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

Cleanup set info. (#10139)

Browse Source 
- Use the array interface internally.
- Deprecate `XGDMatrixSetDenseInfo`.
- Deprecate `XGDMatrixSetUIntInfo`.
- Move the handling of `DataType` into the deprecated C function.

---------

Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
This commit is contained in:
Jiaming Yuan
2024-03-26 23:26:24 +08:00
committed by GitHub
parent 6a7c6a8ae6
commit 230010d9a0
37 changed files with 246 additions and 268 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
tests/cpp/data/test_array_interface.cu
View File

@@ -1,10 +1,11 @@
/**
* Copyright 2021-2023, XGBoost Contributors
* Copyright 2021-2024, XGBoost Contributors
*/
#include <gtest/gtest.h>
#include <xgboost/host_device_vector.h>
#include "../helpers.h"
#include "../../../src/data/array_interface.h"
#include "../helpers.h"
namespace xgboost {

34
tests/cpp/data/test_metainfo.cc
View File

@@ -10,7 +10,6 @@
#include <memory>
#include <string>
#include "../../../src/common/version.h"
#include "../filesystem.h" // dmlc::TemporaryDirectory
#include "../helpers.h" // for GMockTHrow
#include "xgboost/base.h"
@@ -23,23 +22,22 @@ TEST(MetaInfo, GetSet) {
double double2[2] = {1.0, 2.0};
EXPECT_EQ(info.labels.Size(), 0);
info.SetInfo(ctx, "label", double2, xgboost::DataType::kFloat32, 2);
info.SetInfo(ctx, "label", Make1dInterfaceTest(double2, 2));
EXPECT_EQ(info.labels.Size(), 2);
float float2[2] = {1.0f, 2.0f};
EXPECT_EQ(info.GetWeight(1), 1.0f)
<< "When no weights are given, was expecting default value 1";
info.SetInfo(ctx, "weight", float2, xgboost::DataType::kFloat32, 2);
EXPECT_EQ(info.GetWeight(1), 1.0f) << "When no weights are given, was expecting default value 1";
info.SetInfo(ctx, "weight", Make1dInterfaceTest(float2, 2));
EXPECT_EQ(info.GetWeight(1), 2.0f);
uint32_t uint32_t2[2] = {1U, 2U};
EXPECT_EQ(info.base_margin_.Size(), 0);
info.SetInfo(ctx, "base_margin", uint32_t2, xgboost::DataType::kUInt32, 2);
info.SetInfo(ctx, "base_margin", Make1dInterfaceTest(uint32_t2, 2));
EXPECT_EQ(info.base_margin_.Size(), 2);
uint64_t uint64_t2[2] = {1U, 2U};
EXPECT_EQ(info.group_ptr_.size(), 0);
info.SetInfo(ctx, "group", uint64_t2, xgboost::DataType::kUInt64, 2);
info.SetInfo(ctx, "group", Make1dInterfaceTest(uint64_t2, 2));
ASSERT_EQ(info.group_ptr_.size(), 3);
EXPECT_EQ(info.group_ptr_[2], 3);
@@ -135,9 +133,9 @@ TEST(MetaInfo, SaveLoadBinary) {
};
std::vector<float> values (kRows);
std::generate(values.begin(), values.end(), generator);
info.SetInfo(ctx, "label", values.data(), xgboost::DataType::kFloat32, kRows);
info.SetInfo(ctx, "weight", values.data(), xgboost::DataType::kFloat32, kRows);
info.SetInfo(ctx, "base_margin", values.data(), xgboost::DataType::kFloat32, kRows);
info.SetInfo(ctx, "label", Make1dInterfaceTest(values.data(), kRows));
info.SetInfo(ctx, "weight", Make1dInterfaceTest(values.data(), kRows));
info.SetInfo(ctx, "base_margin", Make1dInterfaceTest(values.data(), kRows));
info.num_row_ = kRows;
info.num_col_ = kCols;
@@ -271,7 +269,7 @@ TEST(MetaInfo, CPUQid) {
qid[i] = i;
}
info.SetInfo(ctx, "qid", qid.data(), xgboost::DataType::kUInt32, info.num_row_);
info.SetInfo(ctx, "qid", Make1dInterfaceTest(qid.data(), info.num_row_));
ASSERT_EQ(info.group_ptr_.size(), info.num_row_ + 1);
ASSERT_EQ(info.group_ptr_.front(), 0);
ASSERT_EQ(info.group_ptr_.back(), info.num_row_);
@@ -288,14 +286,12 @@ TEST(MetaInfo, Validate) {
info.num_col_ = 3;
std::vector<xgboost::bst_group_t> groups (11);
Context ctx;
info.SetInfo(ctx, "group", groups.data(), xgboost::DataType::kUInt32, 11);
info.SetInfo(ctx, "group", Make1dInterfaceTest(groups.data(), groups.size()));
EXPECT_THROW(info.Validate(FstCU()), dmlc::Error);
std::vector<float> labels(info.num_row_ + 1);
EXPECT_THROW(
{
info.SetInfo(ctx, "label", labels.data(), xgboost::DataType::kFloat32, info.num_row_ + 1);
},
{ info.SetInfo(ctx, "label", Make1dInterfaceTest(labels.data(), info.num_row_ + 1)); },
dmlc::Error);
// Make overflow data, which can happen when users pass group structure as int
@@ -305,13 +301,13 @@ TEST(MetaInfo, Validate) {
groups.push_back(1562500);
}
groups.push_back(static_cast<xgboost::bst_group_t>(-1));
EXPECT_THROW(info.SetInfo(ctx, "group", groups.data(), xgboost::DataType::kUInt32, groups.size()),
EXPECT_THROW(info.SetInfo(ctx, "group", Make1dInterfaceTest(groups.data(), groups.size())),
dmlc::Error);
#if defined(XGBOOST_USE_CUDA)
info.group_ptr_.clear();
labels.resize(info.num_row_);
info.SetInfo(ctx, "label", labels.data(), xgboost::DataType::kFloat32, info.num_row_);
info.SetInfo(ctx, "label", Make1dInterfaceTest(labels.data(), info.num_row_));
info.labels.SetDevice(FstCU());
EXPECT_THROW(info.Validate(DeviceOrd::CUDA(1)), dmlc::Error);
@@ -340,8 +336,8 @@ TEST(MetaInfo, HostExtend) {
for (size_t g = 0; g < kRows / per_group; ++g) {
groups.emplace_back(per_group);
}
lhs.SetInfo(ctx, "group", groups.data(), xgboost::DataType::kUInt32, groups.size());
rhs.SetInfo(ctx, "group", groups.data(), xgboost::DataType::kUInt32, groups.size());
lhs.SetInfo(ctx, "group", Make1dInterfaceTest(groups.data(), groups.size()));
rhs.SetInfo(ctx, "group", Make1dInterfaceTest(groups.data(), groups.size()));
lhs.Extend(rhs, true, true);
ASSERT_EQ(lhs.num_row_, kRows * 2);