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

Partial rewrite EllpackPage (#5352)

Browse Source
This commit is contained in:
Rory Mitchell
2020-03-11 10:15:53 +13:00
committed by GitHub
parent 7a99f8f27f
commit 3ad4333b0e
23 changed files with 496 additions and 733 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
tests/cpp/tree/gpu_hist/test_gradient_based_sampler.cu
View File

@@ -3,6 +3,7 @@
#include "../../../../src/data/ellpack_page.cuh"
#include "../../../../src/tree/gpu_hist/gradient_based_sampler.cuh"
#include "../../helpers.h"
#include "dmlc/filesystem.h"
namespace xgboost {
namespace tree {
@@ -29,7 +30,7 @@ void VerifySampling(size_t page_size,
BatchParam param{0, 256, 0, page_size};
auto page = (*dmat->GetBatches<EllpackPage>(param).begin()).Impl();
if (page_size != 0) {
EXPECT_NE(page->matrix.n_rows, kRows);
EXPECT_NE(page->n_rows, kRows);
}
GradientBasedSampler sampler(page, kRows, param, subsample, sampling_method);
@@ -37,11 +38,11 @@ void VerifySampling(size_t page_size,
if (fixed_size_sampling) {
EXPECT_EQ(sample.sample_rows, kRows);
EXPECT_EQ(sample.page->matrix.n_rows, kRows);
EXPECT_EQ(sample.page->n_rows, kRows);
EXPECT_EQ(sample.gpair.size(), kRows);
} else {
EXPECT_NEAR(sample.sample_rows, sample_rows, kRows * 0.016f);
EXPECT_NEAR(sample.page->matrix.n_rows, sample_rows, kRows * 0.016f);
EXPECT_NEAR(sample.sample_rows, sample_rows, kRows * 0.016);
EXPECT_NEAR(sample.page->n_rows, sample_rows, kRows * 0.016f);
EXPECT_NEAR(sample.gpair.size(), sample_rows, kRows * 0.016f);
}
@@ -83,7 +84,7 @@ TEST(GradientBasedSampler, NoSampling_ExternalMemory) {
BatchParam param{0, 256, 0, kPageSize};
auto page = (*dmat->GetBatches<EllpackPage>(param).begin()).Impl();
EXPECT_NE(page->matrix.n_rows, kRows);
EXPECT_NE(page->n_rows, kRows);
GradientBasedSampler sampler(page, kRows, param, kSubsample, TrainParam::kUniform);
auto sample = sampler.Sample(gpair.DeviceSpan(), dmat.get());
@@ -91,21 +92,19 @@ TEST(GradientBasedSampler, NoSampling_ExternalMemory) {
EXPECT_EQ(sample.sample_rows, kRows);
EXPECT_EQ(sample.gpair.size(), gpair.Size());
EXPECT_EQ(sample.gpair.data(), gpair.DevicePointer());
EXPECT_EQ(sampled_page->matrix.n_rows, kRows);
EXPECT_EQ(sampled_page->n_rows, kRows);
std::vector<common::CompressedByteT> buffer(sampled_page->gidx_buffer.size());
dh::CopyDeviceSpanToVector(&buffer, sampled_page->gidx_buffer);
std::vector<common::CompressedByteT> buffer(sampled_page->gidx_buffer.HostVector());
common::CompressedIterator<common::CompressedByteT>
ci(buffer.data(), sampled_page->matrix.info.NumSymbols());
ci(buffer.data(), sampled_page->NumSymbols());
size_t offset = 0;
for (auto& batch : dmat->GetBatches<EllpackPage>(param)) {
auto page = batch.Impl();
std::vector<common::CompressedByteT> page_buffer(page->gidx_buffer.size());
dh::CopyDeviceSpanToVector(&page_buffer, page->gidx_buffer);
std::vector<common::CompressedByteT> page_buffer(page->gidx_buffer.HostVector());
common::CompressedIterator<common::CompressedByteT>
page_ci(page_buffer.data(), page->matrix.info.NumSymbols());
size_t num_elements = page->matrix.n_rows * page->matrix.info.row_stride;
page_ci(page_buffer.data(), page->NumSymbols());
size_t num_elements = page->n_rows * page->row_stride;
for (size_t i = 0; i < num_elements; i++) {
EXPECT_EQ(ci[i + offset], page_ci[i]);
}

6
tests/cpp/tree/gpu_hist/test_histogram.cu
View File

@@ -27,7 +27,7 @@ void TestDeterminsticHistogram() {
gpair.SetDevice(0);
auto rounding = CreateRoundingFactor<Gradient>(gpair.DeviceSpan());
BuildGradientHistogram(page->matrix, gpair.DeviceSpan(), ridx,
BuildGradientHistogram(page->GetDeviceAccessor(0), gpair.DeviceSpan(), ridx,
d_histogram, rounding, true);
for (size_t i = 0; i < kRounds; ++i) {
@@ -35,7 +35,7 @@ void TestDeterminsticHistogram() {
auto d_histogram = dh::ToSpan(new_histogram);
auto rounding = CreateRoundingFactor<Gradient>(gpair.DeviceSpan());
BuildGradientHistogram(page->matrix, gpair.DeviceSpan(), ridx,
BuildGradientHistogram(page->GetDeviceAccessor(0), gpair.DeviceSpan(), ridx,
d_histogram, rounding, true);
for (size_t j = 0; j < new_histogram.size(); ++j) {
@@ -50,7 +50,7 @@ void TestDeterminsticHistogram() {
auto gpair = GenerateRandomGradients(kRows, kLower, kUpper);
gpair.SetDevice(0);
dh::device_vector<Gradient> baseline(kBins * kCols);
BuildGradientHistogram(page->matrix, gpair.DeviceSpan(), ridx,
BuildGradientHistogram(page->GetDeviceAccessor(0), gpair.DeviceSpan(), ridx,
dh::ToSpan(baseline), rounding, true);
for (size_t i = 0; i < baseline.size(); ++i) {
EXPECT_NEAR(((Gradient)baseline[i]).GetGrad(), ((Gradient)histogram[i]).GetGrad(),