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

Sort sparse page index when constructing DMatrix. (#7731)

Browse Source
This commit is contained in:
Jiaming Yuan
2022-03-16 18:01:05 +08:00
committed by GitHub
parent 613ec36c5a
commit e78a38b837
4 changed files with 67 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
tests/cpp/data/test_data.cc
View File

@@ -86,6 +86,30 @@ TEST(SparsePage, PushCSCAfterTranspose) {
}
}
TEST(SparsePage, SortIndices) {
auto p_fmat = RandomDataGenerator{100, 10, 0.6}.GenerateDMatrix();
auto n_threads = common::OmpGetNumThreads(0);
SparsePage copy;
for (auto const& page : p_fmat->GetBatches<SparsePage>()) {
ASSERT_TRUE(page.IsIndicesSorted(n_threads));
copy.Push(page);
}
ASSERT_TRUE(copy.IsIndicesSorted(n_threads));
for (size_t ridx = 0; ridx < copy.Size(); ++ridx) {
auto beg = copy.offset.HostVector()[ridx];
auto end = copy.offset.HostVector()[ridx + 1];
auto& h_data = copy.data.HostVector();
if (end - beg >= 2) {
std::swap(h_data[beg], h_data[end - 1]);
}
}
ASSERT_FALSE(copy.IsIndicesSorted(n_threads));
copy.SortIndices(n_threads);
ASSERT_TRUE(copy.IsIndicesSorted(n_threads));
}
TEST(DMatrix, Uri) {
size_t constexpr kRows {16};
size_t constexpr kCols {8};