Remove redundant whitespaces (#5480)

remove redundant whitespaces

src/io/bin.cpp

@@ -508,7 +508,7 @@ namespace LightGBM {
       const double max_sparse_rate =
           static_cast<double>(cnt_in_bin[most_freq_bin_]) / total_sample_cnt;
       // When most_freq_bin_ != default_bin_, there are some additional data loading costs.
-      // so use most_freq_bin_  = default_bin_ when there is not so sparse
+      // so use most_freq_bin_ = default_bin_ when there is not so sparse
       if (most_freq_bin_ != default_bin_ && max_sparse_rate < kSparseThreshold) {
         most_freq_bin_ = default_bin_;
       }
@@ -705,7 +705,7 @@ namespace LightGBM {
         return new MultiValSparseBin<uint32_t, uint32_t>(
             num_data, num_bin, estimate_element_per_row);
       }
-    } else  {
+    } else {
       if (num_bin <= 256) {
         return new MultiValSparseBin<size_t, uint8_t>(
             num_data, num_bin, estimate_element_per_row);

src/network/socket_wrapper.hpp

@@ -248,7 +248,7 @@ class TcpSocket {
   }
 
   inline bool Connect(const char *url, int port) {
-    sockaddr_in  server_addr = GetAddress(url, port);
+    sockaddr_in server_addr = GetAddress(url, port);
     if (connect(sockfd_, reinterpret_cast<const sockaddr*>(&server_addr), sizeof(sockaddr_in)) == 0) {
       return true;
     }

src/objective/binary_objective.hpp

@@ -129,7 +129,7 @@ class BinaryLogloss: public ObjectiveFunction {
         // calculate gradients and hessians
         const double response = -label * sigmoid_ / (1.0f + std::exp(label * sigmoid_ * score[i]));
         const double abs_response = fabs(response);
-        gradients[i] = static_cast<score_t>(response * label_weight  * weights_[i]);
+        gradients[i] = static_cast<score_t>(response * label_weight * weights_[i]);
         hessians[i] = static_cast<score_t>(abs_response * (sigmoid_ - abs_response) * label_weight * weights_[i]);
       }
     }

src/treelearner/cuda/cuda_data_partition.hpp

@@ -346,7 +346,7 @@ class CUDADataPartition {
   data_size_t* cuda_data_indices_;
   /*! \brief start position of each leaf in cuda_data_indices_ */
   data_size_t* cuda_leaf_data_start_;
-  /*! \brief end position of each leaf in cuda_data_indices_  */
+  /*! \brief end position of each leaf in cuda_data_indices_ */
   data_size_t* cuda_leaf_data_end_;
   /*! \brief number of data in each leaf */
   data_size_t* cuda_leaf_num_data_;

src/treelearner/gpu_tree_learner.cpp

@@ -266,7 +266,7 @@ void GPUTreeLearner::AllocateGPUMemory() {
   ptr_pinned_gradients_ = queue_.enqueue_map_buffer(pinned_gradients_, boost::compute::command_queue::map_write_invalidate_region,
                                                     0, allocated_num_data_ * sizeof(score_t));
   pinned_hessians_ = boost::compute::buffer();  // deallocate
-  pinned_hessians_  = boost::compute::buffer(ctx_, allocated_num_data_ * sizeof(score_t),
+  pinned_hessians_ = boost::compute::buffer(ctx_, allocated_num_data_ * sizeof(score_t),
                                              boost::compute::memory_object::read_write | boost::compute::memory_object::use_host_ptr,
                                              ordered_hessians_.data());
   ptr_pinned_hessians_ = queue_.enqueue_map_buffer(pinned_hessians_, boost::compute::command_queue::map_write_invalidate_region,
@@ -277,7 +277,7 @@ void GPUTreeLearner::AllocateGPUMemory() {
   device_gradients_ = boost::compute::buffer(ctx_, allocated_num_data_ * sizeof(score_t),
                       boost::compute::memory_object::read_only, nullptr);
   device_hessians_ = boost::compute::buffer();  // deallocate
-  device_hessians_  = boost::compute::buffer(ctx_, allocated_num_data_ * sizeof(score_t),
+  device_hessians_ = boost::compute::buffer(ctx_, allocated_num_data_ * sizeof(score_t),
                       boost::compute::memory_object::read_only, nullptr);
   // allocate feature mask, for disabling some feature-groups' histogram calculation
   feature_masks_.resize(num_dense_feature4_ * dword_features_);

src/treelearner/monotone_constraints.hpp

@@ -875,7 +875,7 @@ class AdvancedLeafConstraints : public IntermediateLeafConstraints {
     // for example when adding a constraints cstr2 on thresholds [1:2),
     // on an existing constraints cstr1 on thresholds [0, +inf),
     // the thresholds and constraints must become
-    // [0, 1, 2] and  [cstr1, cstr2, cstr1]
+    // [0, 1, 2] and [cstr1, cstr2, cstr1]
     // so since we loop through thresholds only once,
     // the previous constraint that still applies needs to be recorded
     double previous_constraint = use_max_operator