COMPMID-1645  NEL2Normalization for FP32/FP16 & NHWC

Change-Id: I29e35024e29781a6b943b568abec9c73649215e6
diff --git a/src/runtime/CL/functions/CLL2NormalizeLayer.cpp b/src/runtime/CL/functions/CLL2NormalizeLayer.cpp
index 4f709d5..2e3c6d7 100644
--- a/src/runtime/CL/functions/CLL2NormalizeLayer.cpp
+++ b/src/runtime/CL/functions/CLL2NormalizeLayer.cpp
@@ -32,8 +32,8 @@
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "support/ToolchainSupport.h"
 
-using namespace arm_compute;
-
+namespace arm_compute
+{
 CLL2NormalizeLayer::CLL2NormalizeLayer(std::shared_ptr<IMemoryManager> memory_manager)
     : _memory_group(std::move(memory_manager)), _reduce_func(), _normalize_kernel(), _sumsq()
 {
@@ -81,3 +81,4 @@
 
     _memory_group.release();
 }
+} // namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEL2NormalizeLayer.cpp b/src/runtime/NEON/functions/NEL2NormalizeLayer.cpp
index d0b80fb..56da966 100644
--- a/src/runtime/NEON/functions/NEL2NormalizeLayer.cpp
+++ b/src/runtime/NEON/functions/NEL2NormalizeLayer.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
-using namespace arm_compute;
-
+namespace arm_compute
+{
 NEL2NormalizeLayer::NEL2NormalizeLayer(std::shared_ptr<IMemoryManager> memory_manager)
     : _memory_group(std::move(memory_manager)), _reduce_func(), _normalize_kernel(), _sumsq()
 {
@@ -57,8 +57,8 @@
 
     ARM_COMPUTE_RETURN_ON_ERROR(NEReductionOperation::validate(input, &sum_sq, axis, ReductionOperation::SUM_SQUARE));
 
-    // Reduce shape on axis (supported axis is 0)
-    shape.set(0, 1);
+    // Reduce shape on axis
+    shape.set(axis, 1);
     sum_sq.set_tensor_shape(shape);
 
     ARM_COMPUTE_RETURN_ON_ERROR(NEL2NormalizeLayerKernel::validate(input, &sum_sq, output, axis, epsilon));
@@ -75,3 +75,4 @@
 
     _memory_group.release();
 }
+} // namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEReductionOperation.cpp b/src/runtime/NEON/functions/NEReductionOperation.cpp
index 188c2bb..bb27b5d 100644
--- a/src/runtime/NEON/functions/NEReductionOperation.cpp
+++ b/src/runtime/NEON/functions/NEReductionOperation.cpp
@@ -86,7 +86,7 @@
     if(axis == 0)
     {
         // Configure fill border kernel
-        BorderSize fill_border_size = (axis == 0) ? _reduction_kernel.border_size() : BorderSize();
+        BorderSize fill_border_size = _reduction_kernel.border_size();
         BorderMode fill_border_mode = reduction_operation_border_mode(op);
         _fill_border_kernel.configure(input, fill_border_size, fill_border_mode, PixelValue(static_cast<float>(0.f)));
     }