added GetReductionInt to common_utils.h and replaced duplicated code in all loss with reduction gpu op kernels (nll loss, kl div loss, and binary cross entropy)

mindspore/ccsrc/backend/kernel_compiler/common_utils.cc

@@ -571,6 +571,17 @@ int Sign(float x) {
   return 0;
 }
 
+int GetReductionInt(const std::string &reduction) {
+  if (reduction == "none") {
+    return 0;
+  } else if (reduction == "sum") {
+    return 2;
+  } else {
+    // reduction = 'mean'
+    return 1;
+  }
+}
+
 std::pair<AnfNodePtr, size_t> GetKernelInput(const AnfNodePtr &anf_node, size_t index) {
   MS_EXCEPTION_IF_NULL(anf_node);
 

mindspore/ccsrc/backend/kernel_compiler/common_utils.h

@@ -89,6 +89,7 @@ std::string GetProcessor(const AnfNodePtr &anf_node);
 Processor GetProcessor(const string &processor);
 bool IsSameShape(const std::vector<size_t> &shape_a, const std::vector<size_t> &shape_b);
 int Sign(float x);
+int GetReductionInt(const std::string &reduction);
 std::pair<AnfNodePtr, size_t> GetKernelInput(const AnfNodePtr &anf_node, size_t index);
 std::vector<std::pair<AnfNodePtr, std::pair<size_t, size_t>>> GetInputIndex(const std::vector<AnfNodePtr> &node_list,
                                                                             const std::vector<AnfNodePtr> &input_list);

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_gpu_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -67,11 +68,7 @@ class BinaryCrossEntropyGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-    if (reduction == "none") {
-      reduction_ = 0;
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-    }
+    reduction_ = GetReductionInt(reduction);
     workspace_size_ = sizeof(T);
     if (reduction_ != 0) {
       workspace_size_ *= input_size_;

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_grad_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -69,11 +70,7 @@ class BinaryCrossEntropyGradGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-    if (reduction == "none") {
-      reduction_ = 0;
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-    }
+    reduction_ = GetReductionInt(reduction);
     InitSizeLists();
     return true;
   }

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/kl_div_loss_gpu_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -60,11 +61,7 @@ class KLDivLossGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-    if (reduction == "none") {
-      reduction_ = 0;
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-    }
+    reduction_ = GetReductionInt(reduction);
     workspace_size_ = sizeof(T);
     if (reduction_ == 0) {
       workspace_size_ *= input_size_;

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/kl_div_loss_grad_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -61,11 +62,7 @@ class KLDivLossGradGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-    if (reduction == "none") {
-      reduction_ = 0;
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-    }
+    reduction_ = GetReductionInt(reduction);
     InitSizeLists();
     return true;
   }

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/nll_loss_gpu_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -60,19 +61,10 @@ class NLLLossGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-
-    // if reduction is not 'none', tmp_nll is (N,) size
-    if (reduction == "none") {
-      reduction_ = 0;
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-      tmp_loss_size_ = sizeof(T) * n_;
-    } else {
-      // reduction = 'mean'
-      reduction_ = 1;
+    reduction_ = GetReductionInt(reduction);
+    if ((reduction_ == 2) || (reduction_ == 1)) {
       tmp_loss_size_ = sizeof(T) * n_;
     }
-
     tmp_target_weight_size_ = n_ * sizeof(S);
 
     InitSizeLists();

mindspore/ccsrc/backend/kernel_compiler/gpu/nn/nll_loss_grad_gpu_kernel.h

@@ -22,6 +22,7 @@
 #include "backend/kernel_compiler/gpu/gpu_kernel.h"
 #include "backend/kernel_compiler/gpu/gpu_kernel_factory.h"
 #include "backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cuh"
+#include "backend/kernel_compiler/common_utils.h"
 
 namespace mindspore {
 namespace kernel {
@@ -59,16 +60,9 @@ class NLLLossGradGpuKernel : public GpuKernel {
       input_size_ *= input_shape[i];
     }
     string reduction = GetAttr<string>(kernel_node, "reduction");
-
-    // if reduction is not 'none', tmp_nll is (N,) size
-    if (reduction == "none") {
-      reduction_ = 0;
-      num_dloss_ = n_;  // dloss is a vector
-    } else if (reduction == "sum") {
-      reduction_ = 2;
-    } else {
-      // reduction = 'mean'
-      reduction_ = 1;
+    reduction_ = GetReductionInt(reduction);
+    if (reduction_ == 0) {
+      num_dloss_ = n_;
     }
 
     InitSizeLists();