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

Write ELLPACK pages to disk (#4879)

Browse Source 
* add ellpack source
* add batch param
* extract function to parse cache info
* construct ellpack info separately
* push batch to ellpack page
* write ellpack page.
* make sparse page source reusable
This commit is contained in:
Rong Ou
2019-10-22 20:44:32 -07:00
committed by Jiaming Yuan
parent 310fe60b35
commit 5b1715d97c
25 changed files with 935 additions and 408 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
src/data/sparse_page_raw_format.cc
View File

@@ -12,9 +12,10 @@ namespace data {
DMLC_REGISTRY_FILE_TAG(sparse_page_raw_format);
class SparsePageRawFormat : public SparsePageFormat {
template<typename T>
class SparsePageRawFormat : public SparsePageFormat<T> {
public:
bool Read(SparsePage* page, dmlc::SeekStream* fi) override {
bool Read(T* page, dmlc::SeekStream* fi) override {
auto& offset_vec = page->offset.HostVector();
if (!fi->Read(&offset_vec)) return false;
auto& data_vec = page->data.HostVector();
@@ -29,7 +30,7 @@ class SparsePageRawFormat : public SparsePageFormat {
return true;
}
bool Read(SparsePage* page,
bool Read(T* page,
dmlc::SeekStream* fi,
const std::vector<bst_uint>& sorted_index_set) override {
if (!fi->Read(&disk_offset_)) return false;
@@ -79,7 +80,7 @@ class SparsePageRawFormat : public SparsePageFormat {
return true;
}
void Write(const SparsePage& page, dmlc::Stream* fo) override {
void Write(const T& page, dmlc::Stream* fo) override {
const auto& offset_vec = page.offset.HostVector();
const auto& data_vec = page.data.HostVector();
CHECK(page.offset.Size() != 0 && offset_vec[0] == 0);
@@ -98,7 +99,20 @@ class SparsePageRawFormat : public SparsePageFormat {
XGBOOST_REGISTER_SPARSE_PAGE_FORMAT(raw)
.describe("Raw binary data format.")
.set_body([]() {
return new SparsePageRawFormat();
return new SparsePageRawFormat<SparsePage>();
});
XGBOOST_REGISTER_CSC_PAGE_FORMAT(raw)
.describe("Raw binary data format.")
.set_body([]() {
return new SparsePageRawFormat<CSCPage>();
});
XGBOOST_REGISTER_SORTED_CSC_PAGE_FORMAT(raw)
.describe("Raw binary data format.")
.set_body([]() {
return new SparsePageRawFormat<SortedCSCPage>();
});
} // namespace data
} // namespace xgboost