");
}
void DistributedHistSynchronizer::ParallelSubtractionHist(QuantileHistMaker::Builder* builder,
const common::BlockedSpace2d& space,
const std::vector<QuantileHistMaker::Builder::ExpandEntry>& nodes,
const RegTree * p_tree) {
common::ParallelFor2d(space, builder->nthread_, [&](size_t node, common::Range1d r) {
const auto entry = nodes[node];
if (!((*p_tree)[entry.nid].IsLeftChild())) {
auto this_hist = builder->hist_[entry.nid];
if (!(*p_tree)[entry.nid].IsRoot() && entry.sibling_nid > -1) {
auto parent_hist = builder->hist_[(*p_tree)[entry.nid].Parent()];
auto sibling_hist = builder->hist_[entry.sibling_nid];
SubtractionHist(this_hist, parent_hist, sibling_hist, r.begin(), r.end());
}
}
});
}
void BatchHistRowsAdder::AddHistRows(QuantileHistMaker::Builder* builder,
int *starting_index, int *sync_count,
RegTree *p_tree) {
builder->builder_monitor_.Start("AddHistRows");
for (auto const& entry : builder->nodes_for_explicit_hist_build_) {
int nid = entry.nid;
builder->hist_.AddHistRow(nid);
(*starting_index) = std::min(nid, (*starting_index));
}
(*sync_count) = builder->nodes_for_explicit_hist_build_.size();
for (auto const& node : builder->nodes_for_subtraction_trick_) {
builder->hist_.AddHistRow(node.nid);
}
builder_monitor_.Stop("SyncHistograms");
builder->builder_monitor_.Stop("AddHistRows");
}
void DistributedHistRowsAdder::AddHistRows(QuantileHistMaker::Builder* builder,
int *starting_index, int *sync_count,
RegTree *p_tree) {
builder->builder_monitor_.Start("AddHistRows");
const size_t explicit_size = builder->nodes_for_explicit_hist_build_.size();
const size_t subtaction_size = builder->nodes_for_subtraction_trick_.size();
std::vector<int> merged_node_ids(explicit_size + subtaction_size);
for (size_t i = 0; i < explicit_size; ++i) {
merged_node_ids[i] = builder->nodes_for_explicit_hist_build_[i].nid;
}
for (size_t i = 0; i < subtaction_size; ++i) {
merged_node_ids[explicit_size + i] =
builder->nodes_for_subtraction_trick_[i].nid;
}
std::sort(merged_node_ids.begin(), merged_node_ids.end());
int n_left = 0;
for (auto const& nid : merged_node_ids) {
if ((*p_tree)[nid].IsLeftChild()) {
builder->hist_.AddHistRow(nid);
(*starting_index) = std::min(nid, (*starting_index));
n_left++;
builder->hist_local_worker_.AddHistRow(nid);
}
}
for (auto const& nid : merged_node_ids) {
if (!((*p_tree)[nid].IsLeftChild())) {
builder->hist_.AddHistRow(nid);
builder->hist_local_worker_.AddHistRow(nid);
}
}
(*sync_count) = std::max(1, n_left);
builder->builder_monitor_.Stop("AddHistRows");
}
void QuantileHistMaker::Builder::SetHistSynchronizer(HistSynchronizer* sync) {
hist_synchronizer_.reset(sync);
}
void QuantileHistMaker::Builder::SetHistRowsAdder(HistRowsAdder* adder) {
hist_rows_adder_.reset(adder);
}
void QuantileHistMaker::Builder::BuildHistogramsLossGuide(
@@ -152,30 +269,11 @@ void QuantileHistMaker::Builder::BuildHistogramsLossGuide(
int starting_index = std::numeric_limits<int>::max();
int sync_count = 0;
AddHistRows(&starting_index, &sync_count);
hist_rows_adder_->AddHistRows(this, &starting_index, &sync_count, p_tree);
BuildLocalHistograms(gmat, gmatb, p_tree, gpair_h);
SyncHistograms(starting_index, sync_count, p_tree);
hist_synchronizer_->SyncHistograms(this, starting_index, sync_count, p_tree);
}
void QuantileHistMaker::Builder::AddHistRows(int *starting_index, int *sync_count) {
builder_monitor_.Start("AddHistRows");
for (auto const& entry : nodes_for_explicit_hist_build_) {
int nid = entry.nid;
hist_.AddHistRow(nid);
(*starting_index) = std::min(nid, (*starting_index));
}
(*sync_count) = nodes_for_explicit_hist_build_.size();
for (auto const& node : nodes_for_subtraction_trick_) {
hist_.AddHistRow(node.nid);
}
builder_monitor_.Stop("AddHistRows");
}
void QuantileHistMaker::Builder::BuildLocalHistograms(
const GHistIndexMatrix &gmat,
const GHistIndexBlockMatrix &gmatb,
@@ -184,6 +282,7 @@ void QuantileHistMaker::Builder::BuildLocalHistograms(
builder_monitor_.Start("BuildLocalHistograms");
const size_t n_nodes = nodes_for_explicit_hist_build_.size();
// create space of size (# rows in each node)
common::BlockedSpace2d space(n_nodes, [&](size_t node) {
const int32_t nid = nodes_for_explicit_hist_build_[node].nid;
@@ -305,31 +404,28 @@ void QuantileHistMaker::Builder::SplitSiblings(const std::vector<ExpandEntry>& n
std::vector<ExpandEntry>* small_siblings,
std::vector<ExpandEntry>* big_siblings,
RegTree *p_tree) {
builder_monitor_.Start("SplitSiblings");
for (auto const& entry : nodes) {
int nid = entry.nid;
RegTree::Node &node = (*p_tree)[nid];
if (rabit::IsDistributed()) {
if (node.IsRoot() || node.IsLeftChild()) {
small_siblings->push_back(entry);
} else {
big_siblings->push_back(entry);
}
if (node.IsRoot()) {
small_siblings->push_back(entry);
} else {
if (!node.IsRoot() && node.IsLeftChild() &&
(row_set_collection_[nid].Size() <
row_set_collection_[(*p_tree)[node.Parent()].RightChild()].Size())) {
const int32_t left_id = (*p_tree)[node.Parent()].LeftChild();
const int32_t right_id = (*p_tree)[node.Parent()].RightChild();
if (nid == left_id && row_set_collection_[left_id ].Size() <
row_set_collection_[right_id].Size()) {
small_siblings->push_back(entry);
} else if (!node.IsRoot() && !node.IsLeftChild() &&
(row_set_collection_[nid].Size() <=
row_set_collection_[(*p_tree)[node.Parent()].LeftChild()].Size())) {
small_siblings->push_back(entry);
} else if (node.IsRoot()) {
} else if (nid == right_id && row_set_collection_[right_id].Size() <=
row_set_collection_[left_id ].Size()) {
small_siblings->push_back(entry);
} else {
big_siblings->push_back(entry);
}
}
}
builder_monitor_.Stop("SplitSiblings");
}
void QuantileHistMaker::Builder::ExpandWithDepthWise(
@@ -350,17 +446,16 @@ void QuantileHistMaker::Builder::ExpandWithDepthWise(
int starting_index = std::numeric_limits<int>::max();
int sync_count = 0;
std::vector<ExpandEntry> temp_qexpand_depth;
SplitSiblings(qexpand_depth_wise_, &nodes_for_explicit_hist_build_,
&nodes_for_subtraction_trick_, p_tree);
AddHistRows(&starting_index, &sync_count);
hist_rows_adder_->AddHistRows(this, &starting_index, &sync_count, p_tree);
BuildLocalHistograms(gmat, gmatb, p_tree, gpair_h);
SyncHistograms(starting_index, sync_count, p_tree);
hist_synchronizer_->SyncHistograms(this, starting_index, sync_count, p_tree);
BuildNodeStats(gmat, p_fmat, p_tree, gpair_h);
EvaluateAndApplySplits(gmat, column_matrix, p_tree, &num_leaves, depth, &timestamp,
&temp_qexpand_depth);
// clean up
qexpand_depth_wise_.clear();
nodes_for_subtraction_trick_.clear();
@@ -381,7 +476,7 @@ void QuantileHistMaker::Builder::ExpandWithLossGuide(
DMatrix* p_fmat,
RegTree* p_tree,
const std::vector<GradientPair>& gpair_h) {
builder_monitor_.Start("ExpandWithLossGuide");
unsigned timestamp = 0;
int num_leaves = 0;
@@ -424,15 +519,10 @@ void QuantileHistMaker::Builder::ExpandWithLossGuide(
ExpandEntry right_node(cright, cleft, p_tree->GetDepth(cright),
0.0f, timestamp++);
if (rabit::IsDistributed()) {
// in distributed mode, we need to keep consistent across workers
if (row_set_collection_[cleft].Size() < row_set_collection_[cright].Size()) {
BuildHistogramsLossGuide(left_node, gmat, gmatb, p_tree, gpair_h);
} else {
if (row_set_collection_[cleft].Size() < row_set_collection_[cright].Size()) {
BuildHistogramsLossGuide(left_node, gmat, gmatb, p_tree, gpair_h);
} else {
BuildHistogramsLossGuide(right_node, gmat, gmatb, p_tree, gpair_h);
}
BuildHistogramsLossGuide(right_node, gmat, gmatb, p_tree, gpair_h);
}
this->InitNewNode(cleft, gmat, gpair_h, *p_fmat, *p_tree);
@@ -452,6 +542,7 @@ void QuantileHistMaker::Builder::ExpandWithLossGuide(
++num_leaves; // give two and take one, as parent is no longer a leaf
}
}
builder_monitor_.Stop("ExpandWithLossGuide");
}
void QuantileHistMaker::Builder::Update(const GHistIndexMatrix& gmat,
@@ -468,7 +559,6 @@ void QuantileHistMaker::Builder::Update(const GHistIndexMatrix& gmat,
interaction_constraints_.Reset();
this->InitData(gmat, gpair_h, *p_fmat, *p_tree);
if (param_.grow_policy == TrainParam::kLossGuide) {
ExpandWithLossGuide(gmat, gmatb, column_matrix, p_fmat, p_tree, gpair_h);
} else {
@@ -480,7 +570,6 @@ void QuantileHistMaker::Builder::Update(const GHistIndexMatrix& gmat,
p_tree->Stat(nid).base_weight = snode_[nid].weight;
p_tree->Stat(nid).sum_hess = static_cast<float>(snode_[nid].stats.sum_hess);
}
pruner_->Update(gpair, p_fmat, std::vector<RegTree*>{p_tree});
builder_monitor_.Stop("Update");
@@ -615,6 +704,7 @@ void QuantileHistMaker::Builder::InitData(const GHistIndexMatrix& gmat,
// initialize histogram collection
uint32_t nbins = gmat.cut.Ptrs().back();
hist_.Init(nbins);
hist_local_worker_.Init(nbins);
hist_buffer_.Init(nbins);
// initialize histogram builder
@@ -1026,18 +1116,15 @@ void QuantileHistMaker::Builder::ApplySplit(const std::vector<ExpandEntry> nodes
const HistCollection& hist,
RegTree* p_tree) {
builder_monitor_.Start("ApplySplit");
// 1. Find split condition for each split
const size_t n_nodes = nodes.size();
std::vector<int32_t> split_conditions;
FindSplitConditions(nodes, *p_tree, gmat, &split_conditions);
// 2.1 Create a blocked space of size SUM(samples in each node)
common::BlockedSpace2d space(n_nodes, [&](size_t node_in_set) {
int32_t nid = nodes[node_in_set].nid;
return row_set_collection_[nid].Size();
}, kPartitionBlockSize);
// 2.2 Initialize the partition builder
// allocate buffers for storage intermediate results by each thread
partition_builder_.Init(space.Size(), n_nodes, [&](size_t node_in_set) {
@@ -1046,7 +1133,6 @@ void QuantileHistMaker::Builder::ApplySplit(const std::vector<ExpandEntry> nodes
const size_t n_tasks = size / kPartitionBlockSize + !!(size % kPartitionBlockSize);
return n_tasks;
});
// 2.3 Split elements of row_set_collection_ to left and right child-nodes for each node
// Store results in intermediate buffers from partition_builder_
common::ParallelFor2d(space, this->nthread_, [&](size_t node_in_set, common::Range1d r) {
@@ -1068,7 +1154,6 @@ void QuantileHistMaker::Builder::ApplySplit(const std::vector<ExpandEntry> nodes
CHECK(false); // no default behavior
}
});
// 3. Compute offsets to copy blocks of row-indexes
// from partition_builder_ to row_set_collection_
partition_builder_.CalculateRowOffsets();
@@ -1080,10 +1165,8 @@ void QuantileHistMaker::Builder::ApplySplit(const std::vector<ExpandEntry> nodes
partition_builder_.MergeToArray(node_in_set, r.begin(),
const_cast<size_t*>(row_set_collection_[nid].begin));
});
// 5. Add info about splits into row_set_collection_
AddSplitsToRowSet(nodes, p_tree);
builder_monitor_.Stop("ApplySplit");
}

View File

@@ -78,10 +78,14 @@ using xgboost::common::GHistBuilder;
using xgboost::common::ColumnMatrix;
using xgboost::common::Column;
class HistSynchronizer;
class HistRowsAdder;
/*! \brief construct a tree using quantized feature values */
class QuantileHistMaker: public TreeUpdater {
public:
QuantileHistMaker() = default;
QuantileHistMaker() {
updater_monitor_.Init("QuantileHistMaker");
}
void Configure(const Args& args) override;
void Update(HostDeviceVector<GradientPair>* gpair,
@@ -105,6 +109,12 @@ class QuantileHistMaker: public TreeUpdater {
}
protected:
friend class HistSynchronizer;
friend class BatchHistSynchronizer;
friend class DistributedHistSynchronizer;
friend class HistRowsAdder;
friend class BatchHistRowsAdder;
friend class DistributedHistRowsAdder;
// training parameter
TrainParam param_;
// quantized data matrix
@@ -174,8 +184,16 @@ class QuantileHistMaker: public TreeUpdater {
bool UpdatePredictionCache(const DMatrix* data,
HostDeviceVector<bst_float>* p_out_preds);
void SetHistSynchronizer(HistSynchronizer* sync);
void SetHistRowsAdder(HistRowsAdder* adder);
protected:
friend class HistSynchronizer;
friend class BatchHistSynchronizer;
friend class DistributedHistSynchronizer;
friend class HistRowsAdder;
friend class BatchHistRowsAdder;
friend class DistributedHistRowsAdder;
/* tree growing policies */
struct ExpandEntry {
static const int kRootNid = 0;
@@ -259,8 +277,6 @@ class QuantileHistMaker: public TreeUpdater {
RegTree *p_tree,
const std::vector<GradientPair> &gpair_h);
void AddHistRows(int *starting_index, int *sync_count);
void BuildHistogramsLossGuide(
ExpandEntry entry,
const GHistIndexMatrix &gmat,
@@ -276,9 +292,9 @@ class QuantileHistMaker: public TreeUpdater {
std::vector<ExpandEntry>* big_siblings,
RegTree *p_tree);
void SyncHistograms(int starting_index,
int sync_count,
RegTree *p_tree);
void ParallelSubtractionHist(const common::BlockedSpace2d& space,
const std::vector<ExpandEntry>& nodes,
const RegTree * p_tree);
void BuildNodeStats(const GHistIndexMatrix &gmat,
DMatrix *p_fmat,
@@ -316,7 +332,6 @@ class QuantileHistMaker: public TreeUpdater {
return lhs.loss_chg < rhs.loss_chg; // favor large loss_chg
}
}
// --data fields--
const TrainParam& param_;
// number of omp thread used during training
@@ -331,6 +346,8 @@ class QuantileHistMaker: public TreeUpdater {
std::vector<NodeEntry> snode_;
/*! \brief culmulative histogram of gradients. */
HistCollection hist_;
/*! \brief culmulative local parent histogram of gradients. */
HistCollection hist_local_worker_;
/*! \brief feature with least # of bins. to be used for dense specialization
of InitNewNode() */
uint32_t fid_least_bins_;
@@ -367,14 +384,62 @@ class QuantileHistMaker: public TreeUpdater {
common::Monitor builder_monitor_;
common::ParallelGHistBuilder hist_buffer_;
rabit::Reducer<GradStats, GradStats::Reduce> histred_;
std::unique_ptr<HistSynchronizer> hist_synchronizer_;
std::unique_ptr<HistRowsAdder> hist_rows_adder_;
};
common::Monitor updater_monitor_;
std::unique_ptr<Builder> builder_;
std::unique_ptr<TreeUpdater> pruner_;
std::unique_ptr<SplitEvaluator> spliteval_;
FeatureInteractionConstraintHost int_constraint_;
};
class HistSynchronizer {
public:
virtual void SyncHistograms(QuantileHistMaker::Builder* builder,
int starting_index,
int sync_count,
RegTree *p_tree) = 0;
};
class BatchHistSynchronizer: public HistSynchronizer {
public:
void SyncHistograms(QuantileHistMaker::Builder* builder,
int starting_index,
int sync_count,
RegTree *p_tree) override;
};
class DistributedHistSynchronizer: public HistSynchronizer {
public:
void SyncHistograms(QuantileHistMaker::Builder* builder_,
int starting_index, int sync_count, RegTree *p_tree) override;
void ParallelSubtractionHist(QuantileHistMaker::Builder* builder,
const common::BlockedSpace2d& space,
const std::vector<QuantileHistMaker::Builder::ExpandEntry>& nodes,
const RegTree * p_tree);
};
class HistRowsAdder {
public:
virtual void AddHistRows(QuantileHistMaker::Builder* builder,
int *starting_index, int *sync_count, RegTree *p_tree) = 0;
};
class BatchHistRowsAdder: public HistRowsAdder {
public:
void AddHistRows(QuantileHistMaker::Builder* builder,
int *starting_index, int *sync_count, RegTree *p_tree) override;
};
class DistributedHistRowsAdder: public HistRowsAdder {
public:
void AddHistRows(QuantileHistMaker::Builder* builder,
int *starting_index, int *sync_count, RegTree *p_tree) override;
};
} // namespace tree
} // namespace xgboost