Add and test non-in-place scale math functions for CPU and GPU

include/caffe/util/math_functions.hpp

@@ -157,6 +157,12 @@ DEFINE_CAFFE_CPU_UNARY_FUNC(fabs, y[i] = std::fabs(x[i]));
 template <typename Dtype>
 void caffe_gpu_fabs(const int n, const Dtype* x, Dtype* y);
 
+template <typename Dtype>
+void caffe_cpu_scale(const int n, const Dtype alpha, const Dtype *x, Dtype* y);
+
+template <typename Dtype>
+void caffe_gpu_scale(const int n, const Dtype alpha, const Dtype *x, Dtype* y);
+
 }  // namespace caffe
 
 

src/caffe/test/test_math_functions.cpp

@@ -140,4 +140,30 @@ TYPED_TEST(MathFunctionsTest, TestFabsGPU){
   }
 }
 
+TYPED_TEST(MathFunctionsTest, TestScaleCPU){
+  int n = this->blob_bottom_->count();
+  TypeParam alpha = this->blob_bottom_->cpu_diff()[rand() %
+                                               this->blob_bottom_->count()];
+  caffe_cpu_scale<TypeParam>(n, alpha, this->blob_bottom_->cpu_data(),
+                             this->blob_bottom_->mutable_cpu_diff());
+  const TypeParam* scaled = this->blob_bottom_->cpu_diff();
+  const TypeParam* x = this->blob_bottom_->cpu_data();
+  for (int i = 0; i < n; ++i) {
+    CHECK_EQ(scaled[i], x[i] * alpha);
+  }
+}
+
+TYPED_TEST(MathFunctionsTest, TestScaleGPU){
+  int n = this->blob_bottom_->count();
+  TypeParam alpha = this->blob_bottom_->cpu_diff()[rand() %
+                                               this->blob_bottom_->count()];
+  caffe_gpu_scale<TypeParam>(n, alpha, this->blob_bottom_->gpu_data(),
+                             this->blob_bottom_->mutable_gpu_diff());
+  const TypeParam* scaled = this->blob_bottom_->cpu_diff();
+  const TypeParam* x = this->blob_bottom_->cpu_data();
+  for (int i = 0; i < n; ++i) {
+    CHECK_EQ(scaled[i], x[i] * alpha);
+  }
+}
+
 }

src/caffe/util/math_functions.cpp

@@ -413,4 +413,32 @@ void caffe_gpu_asum<double>(const int n, const double* x, double* y) {
 INSTANTIATE_CAFFE_CPU_UNARY_FUNC(sign);
 INSTANTIATE_CAFFE_CPU_UNARY_FUNC(fabs);
 
+template <>
+void caffe_cpu_scale<float>(const int n, const float alpha, const float *x,
+                            float* y) {
+  cblas_scopy(n, x, 1, y, 1);
+  cblas_sscal(n, alpha, y, 1);
+}
+
+template <>
+void caffe_cpu_scale<double>(const int n, const double alpha, const double *x,
+                             double* y) {
+  cblas_dcopy(n, x, 1, y, 1);
+  cblas_dscal(n, alpha, y, 1);
+}
+
+template <>
+void caffe_gpu_scale<float>(const int n, const float alpha, const float *x,
+                            float* y) {
+  CUBLAS_CHECK(cublasScopy(Caffe::cublas_handle(), n, x, 1, y, 1));
+  CUBLAS_CHECK(cublasSscal(Caffe::cublas_handle(), n, &alpha, y, 1));
+}
+
+template <>
+void caffe_gpu_scale<double>(const int n, const double alpha, const double *x,
+                             double* y) {
+  CUBLAS_CHECK(cublasDcopy(Caffe::cublas_handle(), n, x, 1, y, 1));
+  CUBLAS_CHECK(cublasDscal(Caffe::cublas_handle(), n, &alpha, y, 1));
+}
+
 }  // namespace caffe