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

Fix empty partition. (#10559)

Browse Source
This commit is contained in:
Jiaming Yuan 2024-07-10 13:01:47 +08:00 committed by GitHub
parent 8e2b874b4c
commit baba3e9eb0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 16 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/common/hist_util.cc
View File

@ -201,7 +201,7 @@ void RowsWiseBuildHistKernel(Span<GradientPair const> gpair,
auto const &row_ptr = gmat.row_ptr.data(); auto const &row_ptr = gmat.row_ptr.data();
auto base_rowid = gmat.base_rowid; auto base_rowid = gmat.base_rowid;
uint32_t const *offsets = gmat.index.Offset(); std::uint32_t const *offsets = gmat.index.Offset();
// There's no feature-based compression if missing value is present. // There's no feature-based compression if missing value is present.
if (kAnyMissing) { if (kAnyMissing) {
CHECK(!offsets); CHECK(!offsets);
@ -212,8 +212,11 @@ void RowsWiseBuildHistKernel(Span<GradientPair const> gpair,
auto get_row_ptr = [&](bst_idx_t ridx) { auto get_row_ptr = [&](bst_idx_t ridx) {
return kFirstPage ? row_ptr[ridx] : row_ptr[ridx - base_rowid]; return kFirstPage ? row_ptr[ridx] : row_ptr[ridx - base_rowid];
}; };
auto get_rid = [&](bst_idx_t ridx) { return kFirstPage ? ridx : (ridx - base_rowid); }; auto get_rid = [&](bst_idx_t ridx) {
return kFirstPage ? ridx : (ridx - base_rowid);
};
CHECK_NE(row_indices.Size(), 0);
const size_t n_features = const size_t n_features =
get_row_ptr(row_indices.begin[0] + 1) - get_row_ptr(row_indices.begin[0]); get_row_ptr(row_indices.begin[0] + 1) - get_row_ptr(row_indices.begin[0]);
auto hist_data = reinterpret_cast<double *>(hist.data()); auto hist_data = reinterpret_cast<double *>(hist.data());
@ -325,19 +328,23 @@ void BuildHistDispatch(Span<GradientPair const> gpair, const RowSetCollection::E
if (contiguousBlock) { if (contiguousBlock) {
// contiguous memory access, built-in HW prefetching is enough // contiguous memory access, built-in HW prefetching is enough
if (row_indices.Size() == 0) {
return;
}
RowsWiseBuildHistKernel<false, BuildingManager>(gpair, row_indices, gmat, hist); RowsWiseBuildHistKernel<false, BuildingManager>(gpair, row_indices, gmat, hist);
} else { } else {
const RowSetCollection::Elem span1(row_indices.begin, const RowSetCollection::Elem span1(row_indices.begin, row_indices.end - no_prefetch_size);
row_indices.end - no_prefetch_size); if (span1.Size() != 0) {
const RowSetCollection::Elem span2(row_indices.end - no_prefetch_size,
row_indices.end);
RowsWiseBuildHistKernel<true, BuildingManager>(gpair, span1, gmat, hist); RowsWiseBuildHistKernel<true, BuildingManager>(gpair, span1, gmat, hist);
}
// no prefetching to avoid loading extra memory // no prefetching to avoid loading extra memory
const RowSetCollection::Elem span2(row_indices.end - no_prefetch_size, row_indices.end);
if (span2.Size() != 0) {
RowsWiseBuildHistKernel<false, BuildingManager>(gpair, span2, gmat, hist); RowsWiseBuildHistKernel<false, BuildingManager>(gpair, span2, gmat, hist);
} }
} }
} }
}
template <bool any_missing> template <bool any_missing>
void BuildHist(Span<GradientPair const> gpair, const RowSetCollection::Elem row_indices, void BuildHist(Span<GradientPair const> gpair, const RowSetCollection::Elem row_indices,