| 50 | } |
| 51 | |
| 52 | bool NeedFilter(const std::vector<int>& cnt_in_bin, int total_cnt, int filter_cnt, BinType bin_type) { |
| 53 | if (bin_type == BinType::NumericalBin) { |
| 54 | int sum_left = 0; |
| 55 | for (size_t i = 0; i < cnt_in_bin.size() - 1; ++i) { |
| 56 | sum_left += cnt_in_bin[i]; |
| 57 | if (sum_left >= filter_cnt && total_cnt - sum_left >= filter_cnt) { |
| 58 | return false; |
| 59 | } |
| 60 | } |
| 61 | } else { |
| 62 | if (cnt_in_bin.size() <= 2) { |
| 63 | for (size_t i = 0; i < cnt_in_bin.size() - 1; ++i) { |
| 64 | int sum_left = cnt_in_bin[i]; |
| 65 | if (sum_left >= filter_cnt && total_cnt - sum_left >= filter_cnt) { |
| 66 | return false; |
| 67 | } |
| 68 | } |
| 69 | } else { |
| 70 | return false; |
| 71 | } |
| 72 | } |
| 73 | return true; |
| 74 | } |
| 75 | |
| 76 | std::vector<double> GreedyFindBin(const double* distinct_values, const int* counts, |
| 77 | int num_distinct_values, int max_bin, |