lrn coarse presicion

src/operators/kernel/arm/lrn_kernel.cpp

@@ -23,21 +23,27 @@ namespace operators {
 
 template <>
 void LrnKernel<CPU, float>::Compute(const LrnParam &param) const {
+  const float alpha = param.Alpha();
   const Tensor *input_x = param.InputX();
   auto x_dims = input_x->dims();
-  /// data_format = NCHW
-  const int N = x_dims[0];
-  const int C = x_dims[1];
-  const int H = x_dims[2];
-  const int W = x_dims[3];
   Tensor *out = param.Out();
   out->mutable_data<float>();
-  const int n = param.N();
-  const float alpha = param.Alpha();
-  const float beta = param.Beta();
-  const float k = param.K();
-  LRNFunctor<float> lrnFunctor;
-  lrnFunctor(*input_x, out, N, C, H, W, n, k, alpha, beta);
+  if (alpha < 0.001) {
+    // coarse precision
+    out->ShareDataWith(*input_x);
+  } else {
+    /// data_format = NCHW
+    const int N = x_dims[0];
+    const int C = x_dims[1];
+    const int H = x_dims[2];
+    const int W = x_dims[3];
+
+    const int n = param.N();
+    const float beta = param.Beta();
+    const float k = param.K();
+    LRNFunctor<float> lrnFunctor;
+    lrnFunctor(*input_x, out, N, C, H, W, n, k, alpha, beta);
+  }
 }
 
 template class LrnKernel<CPU, float>;