Support categorical data in ellpack. (#6140)

This commit is contained in:
Jiaming Yuan
2020-09-24 19:28:57 +08:00
committed by GitHub
parent 78d72ef936
commit 14afdb4d92
9 changed files with 189 additions and 43 deletions

View File

@@ -60,19 +60,6 @@ inline data::CupyAdapter AdapterFromData(const thrust::device_vector<float> &x,
}
#endif
inline std::vector<float> GenerateRandomCategoricalSingleColumn(int n,
int num_categories) {
std::vector<float> x(n);
std::mt19937 rng(0);
std::uniform_int_distribution<int> dist(0, num_categories - 1);
std::generate(x.begin(), x.end(), [&]() { return dist(rng); });
// Make sure each category is present
for(auto i = 0; i < num_categories; i++) {
x[i] = i;
}
return x;
}
inline std::shared_ptr<data::SimpleDMatrix>
GetDMatrixFromData(const std::vector<float> &x, int num_rows, int num_columns) {
data::DenseAdapter adapter(x.data(), num_rows, num_columns);