slightly reduce the cost of multi-val bin construct. (#2728)

* slightly reduce the cost of multi-val bin construct.

* Update multi_val_sparse_bin.hpp

src/io/dataset.cpp

@@ -493,6 +493,7 @@ void PushDataToMultiValBin(int num_threads, data_size_t num_data, const std::vec
     #pragma omp parallel for schedule(static)
     for (int tid = 0; tid < n_block; ++tid) {
       std::vector<uint32_t> cur_data;
+      cur_data.reserve(most_freq_bins.size());
       data_size_t start = tid * block_size;
       data_size_t end = std::min(num_data, start + block_size);
       for (size_t j = 0; j < most_freq_bins.size(); ++j) {
@@ -517,14 +518,13 @@ void PushDataToMultiValBin(int num_threads, data_size_t num_data, const std::vec
   } else {
     #pragma omp parallel for schedule(static)
     for (int tid = 0; tid < n_block; ++tid) {
-      std::vector<uint32_t> cur_data;
+      std::vector<uint32_t> cur_data(most_freq_bins.size(), 0);
       data_size_t start = tid * block_size;
       data_size_t end = std::min(num_data, start + block_size);
       for (size_t j = 0; j < most_freq_bins.size(); ++j) {
         iters[tid][j]->Reset(start);
       }
       for (data_size_t i = start; i < end; ++i) {
-        cur_data.clear();
         for (size_t j = 0; j < most_freq_bins.size(); ++j) {
           auto cur_bin = iters[tid][j]->Get(i);
           if (cur_bin == most_freq_bins[j]) {
@@ -535,7 +535,7 @@ void PushDataToMultiValBin(int num_threads, data_size_t num_data, const std::vec
               cur_bin -= 1;
             }
           }
-          cur_data.push_back(cur_bin);
+          cur_data[j] = cur_bin;
         }
         ret->PushOneRow(tid, i, cur_data);
       }

src/io/multi_val_dense_bin.hpp

@@ -7,7 +7,6 @@
 
 
 #include <LightGBM/bin.h>
-#include <LightGBM/utils/openmp_wrapper.h>
 
 #include <cstdint>
 #include <cstring>
@@ -39,7 +38,9 @@ public:
 
   void PushOneRow(int , data_size_t idx, const std::vector<uint32_t>& values) override {
     auto start = RowPtr(idx);
+#ifdef DEBUG
     CHECK(num_feature_ == static_cast<int>(values.size()));
+#endif  // DEBUG
     for (auto i = 0; i < num_feature_; ++i) {
       data_[start + i] = static_cast<VAL_T>(values[i]);
     }

src/io/multi_val_sparse_bin.hpp

@@ -65,12 +65,16 @@ public:
       row_ptr_[i + 1] += row_ptr_[i];
     }
     if (t_data_.size() > 0) {
-      size_t offset = data_.size();
+      std::vector<size_t> offsets;
+      offsets.push_back(data_.size());
+      for (size_t tid = 0; tid < t_data_.size() - 1; ++tid) {
+        offsets.push_back(offsets.back() + t_data_[tid].size());
+      }
       data_.resize(row_ptr_[num_data_]);
-      for (size_t tid = 0; tid < t_data_.size(); ++tid) {
-        std::memcpy(data_.data() + offset, t_data_[tid].data(), t_data_[tid].size() * sizeof(VAL_T));
-        offset += t_data_[tid].size();
-        t_data_[tid].clear();
+#pragma omp parallel for schedule(static)
+      for (int tid = 0; tid < static_cast<int>(t_data_.size()); ++tid) {
+        std::copy_n(t_data_[tid].data(), t_data_[tid].size(),
+                    data_.data() + offsets[tid]);
       }
     }
     row_ptr_.shrink_to_fit();