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

Use DataSplitMode to configure data loading (#8434)

Browse Source 
* Use `DataSplitMode` to configure data loading
This commit is contained in:
Rong Ou
2022-11-08 00:21:50 -08:00
committed by GitHub
parent 0d3da9869c
commit 8e76f5f595
13 changed files with 46 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/cpp/data/test_data.cc
View File

@@ -143,7 +143,7 @@ TEST(DMatrix, Uri) {
// EXPECT_THROW(dmat.reset(DMatrix::Load(path, false, true)), dmlc::Error);
std::string uri = path + "?format=csv";
dmat.reset(DMatrix::Load(uri, false, true));
dmat.reset(DMatrix::Load(uri, false, DataSplitMode::kRow));
ASSERT_EQ(dmat->Info().num_col_, kCols);
ASSERT_EQ(dmat->Info().num_row_, kRows);

2
tests/cpp/data/test_metainfo.cc
View File

@@ -175,7 +175,7 @@ TEST(MetaInfo, LoadQid) {
os.set_stream(nullptr);
}
std::unique_ptr<xgboost::DMatrix> dmat(
xgboost::DMatrix::Load(tmp_file, true, false, "libsvm"));
xgboost::DMatrix::Load(tmp_file, true, xgboost::DataSplitMode::kNone, "libsvm"));
const xgboost::MetaInfo& info = dmat->Info();
const std::vector<xgboost::bst_uint> expected_group_ptr{0, 4, 8, 12};

10
tests/cpp/data/test_simple_dmatrix.cc
View File

@@ -15,7 +15,7 @@ TEST(SimpleDMatrix, MetaInfo) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, true, false);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, true, xgboost::DataSplitMode::kNone);
// Test the metadata that was parsed
EXPECT_EQ(dmat->Info().num_row_, 2);
@@ -30,7 +30,7 @@ TEST(SimpleDMatrix, RowAccess) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, false, false);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, false, xgboost::DataSplitMode::kNone);
// Loop over the batches and count the records
int64_t row_count = 0;
@@ -53,7 +53,7 @@ TEST(SimpleDMatrix, ColAccessWithoutBatches) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, true, false);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(tmp_file, true, xgboost::DataSplitMode::kNone);
ASSERT_TRUE(dmat->SingleColBlock());
@@ -304,12 +304,12 @@ TEST(SimpleDMatrix, SaveLoadBinary) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
xgboost::DMatrix * dmat = xgboost::DMatrix::Load(tmp_file, true, false);
xgboost::DMatrix * dmat = xgboost::DMatrix::Load(tmp_file, true, xgboost::DataSplitMode::kNone);
data::SimpleDMatrix *simple_dmat = dynamic_cast<data::SimpleDMatrix*>(dmat);
const std::string tmp_binfile = tempdir.path + "/csr_source.binary";
simple_dmat->SaveToLocalFile(tmp_binfile);
xgboost::DMatrix * dmat_read = xgboost::DMatrix::Load(tmp_binfile, true, false);
xgboost::DMatrix * dmat_read = xgboost::DMatrix::Load(tmp_binfile, true, xgboost::DataSplitMode::kNone);
EXPECT_EQ(dmat->Info().num_col_, dmat_read->Info().num_col_);
EXPECT_EQ(dmat->Info().num_row_, dmat_read->Info().num_row_);

4
tests/cpp/data/test_sparse_page_dmatrix.cc
View File

@@ -109,7 +109,7 @@ TEST(SparsePageDMatrix, MetaInfo) {
CreateBigTestData(tmp_file, kEntries);
xgboost::DMatrix *dmat = xgboost::DMatrix::Load(
tmp_file + "#" + tmp_file + ".cache", false, false);
tmp_file + "#" + tmp_file + ".cache", false, xgboost::DataSplitMode::kNone);
// Test the metadata that was parsed
EXPECT_EQ(dmat->Info().num_row_, 8ul);
@@ -137,7 +137,7 @@ TEST(SparsePageDMatrix, ColAccess) {
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
xgboost::DMatrix *dmat =
xgboost::DMatrix::Load(tmp_file + "#" + tmp_file + ".cache", true, false);
xgboost::DMatrix::Load(tmp_file + "#" + tmp_file + ".cache", true, xgboost::DataSplitMode::kNone);
// Loop over the batches and assert the data is as expected
size_t iter = 0;

2
tests/cpp/data/test_sparse_page_dmatrix.cu
View File

@@ -12,7 +12,7 @@ TEST(SparsePageDMatrix, EllpackPage) {
dmlc::TemporaryDirectory tempdir;
const std::string tmp_file = tempdir.path + "/simple.libsvm";
CreateSimpleTestData(tmp_file);
DMatrix* dmat = DMatrix::Load(tmp_file + "#" + tmp_file + ".cache", true, false);
DMatrix* dmat = DMatrix::Load(tmp_file + "#" + tmp_file + ".cache", true, DataSplitMode::kNone);
// Loop over the batches and assert the data is as expected
size_t n = 0;