Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Improve operation efficiency for single predict (#5016)

Browse Source 
* Improve operation efficiency for single predict
This commit is contained in:
KaiJin Ji
2019-11-10 02:01:28 +08:00
committed by Jiaming Yuan
parent 374648c21a
commit 1733c9e8f7
2 changed files with 39 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
src/predictor/cpu_predictor.cc
View File

@@ -63,24 +63,26 @@ class CPUPredictor : public Predictor {
// Pull to host before entering omp block, as this is not thread safe.
batch.data.HostVector();
batch.offset.HostVector();
if (nsize >= kUnroll) {
#pragma omp parallel for schedule(static)
for (bst_omp_uint i = 0; i < nsize - rest; i += kUnroll) {
const int tid = omp_get_thread_num();
RegTree::FVec& feats = thread_temp[tid];
int64_t ridx[kUnroll];
SparsePage::Inst inst[kUnroll];
for (int k = 0; k < kUnroll; ++k) {
ridx[k] = static_cast<int64_t>(batch.base_rowid + i + k);
}
for (int k = 0; k < kUnroll; ++k) {
inst[k] = batch[i + k];
}
for (int k = 0; k < kUnroll; ++k) {
for (int gid = 0; gid < num_group; ++gid) {
const size_t offset = ridx[k] * num_group + gid;
preds[offset] += this->PredValue(
inst[k], model.trees, model.tree_info, gid,
info.GetRoot(ridx[k]), &feats, tree_begin, tree_end);
for (bst_omp_uint i = 0; i < nsize - rest; i += kUnroll) {
const int tid = omp_get_thread_num();
RegTree::FVec& feats = thread_temp[tid];
int64_t ridx[kUnroll];
SparsePage::Inst inst[kUnroll];
for (int k = 0; k < kUnroll; ++k) {
ridx[k] = static_cast<int64_t>(batch.base_rowid + i + k);
}
for (int k = 0; k < kUnroll; ++k) {
inst[k] = batch[i + k];
}
for (int k = 0; k < kUnroll; ++k) {
for (int gid = 0; gid < num_group; ++gid) {
const size_t offset = ridx[k] * num_group + gid;
preds[offset] += this->PredValue(
inst[k], model.trees, model.tree_info, gid,
info.GetRoot(ridx[k]), &feats, tree_begin, tree_end);
}
}
}
}