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

Initial support for column-wise data split (#8468)

Browse Source
This commit is contained in:
Rong Ou
2022-12-03 09:37:51 -08:00
committed by GitHub
parent c0609b98f1
commit 78d65a1928
8 changed files with 135 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

63
tests/cpp/data/test_simple_dmatrix.cc
View File

@@ -300,6 +300,69 @@ TEST(SimpleDMatrix, Slice) {
ASSERT_EQ(out->Info().num_nonzero_, ridxs.size() * kCols); // dense
}
TEST(SimpleDMatrix, SliceCol) {
size_t constexpr kRows {16};
size_t constexpr kCols {8};
size_t constexpr kClasses {3};
auto p_m = RandomDataGenerator{kRows, kCols, 0}.GenerateDMatrix(true);
auto& weights = p_m->Info().weights_.HostVector();
weights.resize(kRows);
std::iota(weights.begin(), weights.end(), 0.0f);
auto& lower = p_m->Info().labels_lower_bound_.HostVector();
auto& upper = p_m->Info().labels_upper_bound_.HostVector();
lower.resize(kRows);
upper.resize(kRows);
std::iota(lower.begin(), lower.end(), 0.0f);
std::iota(upper.begin(), upper.end(), 1.0f);
auto& margin = p_m->Info().base_margin_;
margin = decltype(p_m->Info().base_margin_){{kRows, kClasses}, GenericParameter::kCpuId};
size_t constexpr kSlicCols {4};
for (auto slice = 0; slice < 2; slice++) {
auto const slice_start = slice * kSlicCols;
std::unique_ptr<DMatrix> out { p_m->SliceCol(slice_start, kSlicCols) };
ASSERT_EQ(out->Info().labels.Size(), kRows);
ASSERT_EQ(out->Info().labels_lower_bound_.Size(), kRows);
ASSERT_EQ(out->Info().labels_upper_bound_.Size(), kRows);
ASSERT_EQ(out->Info().base_margin_.Size(), kRows * kClasses);
for (auto const &in_batch : p_m->GetBatches<SparsePage>()) {
auto in_page = in_batch.GetView();
for (auto const &out_batch : out->GetBatches<SparsePage>()) {
auto out_page = out_batch.GetView();
for (size_t i = 0; i < kRows; ++i) {
auto out_inst = out_page[i];
auto in_inst = in_page[i];
ASSERT_EQ(out_inst.size() * 2, in_inst.size()) << i;
for (size_t j = 0; j < kSlicCols; ++j) {
ASSERT_EQ(in_inst[slice_start + j].fvalue, out_inst[j].fvalue);
ASSERT_EQ(in_inst[slice_start + j].index, out_inst[j].index);
}
ASSERT_EQ(p_m->Info().labels_lower_bound_.HostVector().at(i),
out->Info().labels_lower_bound_.HostVector().at(i));
ASSERT_EQ(p_m->Info().labels_upper_bound_.HostVector().at(i),
out->Info().labels_upper_bound_.HostVector().at(i));
ASSERT_EQ(p_m->Info().weights_.HostVector().at(i), out->Info().weights_.HostVector().at(i));
auto out_margin = out->Info().base_margin_.View(GenericParameter::kCpuId);
auto in_margin = margin.View(GenericParameter::kCpuId);
for (size_t j = 0; j < kClasses; ++j) {
ASSERT_EQ(out_margin(i, j), in_margin(i, j));
}
}
}
}
ASSERT_EQ(out->Info().num_col_, out->Info().num_col_);
ASSERT_EQ(out->Info().num_row_, kRows);
ASSERT_EQ(out->Info().num_nonzero_, kRows * kSlicCols); // dense
}
}
TEST(SimpleDMatrix, SaveLoadBinary) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";