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

Rename IterativeDMatrix. (#8045)

Browse Source
This commit is contained in:
Jiaming Yuan 2022-07-04 18:52:31 +08:00 committed by GitHub
parent f24bfc7684
commit 8746f9cddf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 29 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/data/data.cc
View File

@ -26,7 +26,7 @@
#include "../common/group_data.h" #include "../common/group_data.h"
#include "../common/threading_utils.h" #include "../common/threading_utils.h"
#include "../data/adapter.h" #include "../data/adapter.h"
#include "../data/iterative_device_dmatrix.h" #include "../data/iterative_dmatrix.h"
#include "file_iterator.h" #include "file_iterator.h"
#include "validation.h" #include "validation.h"
@ -938,7 +938,7 @@ DMatrix *DMatrix::Create(DataIterHandle iter, DMatrixHandle proxy,
XGDMatrixCallbackNext *next, float missing, XGDMatrixCallbackNext *next, float missing,
int nthread, int nthread,
int max_bin) { int max_bin) {
return new data::IterativeDeviceDMatrix(iter, proxy, reset, next, missing, return new data::IterativeDMatrix(iter, proxy, reset, next, missing,
nthread, max_bin); nthread, max_bin);
} }

8
src/data/iterative_device_dmatrix.cu → src/data/iterative_dmatrix.cu
View File

@ -1,5 +1,5 @@
/*! /*!
* Copyright 2020 XGBoost contributors * Copyright 2020-2022 XGBoost contributors
*/ */
#include <memory> #include <memory>
#include <type_traits> #include <type_traits>
@ -7,7 +7,7 @@
#include "../common/hist_util.cuh" #include "../common/hist_util.cuh"
#include "simple_batch_iterator.h" #include "simple_batch_iterator.h"
#include "iterative_device_dmatrix.h" #include "iterative_dmatrix.h"
#include "sparse_page_source.h" #include "sparse_page_source.h"
#include "ellpack_page.cuh" #include "ellpack_page.cuh"
#include "proxy_dmatrix.h" #include "proxy_dmatrix.h"
@ -16,7 +16,7 @@
namespace xgboost { namespace xgboost {
namespace data { namespace data {
void IterativeDeviceDMatrix::Initialize(DataIterHandle iter_handle, float missing) { void IterativeDMatrix::InitFromCUDA(DataIterHandle iter_handle, float missing) {
// A handle passed to external iterator. // A handle passed to external iterator.
DMatrixProxy* proxy = MakeProxy(proxy_); DMatrixProxy* proxy = MakeProxy(proxy_);
CHECK(proxy); CHECK(proxy);
@ -160,7 +160,7 @@ void IterativeDeviceDMatrix::Initialize(DataIterHandle iter_handle, float missin
rabit::Allreduce<rabit::op::Max>(&info_.num_col_, 1); rabit::Allreduce<rabit::op::Max>(&info_.num_col_, 1);
} }
BatchSet<EllpackPage> IterativeDeviceDMatrix::GetEllpackBatches(const BatchParam& param) { BatchSet<EllpackPage> IterativeDMatrix::GetEllpackBatches(const BatchParam& param) {
CHECK(page_); CHECK(page_);
// FIXME(Jiamingy): https://github.com/dmlc/xgboost/issues/7976 // FIXME(Jiamingy): https://github.com/dmlc/xgboost/issues/7976
if (param.max_bin != batch_param_.max_bin) { if (param.max_bin != batch_param_.max_bin) {

29
src/data/iterative_device_dmatrix.h → src/data/iterative_dmatrix.h
View File

@ -1,9 +1,9 @@
/*! /*!
* Copyright 2020 by Contributors * Copyright 2020-2022 by Contributors
* \file iterative_device_dmatrix.h * \file iterative_dmatrix.h
*/ */
#ifndef XGBOOST_DATA_ITERATIVE_DEVICE_DMATRIX_H_ #ifndef XGBOOST_DATA_ITERATIVE_DMATRIX_H_
#define XGBOOST_DATA_ITERATIVE_DEVICE_DMATRIX_H_ #define XGBOOST_DATA_ITERATIVE_DMATRIX_H_
#include <vector> #include <vector>
#include <string> #include <string>
@ -19,7 +19,7 @@
namespace xgboost { namespace xgboost {
namespace data { namespace data {
class IterativeDeviceDMatrix : public DMatrix { class IterativeDMatrix : public DMatrix {
MetaInfo info_; MetaInfo info_;
Context ctx_; Context ctx_;
BatchParam batch_param_; BatchParam batch_param_;
@ -30,18 +30,17 @@ class IterativeDeviceDMatrix : public DMatrix {
XGDMatrixCallbackNext *next_; XGDMatrixCallbackNext *next_;
public: public:
void Initialize(DataIterHandle iter, float missing); void InitFromCUDA(DataIterHandle iter, float missing);
public: public:
explicit IterativeDeviceDMatrix(DataIterHandle iter, DMatrixHandle proxy, explicit IterativeDMatrix(DataIterHandle iter, DMatrixHandle proxy, DataIterResetCallback *reset,
DataIterResetCallback *reset, XGDMatrixCallbackNext *next, XGDMatrixCallbackNext *next, float missing, int nthread, int max_bin)
float missing, int nthread, int max_bin)
: proxy_{proxy}, reset_{reset}, next_{next} { : proxy_{proxy}, reset_{reset}, next_{next} {
batch_param_ = BatchParam{0, max_bin}; batch_param_ = BatchParam{MakeProxy(proxy_)->DeviceIdx(), max_bin};
ctx_.UpdateAllowUnknown(Args{{"nthread", std::to_string(nthread)}}); ctx_.UpdateAllowUnknown(Args{{"nthread", std::to_string(nthread)}});
this->Initialize(iter, missing); this->InitFromCUDA(iter, missing);
} }
~IterativeDeviceDMatrix() override = default; ~IterativeDMatrix() override = default;
bool EllpackExists() const override { return true; } bool EllpackExists() const override { return true; }
bool SparsePageExists() const override { return false; } bool SparsePageExists() const override { return false; }
@ -77,14 +76,14 @@ class IterativeDeviceDMatrix : public DMatrix {
}; };
#if !defined(XGBOOST_USE_CUDA) #if !defined(XGBOOST_USE_CUDA)
inline void IterativeDeviceDMatrix::Initialize(DataIterHandle iter, float missing) { inline void IterativeDMatrix::InitFromCUDA(DataIterHandle iter, float missing) {
// silent the warning about unused variables. // silent the warning about unused variables.
(void)(proxy_); (void)(proxy_);
(void)(reset_); (void)(reset_);
(void)(next_); (void)(next_);
common::AssertGPUSupport(); common::AssertGPUSupport();
} }
inline BatchSet<EllpackPage> IterativeDeviceDMatrix::GetEllpackBatches(const BatchParam& param) { inline BatchSet<EllpackPage> IterativeDMatrix::GetEllpackBatches(const BatchParam& param) {
common::AssertGPUSupport(); common::AssertGPUSupport();
auto begin_iter = auto begin_iter =
BatchIterator<EllpackPage>(new SimpleBatchIteratorImpl<EllpackPage>(page_)); BatchIterator<EllpackPage>(new SimpleBatchIteratorImpl<EllpackPage>(page_));
@ -94,4 +93,4 @@ inline BatchSet<EllpackPage> IterativeDeviceDMatrix::GetEllpackBatches(const Bat
} // namespace data } // namespace data
} // namespace xgboost } // namespace xgboost
#endif // XGBOOST_DATA_ITERATIVE_DEVICE_DMATRIX_H_ #endif // XGBOOST_DATA_ITERATIVE_DMATRIX_H_

12
tests/cpp/data/test_iterative_device_dmatrix.cu → tests/cpp/data/test_iterative_dmatrix.cu
View File

@ -1,10 +1,10 @@
/*! /*!
* Copyright 2020 XGBoost contributors * Copyright 2020-2022 XGBoost contributors
*/ */
#include <gtest/gtest.h> #include <gtest/gtest.h>
#include "../helpers.h" #include "../helpers.h"
#include "../../../src/data/iterative_device_dmatrix.h" #include "../../../src/data/iterative_dmatrix.h"
#include "../../../src/data/ellpack_page.cuh" #include "../../../src/data/ellpack_page.cuh"
#include "../../../src/data/device_adapter.cuh" #include "../../../src/data/device_adapter.cuh"
@ -13,7 +13,7 @@ namespace data {
void TestEquivalent(float sparsity) { void TestEquivalent(float sparsity) {
CudaArrayIterForTest iter{sparsity}; CudaArrayIterForTest iter{sparsity};
IterativeDeviceDMatrix m( IterativeDMatrix m(
&iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), &iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(),
0, 256); 0, 256);
size_t offset = 0; size_t offset = 0;
@ -88,7 +88,7 @@ TEST(IterativeDeviceDMatrix, Basic) {
TEST(IterativeDeviceDMatrix, RowMajor) { TEST(IterativeDeviceDMatrix, RowMajor) {
CudaArrayIterForTest iter(0.0f); CudaArrayIterForTest iter(0.0f);
IterativeDeviceDMatrix m( IterativeDMatrix m(
&iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), &iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(),
0, 256); 0, 256);
size_t n_batches = 0; size_t n_batches = 0;
@ -139,7 +139,7 @@ TEST(IterativeDeviceDMatrix, RowMajorMissing) {
reinterpret_cast<float *>(get<Integer>(j_interface["data"][0]))); reinterpret_cast<float *>(get<Integer>(j_interface["data"][0])));
thrust::copy(h_data.cbegin(), h_data.cend(), ptr); thrust::copy(h_data.cbegin(), h_data.cend(), ptr);
IterativeDeviceDMatrix m( IterativeDMatrix m(
&iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), &iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(),
0, 256); 0, 256);
auto &ellpack = *m.GetBatches<EllpackPage>({0, 256}).begin(); auto &ellpack = *m.GetBatches<EllpackPage>({0, 256}).begin();
@ -159,7 +159,7 @@ TEST(IterativeDeviceDMatrix, IsDense) {
int num_bins = 16; int num_bins = 16;
auto test = [num_bins] (float sparsity) { auto test = [num_bins] (float sparsity) {
CudaArrayIterForTest iter(sparsity); CudaArrayIterForTest iter(sparsity);
IterativeDeviceDMatrix m( IterativeDMatrix m(
&iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), &iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(),
0, 256); 0, 256);
if (sparsity == 0.0) { if (sparsity == 0.0) {

8
tests/cpp/helpers.cu
View File

@ -2,7 +2,7 @@
#include "helpers.h" #include "helpers.h"
#include "../../src/data/device_adapter.cuh" #include "../../src/data/device_adapter.cuh"
#include "../../src/data/iterative_device_dmatrix.h" #include "../../src/data/iterative_dmatrix.h"
namespace xgboost { namespace xgboost {
@ -28,12 +28,10 @@ int CudaArrayIterForTest::Next() {
return 1; return 1;
} }
std::shared_ptr<DMatrix> RandomDataGenerator::GenerateDeviceDMatrix() { std::shared_ptr<DMatrix> RandomDataGenerator::GenerateDeviceDMatrix() {
CudaArrayIterForTest iter{this->sparsity_, this->rows_, this->cols_, 1}; CudaArrayIterForTest iter{this->sparsity_, this->rows_, this->cols_, 1};
auto m = std::make_shared<data::IterativeDeviceDMatrix>( auto m = std::make_shared<data::IterativeDMatrix>(
&iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), &iter, iter.Proxy(), Reset, Next, std::numeric_limits<float>::quiet_NaN(), 0, bins_);
0, bins_);
return m; return m;
} }
} // namespace xgboost } // namespace xgboost