Rewrite sparse dmatrix using callbacks. (#7092)

- Reduce dependency on dmlc parsers and provide an interface for users to load data by themselves.
- Remove use of threaded iterator and IO queue.
- Remove `page_size`.
- Make sure the number of pages in memory is bounded.
- Make sure the cache can not be violated.
- Provide an interface for internal algorithms to process data asynchronously.
This commit is contained in:
Jiaming Yuan
2021-07-16 12:33:31 +08:00
committed by GitHub
parent 2f524e9f41
commit bd1f3a38f0
51 changed files with 1445 additions and 1391 deletions

View File

@@ -83,7 +83,7 @@ inline std::shared_ptr<DMatrix> GetExternalMemoryDMatrixFromData(
}
fo.close();
return std::shared_ptr<DMatrix>(DMatrix::Load(
tmp_file + "#" + tmp_file + ".cache", true, false, "auto", page_size));
tmp_file + "#" + tmp_file + ".cache", true, false, "auto"));
}
// Test that elements are approximately equally distributed among bins