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

Fix CPU bin compression with categorical data. (#8809)

Browse Source 
* Fix CPU bin compression with categorical data.

* The bug causes the maximum category to be lesser than 256 or the maximum number of bins when
the input data is dense.
This commit is contained in:
Jiaming Yuan
2023-02-16 04:20:34 +08:00
committed by GitHub
parent cce4af4acf
commit c0afdb6786
6 changed files with 49 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/common/column_matrix.cc
View File

@@ -46,7 +46,7 @@ void ColumnMatrix::InitStorage(GHistIndexMatrix const& gmat, double sparse_thres
feature_offsets_[fid] = accum_index;
}
SetTypeSize(gmat.max_num_bins);
SetTypeSize(gmat.MaxNumBinPerFeat());
auto storage_size =
feature_offsets_.back() * static_cast<std::underlying_type_t<BinTypeSize>>(bins_type_size_);
index_.resize(storage_size, 0);