Move CPU/GPU files from Core/Runtime to the respective backend folders

Legacy structure contained two libraries core/runtime with two backends
in each.
We reduce the core/runtime libraries to a single library thus merging
the backend files

Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I69545765fe7a730368105cdbd067d3135ec7a174
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/6155
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

518 files changed, 1210 insertions, 1210 deletions

diff --git a/Android.bp b/Android.bp
index c1adf32893..14ab3da246 100644
--- a/Android.bp
+++ b/Android.bp
@@ -50,7 +50,7 @@ cc_library_static {
                          "src/core/NEON/kernels/assembly",
                          "src/core/NEON/kernels/convolution/common",
                          "src/core/NEON/kernels/convolution/winograd",
-                         "src/core/cpu/kernels/assembly"],
+                         "src/cpu/kernels/assembly"],
     export_include_dirs: [".", "./include"],
     srcs: [
         "src/c/AclContext.cpp",
@@ -245,145 +245,6 @@ cc_library_static {
         "src/core/Utils.cpp",
         "src/core/Validate.cpp",
         "src/core/Version.cpp",
-        "src/core/cpu/kernels/CpuActivationKernel.cpp",
-        "src/core/cpu/kernels/CpuAddKernel.cpp",
-        "src/core/cpu/kernels/CpuCastKernel.cpp",
-        "src/core/cpu/kernels/CpuCol2ImKernel.cpp",
-        "src/core/cpu/kernels/CpuConcatenateBatchKernel.cpp",
-        "src/core/cpu/kernels/CpuConcatenateDepthKernel.cpp",
-        "src/core/cpu/kernels/CpuConcatenateHeightKernel.cpp",
-        "src/core/cpu/kernels/CpuConcatenateWidthKernel.cpp",
-        "src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp",
-        "src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp",
-        "src/core/cpu/kernels/CpuCopyKernel.cpp",
-        "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp",
-        "src/core/cpu/kernels/CpuDequantizeKernel.cpp",
-        "src/core/cpu/kernels/CpuDirectConv2dKernel.cpp",
-        "src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp",
-        "src/core/cpu/kernels/CpuElementwiseKernel.cpp",
-        "src/core/cpu/kernels/CpuElementwiseUnaryKernel.cpp",
-        "src/core/cpu/kernels/CpuFillKernel.cpp",
-        "src/core/cpu/kernels/CpuFloorKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp",
-        "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.cpp",
-        "src/core/cpu/kernels/CpuIm2ColKernel.cpp",
-        "src/core/cpu/kernels/CpuMulKernel.cpp",
-        "src/core/cpu/kernels/CpuPermuteKernel.cpp",
-        "src/core/cpu/kernels/CpuPool2dKernel.cpp",
-        "src/core/cpu/kernels/CpuQuantizeKernel.cpp",
-        "src/core/cpu/kernels/CpuReshapeKernel.cpp",
-        "src/core/cpu/kernels/CpuScaleKernel.cpp",
-        "src/core/cpu/kernels/CpuSoftmaxKernel.cpp",
-        "src/core/cpu/kernels/CpuSubKernel.cpp",
-        "src/core/cpu/kernels/CpuTransposeKernel.cpp",
-        "src/core/cpu/kernels/CpuWeightsReshapeKernel.cpp",
-        "src/core/cpu/kernels/CpuWinogradConv2dKernel.cpp",
-        "src/core/cpu/kernels/activation/neon/fp16.cpp",
-        "src/core/cpu/kernels/activation/neon/fp32.cpp",
-        "src/core/cpu/kernels/activation/neon/qasymm8.cpp",
-        "src/core/cpu/kernels/activation/neon/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/activation/neon/qsymm16.cpp",
-        "src/core/cpu/kernels/activation/sve/fp16.cpp",
-        "src/core/cpu/kernels/activation/sve/fp32.cpp",
-        "src/core/cpu/kernels/activation/sve/qasymm8.cpp",
-        "src/core/cpu/kernels/activation/sve/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/activation/sve/qsymm16.cpp",
-        "src/core/cpu/kernels/add/neon/qasymm8.cpp",
-        "src/core/cpu/kernels/add/neon/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/add/neon/qsymm16.cpp",
-        "src/core/cpu/kernels/add/sve/impl.cpp",
-        "src/core/cpu/kernels/add/sve/qasymm8.cpp",
-        "src/core/cpu/kernels/add/sve/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/add/sve/qsymm16.cpp",
-        "src/core/cpu/kernels/elementwise/sve/elementwise.cpp",
-        "src/core/cpu/kernels/elementwise/sve/elementwise_unary.cpp",
-        "src/core/cpu/kernels/floor/neon/fp16.cpp",
-        "src/core/cpu/kernels/floor/neon/fp32.cpp",
-        "src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp",
-        "src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp",
-        "src/core/cpu/kernels/pool2d/neon/fp16.cpp",
-        "src/core/cpu/kernels/pool2d/neon/fp32.cpp",
-        "src/core/cpu/kernels/pool2d/neon/nchw/all.cpp",
-        "src/core/cpu/kernels/pool2d/neon/qasymm8.cpp",
-        "src/core/cpu/kernels/pool2d/neon/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/scale/neon/fp16.cpp",
-        "src/core/cpu/kernels/scale/neon/integer.cpp",
-        "src/core/cpu/kernels/scale/neon/qasymm8.cpp",
-        "src/core/cpu/kernels/scale/neon/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/scale/sve/fp16.cpp",
-        "src/core/cpu/kernels/scale/sve/fp32.cpp",
-        "src/core/cpu/kernels/scale/sve/integer.cpp",
-        "src/core/cpu/kernels/scale/sve/qasymm8.cpp",
-        "src/core/cpu/kernels/scale/sve/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/softmax/impl/sve/impl.cpp",
-        "src/core/cpu/kernels/sub/neon/qasymm8.cpp",
-        "src/core/cpu/kernels/sub/neon/qasymm8_signed.cpp",
-        "src/core/cpu/kernels/sub/neon/qsymm16.cpp",
-        "src/core/gpu/cl/ClKernelLibrary.cpp",
-        "src/core/gpu/cl/kernels/ClActivationKernel.cpp",
-        "src/core/gpu/cl/kernels/ClBatchConcatenateKernel.cpp",
-        "src/core/gpu/cl/kernels/ClCastKernel.cpp",
-        "src/core/gpu/cl/kernels/ClCol2ImKernel.cpp",
-        "src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp",
-        "src/core/gpu/cl/kernels/ClCopyKernel.cpp",
-        "src/core/gpu/cl/kernels/ClCropKernel.cpp",
-        "src/core/gpu/cl/kernels/ClDepthConcatenateKernel.cpp",
-        "src/core/gpu/cl/kernels/ClDequantizeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClDirectConv2dKernel.cpp",
-        "src/core/gpu/cl/kernels/ClElementwiseKernel.cpp",
-        "src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp",
-        "src/core/gpu/cl/kernels/ClFillKernel.cpp",
-        "src/core/gpu/cl/kernels/ClFloorKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp",
-        "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp",
-        "src/core/gpu/cl/kernels/ClHeightConcatenateKernel.cpp",
-        "src/core/gpu/cl/kernels/ClIm2ColKernel.cpp",
-        "src/core/gpu/cl/kernels/ClMulKernel.cpp",
-        "src/core/gpu/cl/kernels/ClPermuteKernel.cpp",
-        "src/core/gpu/cl/kernels/ClPool2dKernel.cpp",
-        "src/core/gpu/cl/kernels/ClQuantizeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClReshapeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClScaleKernel.cpp",
-        "src/core/gpu/cl/kernels/ClSoftmaxKernel.cpp",
-        "src/core/gpu/cl/kernels/ClTransposeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWidthConcatenateKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp",
-        "src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp",
-        "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.cpp",
-        "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp",
-        "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp",
-        "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp",
-        "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp",
-        "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp",
-        "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp",
-        "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp",
         "src/core/helpers/SoftmaxHelpers.cpp",
         "src/core/helpers/WindowHelpers.cpp",
         "src/core/utils/AssemblyUtils.cpp",
@@ -400,9 +261,213 @@ cc_library_static {
         "src/cpu/CpuContext.cpp",
         "src/cpu/CpuQueue.cpp",
         "src/cpu/CpuTensor.cpp",
+        "src/cpu/kernels/CpuActivationKernel.cpp",
+        "src/cpu/kernels/CpuAddKernel.cpp",
+        "src/cpu/kernels/CpuCastKernel.cpp",
+        "src/cpu/kernels/CpuCol2ImKernel.cpp",
+        "src/cpu/kernels/CpuConcatenateBatchKernel.cpp",
+        "src/cpu/kernels/CpuConcatenateDepthKernel.cpp",
+        "src/cpu/kernels/CpuConcatenateHeightKernel.cpp",
+        "src/cpu/kernels/CpuConcatenateWidthKernel.cpp",
+        "src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp",
+        "src/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp",
+        "src/cpu/kernels/CpuCopyKernel.cpp",
+        "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp",
+        "src/cpu/kernels/CpuDequantizeKernel.cpp",
+        "src/cpu/kernels/CpuDirectConv2dKernel.cpp",
+        "src/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp",
+        "src/cpu/kernels/CpuElementwiseKernel.cpp",
+        "src/cpu/kernels/CpuElementwiseUnaryKernel.cpp",
+        "src/cpu/kernels/CpuFillKernel.cpp",
+        "src/cpu/kernels/CpuFloorKernel.cpp",
+        "src/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp",
+        "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp",
+        "src/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp",
+        "src/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp",
+        "src/cpu/kernels/CpuGemmTranspose1xWKernel.cpp",
+        "src/cpu/kernels/CpuIm2ColKernel.cpp",
+        "src/cpu/kernels/CpuMulKernel.cpp",
+        "src/cpu/kernels/CpuPermuteKernel.cpp",
+        "src/cpu/kernels/CpuPool2dKernel.cpp",
+        "src/cpu/kernels/CpuQuantizeKernel.cpp",
+        "src/cpu/kernels/CpuReshapeKernel.cpp",
+        "src/cpu/kernels/CpuScaleKernel.cpp",
+        "src/cpu/kernels/CpuSoftmaxKernel.cpp",
+        "src/cpu/kernels/CpuSubKernel.cpp",
+        "src/cpu/kernels/CpuTransposeKernel.cpp",
+        "src/cpu/kernels/CpuWeightsReshapeKernel.cpp",
+        "src/cpu/kernels/CpuWinogradConv2dKernel.cpp",
+        "src/cpu/kernels/activation/neon/fp16.cpp",
+        "src/cpu/kernels/activation/neon/fp32.cpp",
+        "src/cpu/kernels/activation/neon/qasymm8.cpp",
+        "src/cpu/kernels/activation/neon/qasymm8_signed.cpp",
+        "src/cpu/kernels/activation/neon/qsymm16.cpp",
+        "src/cpu/kernels/activation/sve/fp16.cpp",
+        "src/cpu/kernels/activation/sve/fp32.cpp",
+        "src/cpu/kernels/activation/sve/qasymm8.cpp",
+        "src/cpu/kernels/activation/sve/qasymm8_signed.cpp",
+        "src/cpu/kernels/activation/sve/qsymm16.cpp",
+        "src/cpu/kernels/add/neon/qasymm8.cpp",
+        "src/cpu/kernels/add/neon/qasymm8_signed.cpp",
+        "src/cpu/kernels/add/neon/qsymm16.cpp",
+        "src/cpu/kernels/add/sve/impl.cpp",
+        "src/cpu/kernels/add/sve/qasymm8.cpp",
+        "src/cpu/kernels/add/sve/qasymm8_signed.cpp",
+        "src/cpu/kernels/add/sve/qsymm16.cpp",
+        "src/cpu/kernels/elementwise/sve/elementwise.cpp",
+        "src/cpu/kernels/elementwise/sve/elementwise_unary.cpp",
+        "src/cpu/kernels/floor/neon/fp16.cpp",
+        "src/cpu/kernels/floor/neon/fp32.cpp",
+        "src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp",
+        "src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp",
+        "src/cpu/kernels/pool2d/neon/fp16.cpp",
+        "src/cpu/kernels/pool2d/neon/fp32.cpp",
+        "src/cpu/kernels/pool2d/neon/nchw/all.cpp",
+        "src/cpu/kernels/pool2d/neon/qasymm8.cpp",
+        "src/cpu/kernels/pool2d/neon/qasymm8_signed.cpp",
+        "src/cpu/kernels/scale/neon/fp16.cpp",
+        "src/cpu/kernels/scale/neon/integer.cpp",
+        "src/cpu/kernels/scale/neon/qasymm8.cpp",
+        "src/cpu/kernels/scale/neon/qasymm8_signed.cpp",
+        "src/cpu/kernels/scale/sve/fp16.cpp",
+        "src/cpu/kernels/scale/sve/fp32.cpp",
+        "src/cpu/kernels/scale/sve/integer.cpp",
+        "src/cpu/kernels/scale/sve/qasymm8.cpp",
+        "src/cpu/kernels/scale/sve/qasymm8_signed.cpp",
+        "src/cpu/kernels/softmax/impl/sve/impl.cpp",
+        "src/cpu/kernels/sub/neon/qasymm8.cpp",
+        "src/cpu/kernels/sub/neon/qasymm8_signed.cpp",
+        "src/cpu/kernels/sub/neon/qsymm16.cpp",
+        "src/cpu/operators/CpuActivation.cpp",
+        "src/cpu/operators/CpuAdd.cpp",
+        "src/cpu/operators/CpuCast.cpp",
+        "src/cpu/operators/CpuConcatenate.cpp",
+        "src/cpu/operators/CpuConv2d.cpp",
+        "src/cpu/operators/CpuConvertFullyConnectedWeights.cpp",
+        "src/cpu/operators/CpuCopy.cpp",
+        "src/cpu/operators/CpuDepthwiseConv2d.cpp",
+        "src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp",
+        "src/cpu/operators/CpuDequantize.cpp",
+        "src/cpu/operators/CpuDirectConv2d.cpp",
+        "src/cpu/operators/CpuElementwise.cpp",
+        "src/cpu/operators/CpuElementwiseUnary.cpp",
+        "src/cpu/operators/CpuFill.cpp",
+        "src/cpu/operators/CpuFlatten.cpp",
+        "src/cpu/operators/CpuFloor.cpp",
+        "src/cpu/operators/CpuFullyConnected.cpp",
+        "src/cpu/operators/CpuGemm.cpp",
+        "src/cpu/operators/CpuGemmConv2d.cpp",
+        "src/cpu/operators/CpuGemmDirectConv2d.cpp",
+        "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp",
+        "src/cpu/operators/CpuGemmLowpOutputStage.cpp",
+        "src/cpu/operators/CpuMul.cpp",
+        "src/cpu/operators/CpuPermute.cpp",
+        "src/cpu/operators/CpuPool2d.cpp",
+        "src/cpu/operators/CpuQuantize.cpp",
+        "src/cpu/operators/CpuReshape.cpp",
+        "src/cpu/operators/CpuScale.cpp",
+        "src/cpu/operators/CpuSoftmax.cpp",
+        "src/cpu/operators/CpuSub.cpp",
+        "src/cpu/operators/CpuTranspose.cpp",
+        "src/cpu/operators/CpuWinogradConv2d.cpp",
+        "src/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp",
         "src/gpu/cl/ClContext.cpp",
+        "src/gpu/cl/ClKernelLibrary.cpp",
         "src/gpu/cl/ClQueue.cpp",
         "src/gpu/cl/ClTensor.cpp",
+        "src/gpu/cl/kernels/ClActivationKernel.cpp",
+        "src/gpu/cl/kernels/ClBatchConcatenateKernel.cpp",
+        "src/gpu/cl/kernels/ClCastKernel.cpp",
+        "src/gpu/cl/kernels/ClCol2ImKernel.cpp",
+        "src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp",
+        "src/gpu/cl/kernels/ClCopyKernel.cpp",
+        "src/gpu/cl/kernels/ClCropKernel.cpp",
+        "src/gpu/cl/kernels/ClDepthConcatenateKernel.cpp",
+        "src/gpu/cl/kernels/ClDequantizeKernel.cpp",
+        "src/gpu/cl/kernels/ClDirectConv2dKernel.cpp",
+        "src/gpu/cl/kernels/ClElementwiseKernel.cpp",
+        "src/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp",
+        "src/gpu/cl/kernels/ClFillKernel.cpp",
+        "src/gpu/cl/kernels/ClFloorKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp",
+        "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp",
+        "src/gpu/cl/kernels/ClHeightConcatenateKernel.cpp",
+        "src/gpu/cl/kernels/ClIm2ColKernel.cpp",
+        "src/gpu/cl/kernels/ClMulKernel.cpp",
+        "src/gpu/cl/kernels/ClPermuteKernel.cpp",
+        "src/gpu/cl/kernels/ClPool2dKernel.cpp",
+        "src/gpu/cl/kernels/ClQuantizeKernel.cpp",
+        "src/gpu/cl/kernels/ClReshapeKernel.cpp",
+        "src/gpu/cl/kernels/ClScaleKernel.cpp",
+        "src/gpu/cl/kernels/ClSoftmaxKernel.cpp",
+        "src/gpu/cl/kernels/ClTransposeKernel.cpp",
+        "src/gpu/cl/kernels/ClWeightsReshapeKernel.cpp",
+        "src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp",
+        "src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp",
+        "src/gpu/cl/kernels/ClWidthConcatenateKernel.cpp",
+        "src/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp",
+        "src/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp",
+        "src/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp",
+        "src/gpu/cl/kernels/gemm/ClGemmHelpers.cpp",
+        "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp",
+        "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp",
+        "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp",
+        "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp",
+        "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp",
+        "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp",
+        "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp",
+        "src/gpu/cl/operators/ClActivation.cpp",
+        "src/gpu/cl/operators/ClAdd.cpp",
+        "src/gpu/cl/operators/ClCast.cpp",
+        "src/gpu/cl/operators/ClConcatenate.cpp",
+        "src/gpu/cl/operators/ClConv2d.cpp",
+        "src/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp",
+        "src/gpu/cl/operators/ClCopy.cpp",
+        "src/gpu/cl/operators/ClCrop.cpp",
+        "src/gpu/cl/operators/ClDequantize.cpp",
+        "src/gpu/cl/operators/ClDirectConv2d.cpp",
+        "src/gpu/cl/operators/ClElementwiseOperations.cpp",
+        "src/gpu/cl/operators/ClElementwiseUnary.cpp",
+        "src/gpu/cl/operators/ClFill.cpp",
+        "src/gpu/cl/operators/ClFlatten.cpp",
+        "src/gpu/cl/operators/ClFloor.cpp",
+        "src/gpu/cl/operators/ClFullyConnected.cpp",
+        "src/gpu/cl/operators/ClGemm.cpp",
+        "src/gpu/cl/operators/ClGemmConv2d.cpp",
+        "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp",
+        "src/gpu/cl/operators/ClGemmLowpOutputStage.cpp",
+        "src/gpu/cl/operators/ClLogicalNot.cpp",
+        "src/gpu/cl/operators/ClMul.cpp",
+        "src/gpu/cl/operators/ClPRelu.cpp",
+        "src/gpu/cl/operators/ClPermute.cpp",
+        "src/gpu/cl/operators/ClPool2d.cpp",
+        "src/gpu/cl/operators/ClQuantize.cpp",
+        "src/gpu/cl/operators/ClReshape.cpp",
+        "src/gpu/cl/operators/ClScale.cpp",
+        "src/gpu/cl/operators/ClSoftmax.cpp",
+        "src/gpu/cl/operators/ClSub.cpp",
+        "src/gpu/cl/operators/ClTranspose.cpp",
+        "src/gpu/cl/operators/ClWinogradConv2d.cpp",
         "src/runtime/Allocator.cpp",
         "src/runtime/BlobLifetimeManager.cpp",
         "src/runtime/BlobMemoryPool.cpp",
@@ -629,71 +694,6 @@ cc_library_static {
         "src/runtime/Tensor.cpp",
         "src/runtime/TensorAllocator.cpp",
         "src/runtime/Utils.cpp",
-        "src/runtime/cpu/operators/CpuActivation.cpp",
-        "src/runtime/cpu/operators/CpuAdd.cpp",
-        "src/runtime/cpu/operators/CpuCast.cpp",
-        "src/runtime/cpu/operators/CpuConcatenate.cpp",
-        "src/runtime/cpu/operators/CpuConv2d.cpp",
-        "src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.cpp",
-        "src/runtime/cpu/operators/CpuCopy.cpp",
-        "src/runtime/cpu/operators/CpuDepthwiseConv2d.cpp",
-        "src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp",
-        "src/runtime/cpu/operators/CpuDequantize.cpp",
-        "src/runtime/cpu/operators/CpuDirectConv2d.cpp",
-        "src/runtime/cpu/operators/CpuElementwise.cpp",
-        "src/runtime/cpu/operators/CpuElementwiseUnary.cpp",
-        "src/runtime/cpu/operators/CpuFill.cpp",
-        "src/runtime/cpu/operators/CpuFlatten.cpp",
-        "src/runtime/cpu/operators/CpuFloor.cpp",
-        "src/runtime/cpu/operators/CpuFullyConnected.cpp",
-        "src/runtime/cpu/operators/CpuGemm.cpp",
-        "src/runtime/cpu/operators/CpuGemmConv2d.cpp",
-        "src/runtime/cpu/operators/CpuGemmDirectConv2d.cpp",
-        "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp",
-        "src/runtime/cpu/operators/CpuGemmLowpOutputStage.cpp",
-        "src/runtime/cpu/operators/CpuMul.cpp",
-        "src/runtime/cpu/operators/CpuPermute.cpp",
-        "src/runtime/cpu/operators/CpuPool2d.cpp",
-        "src/runtime/cpu/operators/CpuQuantize.cpp",
-        "src/runtime/cpu/operators/CpuReshape.cpp",
-        "src/runtime/cpu/operators/CpuScale.cpp",
-        "src/runtime/cpu/operators/CpuSoftmax.cpp",
-        "src/runtime/cpu/operators/CpuSub.cpp",
-        "src/runtime/cpu/operators/CpuTranspose.cpp",
-        "src/runtime/cpu/operators/CpuWinogradConv2d.cpp",
-        "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp",
-        "src/runtime/gpu/cl/operators/ClActivation.cpp",
-        "src/runtime/gpu/cl/operators/ClAdd.cpp",
-        "src/runtime/gpu/cl/operators/ClCast.cpp",
-        "src/runtime/gpu/cl/operators/ClConcatenate.cpp",
-        "src/runtime/gpu/cl/operators/ClConv2d.cpp",
-        "src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp",
-        "src/runtime/gpu/cl/operators/ClCopy.cpp",
-        "src/runtime/gpu/cl/operators/ClCrop.cpp",
-        "src/runtime/gpu/cl/operators/ClDequantize.cpp",
-        "src/runtime/gpu/cl/operators/ClDirectConv2d.cpp",
-        "src/runtime/gpu/cl/operators/ClElementwiseOperations.cpp",
-        "src/runtime/gpu/cl/operators/ClElementwiseUnary.cpp",
-        "src/runtime/gpu/cl/operators/ClFill.cpp",
-        "src/runtime/gpu/cl/operators/ClFlatten.cpp",
-        "src/runtime/gpu/cl/operators/ClFloor.cpp",
-        "src/runtime/gpu/cl/operators/ClFullyConnected.cpp",
-        "src/runtime/gpu/cl/operators/ClGemm.cpp",
-        "src/runtime/gpu/cl/operators/ClGemmConv2d.cpp",
-        "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp",
-        "src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.cpp",
-        "src/runtime/gpu/cl/operators/ClLogicalNot.cpp",
-        "src/runtime/gpu/cl/operators/ClMul.cpp",
-        "src/runtime/gpu/cl/operators/ClPRelu.cpp",
-        "src/runtime/gpu/cl/operators/ClPermute.cpp",
-        "src/runtime/gpu/cl/operators/ClPool2d.cpp",
-        "src/runtime/gpu/cl/operators/ClQuantize.cpp",
-        "src/runtime/gpu/cl/operators/ClReshape.cpp",
-        "src/runtime/gpu/cl/operators/ClScale.cpp",
-        "src/runtime/gpu/cl/operators/ClSoftmax.cpp",
-        "src/runtime/gpu/cl/operators/ClSub.cpp",
-        "src/runtime/gpu/cl/operators/ClTranspose.cpp",
-        "src/runtime/gpu/cl/operators/ClWinogradConv2d.cpp",
         "utils/CommonGraphOptions.cpp",
         "utils/GraphUtils.cpp",
         "utils/Utils.cpp",
diff --git a/SConscript b/SConscript
index c384597752..62ff266d34 100644
--- a/SConscript
+++ b/SConscript
@@ -457,7 +457,7 @@ if env['neon']:
                                       "src/core/NEON/kernels/convolution/depthwise/",
                                       "src/core/NEON/kernels/assembly/",
                                       "arm_compute/core/NEON/kernels/assembly/",
-                                      "src/core/cpu/kernels/assembly/",])
+                                      "src/cpu/kernels/assembly/",])
 
     # Load files based on user's options
     operators = filelist['cpu']['operators']
diff --git a/docs/ComputeLibrary.dir b/docs/ComputeLibrary.dir
index e08f05eb2d..a57b3cb3da 100644
--- a/docs/ComputeLibrary.dir
+++ b/docs/ComputeLibrary.dir
@@ -230,7 +230,7 @@
  *  @brief Scalar operations
  */
 
-/** @dir src/core/gpu/cl/kernels/gemm
+/** @dir src/gpu/cl/kernels/gemm
  *  @brief Folder containing all the configuration files for GEMM
  */
 
diff --git a/docs/contributor_guide/adding_operator.dox b/docs/contributor_guide/adding_operator.dox
index 772d4362c8..ab174838cc 100644
--- a/docs/contributor_guide/adding_operator.dox
+++ b/docs/contributor_guide/adding_operator.dox
@@ -119,11 +119,11 @@ Each kernel will have to implement the method:
 
 The structure of the kernel .cpp file should be similar to the next ones.
 For OpenCL:
-@snippet src/core/gpu/cl/kernels/ClReshapeKernel.cpp ClReshapeKernel Kernel
+@snippet src/gpu/cl/kernels/ClReshapeKernel.cpp ClReshapeKernel Kernel
 The run will call the function defined in the .cl file.
 
 For the Arm® Neon™ backend case:
-@snippet src/core/cpu/kernels/CpuReshapeKernel.cpp NEReshapeLayerKernel Kernel
+@snippet src/cpu/kernels/CpuReshapeKernel.cpp NEReshapeLayerKernel Kernel
 
 In the Arm® Neon™ case, there is no need to add an extra file and we implement the kernel in the same NEReshapeLayerKernel.cpp file.
 If the tests are already in place, the new kernel can be tested using the existing tests by adding the configure and run of the kernel to the compute_target() in the fixture.
diff --git a/examples/gemm_tuner/cl_gemm_native.cpp b/examples/gemm_tuner/cl_gemm_native.cpp
index 093935f716..dd03873921 100644
--- a/examples/gemm_tuner/cl_gemm_native.cpp
+++ b/examples/gemm_tuner/cl_gemm_native.cpp
@@ -32,7 +32,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/CLTuner.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
diff --git a/examples/gemm_tuner/cl_gemm_reshaped.cpp b/examples/gemm_tuner/cl_gemm_reshaped.cpp
index 08d46feb5f..bbc42a7e55 100644
--- a/examples/gemm_tuner/cl_gemm_reshaped.cpp
+++ b/examples/gemm_tuner/cl_gemm_reshaped.cpp
@@ -33,8 +33,8 @@
 #include "arm_compute/runtime/CL/CLTuner.h"
 #include "examples/gemm_tuner/CommonGemmExampleOptions.h"
 #include "examples/gemm_tuner/GemmTunerHelpers.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
diff --git a/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp b/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
index dbaaca6048..79f251ae7d 100644
--- a/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
+++ b/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
@@ -33,7 +33,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/CLTuner.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
diff --git a/examples/gemm_tuner/cl_gemmlowp_reshaped.cpp b/examples/gemm_tuner/cl_gemmlowp_reshaped.cpp
index 3c8976fddd..115ce6ba23 100644
--- a/examples/gemm_tuner/cl_gemmlowp_reshaped.cpp
+++ b/examples/gemm_tuner/cl_gemmlowp_reshaped.cpp
@@ -33,8 +33,8 @@
 #include "arm_compute/runtime/CL/CLTuner.h"
 #include "examples/gemm_tuner/CommonGemmExampleOptions.h"
 #include "examples/gemm_tuner/GemmTunerHelpers.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
diff --git a/examples/gemm_tuner/cl_gemmlowp_reshaped_rhs_only_fused_output_stage_fixedpoint.cpp b/examples/gemm_tuner/cl_gemmlowp_reshaped_rhs_only_fused_output_stage_fixedpoint.cpp
index 15c1b86c61..0e0917daa9 100644
--- a/examples/gemm_tuner/cl_gemmlowp_reshaped_rhs_only_fused_output_stage_fixedpoint.cpp
+++ b/examples/gemm_tuner/cl_gemmlowp_reshaped_rhs_only_fused_output_stage_fixedpoint.cpp
@@ -35,8 +35,8 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/CLTuner.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
diff --git a/filelist.json b/filelist.json
index d9c67126c6..8aa869a909 100644
--- a/filelist.json
+++ b/filelist.json
@@ -41,15 +41,15 @@
       "src/core/CL/ICLSimpleKernel.cpp",
       "src/core/CL/ICLTensor.cpp",
       "src/core/CL/OpenCL.cpp",
-      "src/core/gpu/cl/ClKernelLibrary.cpp",
-      "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.cpp",
-      "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp",
-      "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp",
-      "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp",
-      "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp",
-      "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp",
-      "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp",
-      "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp",
+      "src/gpu/cl/ClKernelLibrary.cpp",
+      "src/gpu/cl/kernels/gemm/ClGemmHelpers.cpp",
+      "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp",
+      "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp",
+      "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp",
+      "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp",
+      "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp",
+      "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp",
+      "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp",
       "src/core/CL/kernels/CLFillBorderKernel.cpp",
       "src/gpu/cl/ClContext.cpp",
       "src/gpu/cl/ClQueue.cpp",
@@ -90,52 +90,52 @@
       "Activation": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClActivation.cpp"
+            "src/gpu/cl/operators/ClActivation.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClActivationKernel.cpp"
+            "src/gpu/cl/kernels/ClActivationKernel.cpp"
           ]
         }
       },
       "Add": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClAdd.cpp"
+            "src/gpu/cl/operators/ClAdd.cpp"
           ]
         }
       },
       "Cast": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClCast.cpp"
+            "src/gpu/cl/operators/ClCast.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClCastKernel.cpp"
+            "src/gpu/cl/kernels/ClCastKernel.cpp"
           ]
         }
       },
       "Concatenate": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClConcatenate.cpp"
+            "src/gpu/cl/operators/ClConcatenate.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp",
-            "src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp",
-            "src/core/gpu/cl/kernels/ClWidthConcatenateKernel.cpp",
-            "src/core/gpu/cl/kernels/ClHeightConcatenateKernel.cpp",
-            "src/core/gpu/cl/kernels/ClDepthConcatenateKernel.cpp",
-            "src/core/gpu/cl/kernels/ClBatchConcatenateKernel.cpp"
+            "src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp",
+            "src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp",
+            "src/gpu/cl/kernels/ClWidthConcatenateKernel.cpp",
+            "src/gpu/cl/kernels/ClHeightConcatenateKernel.cpp",
+            "src/gpu/cl/kernels/ClDepthConcatenateKernel.cpp",
+            "src/gpu/cl/kernels/ClBatchConcatenateKernel.cpp"
           ]
         }
       },
       "DirectConv2d": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClDirectConv2d.cpp"
+            "src/gpu/cl/operators/ClDirectConv2d.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClDirectConv2dKernel.cpp"
+            "src/gpu/cl/kernels/ClDirectConv2dKernel.cpp"
           ]
         }
       },
@@ -149,230 +149,230 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClFullyConnected.cpp"
+            "src/gpu/cl/operators/ClFullyConnected.cpp"
           ]
         }
       },
       "ConvertFullyConnectedWeights": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp"
+            "src/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp"
+            "src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp"
           ]
         }
       },
       "Permute": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClPermute.cpp"
+            "src/gpu/cl/operators/ClPermute.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClPermuteKernel.cpp"
+            "src/gpu/cl/kernels/ClPermuteKernel.cpp"
           ]
         }
       },
       "Pool2d": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClPool2d.cpp"
+            "src/gpu/cl/operators/ClPool2d.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClPool2dKernel.cpp"
+            "src/gpu/cl/kernels/ClPool2dKernel.cpp"
           ]
         }
       },
       "Conv2d": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClConv2d.cpp"
+            "src/gpu/cl/operators/ClConv2d.cpp"
           ]
         }
       },
       "PRelu": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClPRelu.cpp"
+            "src/gpu/cl/operators/ClPRelu.cpp"
           ]
         }
       },
       "Reshape": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClReshape.cpp"
+            "src/gpu/cl/operators/ClReshape.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClReshapeKernel.cpp"
+            "src/gpu/cl/kernels/ClReshapeKernel.cpp"
           ]
         }
       },
       "Copy": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClCopy.cpp"
+            "src/gpu/cl/operators/ClCopy.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClCopyKernel.cpp"
+            "src/gpu/cl/kernels/ClCopyKernel.cpp"
           ]
         }
       },
       "Crop": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClCrop.cpp"
+            "src/gpu/cl/operators/ClCrop.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClCropKernel.cpp"
+            "src/gpu/cl/kernels/ClCropKernel.cpp"
           ]
         }
       },
       "Dequantize": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClDequantize.cpp"
+            "src/gpu/cl/operators/ClDequantize.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClDequantizeKernel.cpp"
+            "src/gpu/cl/kernels/ClDequantizeKernel.cpp"
           ]
         }
       },
       "Elementwise": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClElementwiseOperations.cpp"
+            "src/gpu/cl/operators/ClElementwiseOperations.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClElementwiseKernel.cpp"
+            "src/gpu/cl/kernels/ClElementwiseKernel.cpp"
           ]
         }
       },
       "ElementwiseUnary": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClElementwiseUnary.cpp"
+            "src/gpu/cl/operators/ClElementwiseUnary.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp"
+            "src/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp"
           ]
         }
       },
       "Fill": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClFill.cpp"
+            "src/gpu/cl/operators/ClFill.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClFillKernel.cpp"
+            "src/gpu/cl/kernels/ClFillKernel.cpp"
           ]
         }
       },
       "Flatten": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClFlatten.cpp"
+            "src/gpu/cl/operators/ClFlatten.cpp"
           ]
         }
       },
       "Floor": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClFloor.cpp"
+            "src/gpu/cl/operators/ClFloor.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClFloorKernel.cpp"
+            "src/gpu/cl/kernels/ClFloorKernel.cpp"
           ]
         }
       },
       "GEMM": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClGemm.cpp",
-            "src/runtime/gpu/cl/operators/ClGemmConv2d.cpp"
+            "src/gpu/cl/operators/ClGemm.cpp",
+            "src/gpu/cl/operators/ClGemmConv2d.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp"
+            "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp"
           ]
         }
       },
       "GEMMLowp": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp",
-            "src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.cpp"
+            "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp",
+            "src/gpu/cl/operators/ClGemmLowpOutputStage.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp",
-            "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp" 
+            "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp",
+            "src/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp" 
           ]
         }
       },
       "Mul": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClMul.cpp"
+            "src/gpu/cl/operators/ClMul.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClMulKernel.cpp"
+            "src/gpu/cl/kernels/ClMulKernel.cpp"
           ]
         }
       },
       "Quantize": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClQuantize.cpp"
+            "src/gpu/cl/operators/ClQuantize.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClQuantizeKernel.cpp"
+            "src/gpu/cl/kernels/ClQuantizeKernel.cpp"
           ]
         }
       },
       "Scale": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClScale.cpp"
+            "src/gpu/cl/operators/ClScale.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClScaleKernel.cpp"
+            "src/gpu/cl/kernels/ClScaleKernel.cpp"
           ]
         }
       },
       "Softmax": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClSoftmax.cpp"
+            "src/gpu/cl/operators/ClSoftmax.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClSoftmaxKernel.cpp"
+            "src/gpu/cl/kernels/ClSoftmaxKernel.cpp"
           ]
         }
       },
       "Sub": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClSub.cpp"
+            "src/gpu/cl/operators/ClSub.cpp"
           ]
         }
       },
       "Transpose": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClTranspose.cpp"
+            "src/gpu/cl/operators/ClTranspose.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClTransposeKernel.cpp"
+            "src/gpu/cl/kernels/ClTransposeKernel.cpp"
           ]
         }
       },
@@ -428,8 +428,8 @@
       "GEMMConv2d": {
         "files": {
           "kernel": [
-            "src/core/gpu/cl/kernels/ClCol2ImKernel.cpp",
-            "src/core/gpu/cl/kernels/ClIm2ColKernel.cpp"
+            "src/gpu/cl/kernels/ClCol2ImKernel.cpp",
+            "src/gpu/cl/kernels/ClIm2ColKernel.cpp"
           ]
         }
       },
@@ -520,7 +520,7 @@
       "LogicalNot": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClLogicalNot.cpp"
+            "src/gpu/cl/operators/ClLogicalNot.cpp"
           ]
         }
       },
@@ -674,19 +674,19 @@
       "WeightsReshape": {
         "files": {
           "kernel": [
-            "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.cpp"
+            "src/gpu/cl/kernels/ClWeightsReshapeKernel.cpp"
           ]
         }
       },
       "WinogradConv2d": {
         "files": {
           "operator": [
-            "src/runtime/gpu/cl/operators/ClWinogradConv2d.cpp"
+            "src/gpu/cl/operators/ClWinogradConv2d.cpp"
           ],
           "kernel": [
-            "src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp",
-            "src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp",
-            "src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp"
+            "src/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp",
+            "src/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp",
+            "src/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp"
           ]
         }
       }
@@ -711,43 +711,43 @@
       "Activation": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuActivation.cpp"
+            "src/cpu/operators/CpuActivation.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuActivationKernel.cpp"
+            "src/cpu/kernels/CpuActivationKernel.cpp"
           ],
           "sve": {
             "fp32": [
-              "src/core/cpu/kernels/activation/sve/fp32.cpp"
+              "src/cpu/kernels/activation/sve/fp32.cpp"
             ],
             "fp16": [
-              "src/core/cpu/kernels/activation/sve/fp16.cpp"
+              "src/cpu/kernels/activation/sve/fp16.cpp"
             ],
             "qsymm16": [
-              "src/core/cpu/kernels/activation/sve/qsymm16.cpp"
+              "src/cpu/kernels/activation/sve/qsymm16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/activation/sve/qasymm8.cpp"
+              "src/cpu/kernels/activation/sve/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/activation/sve/qasymm8_signed.cpp"
+              "src/cpu/kernels/activation/sve/qasymm8_signed.cpp"
             ]
           },
           "neon": {
             "fp32": [
-              "src/core/cpu/kernels/activation/neon/fp32.cpp"
+              "src/cpu/kernels/activation/neon/fp32.cpp"
             ],
             "fp16": [
-              "src/core/cpu/kernels/activation/neon/fp16.cpp"
+              "src/cpu/kernels/activation/neon/fp16.cpp"
             ],
             "qsymm16": [
-              "src/core/cpu/kernels/activation/neon/qsymm16.cpp"
+              "src/cpu/kernels/activation/neon/qsymm16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/activation/neon/qasymm8.cpp"
+              "src/cpu/kernels/activation/neon/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/activation/neon/qasymm8_signed.cpp"
+              "src/cpu/kernels/activation/neon/qasymm8_signed.cpp"
             ]
           }
         }
@@ -755,34 +755,34 @@
       "Add": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuAdd.cpp"
+            "src/cpu/operators/CpuAdd.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuAddKernel.cpp"
+            "src/cpu/kernels/CpuAddKernel.cpp"
           ],
           "sve": {
             "all": [
-              "src/core/cpu/kernels/add/sve/impl.cpp"
+              "src/cpu/kernels/add/sve/impl.cpp"
             ],
             "qsymm16": [
-              "src/core/cpu/kernels/add/sve/qsymm16.cpp"
+              "src/cpu/kernels/add/sve/qsymm16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/add/sve/qasymm8.cpp"
+              "src/cpu/kernels/add/sve/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/add/sve/qasymm8_signed.cpp"
+              "src/cpu/kernels/add/sve/qasymm8_signed.cpp"
             ]
           },
           "neon": {
             "qsymm16": [
-              "src/core/cpu/kernels/add/neon/qsymm16.cpp"
+              "src/cpu/kernels/add/neon/qsymm16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/add/neon/qasymm8.cpp"
+              "src/cpu/kernels/add/neon/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/add/neon/qasymm8_signed.cpp"
+              "src/cpu/kernels/add/neon/qasymm8_signed.cpp"
             ]
           }
         }
@@ -862,64 +862,64 @@
       "Col2Im": {
         "files": {
           "kernel": [
-            "src/core/cpu/kernels/CpuCol2ImKernel.cpp"
+            "src/cpu/kernels/CpuCol2ImKernel.cpp"
           ]
         }
       },
       "Cast": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuCast.cpp"
+            "src/cpu/operators/CpuCast.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuCastKernel.cpp"
+            "src/cpu/kernels/CpuCastKernel.cpp"
           ]
         }
       },
       "Concatenate": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuConcatenate.cpp"
+            "src/cpu/operators/CpuConcatenate.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuConcatenateWidthKernel.cpp",
-            "src/core/cpu/kernels/CpuConcatenateBatchKernel.cpp",
-            "src/core/cpu/kernels/CpuConcatenateDepthKernel.cpp",
-            "src/core/cpu/kernels/CpuConcatenateHeightKernel.cpp"
+            "src/cpu/kernels/CpuConcatenateWidthKernel.cpp",
+            "src/cpu/kernels/CpuConcatenateBatchKernel.cpp",
+            "src/cpu/kernels/CpuConcatenateDepthKernel.cpp",
+            "src/cpu/kernels/CpuConcatenateHeightKernel.cpp"
           ]
         }
       },
       "ConvertFullyConnectedWeights": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.cpp"
+            "src/cpu/operators/CpuConvertFullyConnectedWeights.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp"
+            "src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp"
           ]
         }
       },
       "ConvertQuantizedSignedness": {
         "files": {
           "kernel": [
-            "src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp"
+            "src/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp"
           ]
         }
       },
       "Convolution": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuConv2d.cpp"
+            "src/cpu/operators/CpuConv2d.cpp"
           ]
         }
       },
       "Copy": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuCopy.cpp"
+            "src/cpu/operators/CpuCopy.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuCopyKernel.cpp"
+            "src/cpu/kernels/CpuCopyKernel.cpp"
           ]
         }
       },
@@ -937,16 +937,16 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuDepthwiseConv2d.cpp",
-            "src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp",
-            "src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp"
+            "src/cpu/operators/CpuDepthwiseConv2d.cpp",
+            "src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp",
+            "src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp"
           ],
           "kernel": [
             "src/core/NEON/kernels/convolution/common/padding.cpp",
             "src/core/NEON/kernels/convolution/common/qasymm8.cpp",
             "src/core/NEON/kernels/convolution/common/qsymm8.cpp",
             "src/core/NEON/kernels/convolution/common/utils.cpp",
-            "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp"
+            "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp"
           ],
           "sve": {
             "all": [
@@ -1068,10 +1068,10 @@
       "Dequantize": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuDequantize.cpp"
+            "src/cpu/operators/CpuDequantize.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuDequantizeKernel.cpp"
+            "src/cpu/kernels/CpuDequantizeKernel.cpp"
           ]
         }
       },
@@ -1082,25 +1082,25 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuDirectConv2d.cpp"
+            "src/cpu/operators/CpuDirectConv2d.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuDirectConv2dKernel.cpp",
-            "src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp"
+            "src/cpu/kernels/CpuDirectConv2dKernel.cpp",
+            "src/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp"
           ]
         }
       },
       "Elementwise": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuElementwise.cpp"
+            "src/cpu/operators/CpuElementwise.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuElementwiseKernel.cpp"
+            "src/cpu/kernels/CpuElementwiseKernel.cpp"
           ],
           "sve": {
             "all": [
-              "src/core/cpu/kernels/elementwise/sve/elementwise.cpp"
+              "src/cpu/kernels/elementwise/sve/elementwise.cpp"
             ]
           }
         }
@@ -1108,14 +1108,14 @@
       "ElementwiseUnary": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuElementwiseUnary.cpp"
+            "src/cpu/operators/CpuElementwiseUnary.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuElementwiseUnaryKernel.cpp"
+            "src/cpu/kernels/CpuElementwiseUnaryKernel.cpp"
           ],
           "sve": {
             "all": [
-              "src/core/cpu/kernels/elementwise/sve/elementwise_unary.cpp"
+              "src/cpu/kernels/elementwise/sve/elementwise_unary.cpp"
             ]
           }
         }
@@ -1142,34 +1142,34 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuFlatten.cpp"
+            "src/cpu/operators/CpuFlatten.cpp"
           ]
         }
       },
       "Fill": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuFill.cpp"
+            "src/cpu/operators/CpuFill.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuFillKernel.cpp"
+            "src/cpu/kernels/CpuFillKernel.cpp"
           ]
         }
       },
       "Floor": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuFloor.cpp"
+            "src/cpu/operators/CpuFloor.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuFloorKernel.cpp"
+            "src/cpu/kernels/CpuFloorKernel.cpp"
           ],
           "neon": {
             "fp32": [
-              "src/core/cpu/kernels/floor/neon/fp32.cpp"
+              "src/cpu/kernels/floor/neon/fp32.cpp"
             ],
             "fp16": [
-              "src/core/cpu/kernels/floor/neon/fp16.cpp"
+              "src/cpu/kernels/floor/neon/fp16.cpp"
             ]
           }
         }
@@ -1183,7 +1183,7 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuFullyConnected.cpp"
+            "src/cpu/operators/CpuFullyConnected.cpp"
           ]
         },
         "kernel": [
@@ -1199,12 +1199,12 @@
       },
       "GEMM": {
         "files": {
-          "operator" : ["src/runtime/cpu/operators/CpuGemm.cpp"],
+          "operator" : ["src/cpu/operators/CpuGemm.cpp"],
           "kernel": [
-            "src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp"
+            "src/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp",
+            "src/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp",
+            "src/cpu/kernels/CpuGemmTranspose1xWKernel.cpp",
+            "src/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp"
           ]
         }
       },
@@ -1214,18 +1214,18 @@
         ],
         "files": {
           "operator" : [
-              "src/runtime/cpu/operators/CpuGemmLowpOutputStage.cpp",
-              "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp"
+              "src/cpu/operators/CpuGemmLowpOutputStage.cpp",
+              "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp",
-            "src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp"
+            "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp",
+            "src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp"
           ]
         }
       },
@@ -1240,17 +1240,17 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuGemmConv2d.cpp"
+            "src/cpu/operators/CpuGemmConv2d.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuWeightsReshapeKernel.cpp"
+            "src/cpu/kernels/CpuWeightsReshapeKernel.cpp"
           ]
         }
       },
       "GemmAssemblyDispatch": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp"
+            "src/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp"
           ],
           "kernel": [
             "src/core/NEON/kernels/arm_gemm/gemm_fp16.cpp",
@@ -1396,37 +1396,37 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuGemmDirectConv2d.cpp"
+            "src/cpu/operators/CpuGemmDirectConv2d.cpp"
           ]
         }
       },
       "Mul": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuMul.cpp"
+            "src/cpu/operators/CpuMul.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuMulKernel.cpp"
+            "src/cpu/kernels/CpuMulKernel.cpp"
           ]
         }
       },
       "Quantize": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuQuantize.cpp"
+            "src/cpu/operators/CpuQuantize.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuQuantizeKernel.cpp"
+            "src/cpu/kernels/CpuQuantizeKernel.cpp"
           ]
         }
       },
       "Reshape": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuReshape.cpp"
+            "src/cpu/operators/CpuReshape.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuReshapeKernel.cpp"
+            "src/cpu/kernels/CpuReshapeKernel.cpp"
           ]
         }
       },
@@ -1447,7 +1447,7 @@
       "Im2Col": {
         "files": {
           "kernel": [
-            "src/core/cpu/kernels/CpuIm2ColKernel.cpp"
+            "src/cpu/kernels/CpuIm2ColKernel.cpp"
           ]
         }
       },
@@ -1516,21 +1516,21 @@
       "Permute": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuPermute.cpp"
+            "src/cpu/operators/CpuPermute.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuPermuteKernel.cpp"
+            "src/cpu/kernels/CpuPermuteKernel.cpp"
           ]
         }
       },
       "Pool2d": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuPool2d.cpp"
+            "src/cpu/operators/CpuPool2d.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuPool2dKernel.cpp",
-            "src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp",
+            "src/cpu/kernels/CpuPool2dKernel.cpp",
+            "src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp",
             "src/core/NEON/kernels/arm_conv/pooling/kernels/cpp_nhwc_1x1_stride_any_depthfirst/generic.cpp",
             "src/core/NEON/kernels/arm_conv/pooling/pooling_fp16.cpp",
             "src/core/NEON/kernels/arm_conv/pooling/pooling_fp32.cpp",
@@ -1541,19 +1541,19 @@
           ],
           "neon": {
             "nchw": [
-              "src/core/cpu/kernels/pool2d/neon/nchw/all.cpp"
+              "src/cpu/kernels/pool2d/neon/nchw/all.cpp"
             ],
             "fp32": [
-              "src/core/cpu/kernels/pool2d/neon/fp32.cpp"
+              "src/cpu/kernels/pool2d/neon/fp32.cpp"
             ],
             "fp16": [
-              "src/core/cpu/kernels/pool2d/neon/fp16.cpp"
+              "src/cpu/kernels/pool2d/neon/fp16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/pool2d/neon/qasymm8.cpp"
+              "src/cpu/kernels/pool2d/neon/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/pool2d/neon/qasymm8_signed.cpp"
+              "src/cpu/kernels/pool2d/neon/qasymm8_signed.cpp"
             ],
             "estate64": [
               "src/core/NEON/kernels/arm_conv/pooling/kernels/a64_fp16_nhwc_avg_3x3_s1_output2x2_depthfirst/generic.cpp",
@@ -1701,40 +1701,40 @@
       "Scale": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuScale.cpp"
+            "src/cpu/operators/CpuScale.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuScaleKernel.cpp"
+            "src/cpu/kernels/CpuScaleKernel.cpp"
           ],
           "sve": {
             "fp32": [
-              "src/core/cpu/kernels/scale/sve/fp32.cpp"
+              "src/cpu/kernels/scale/sve/fp32.cpp"
             ],
             "fp16": [
-              "src/core/cpu/kernels/scale/sve/fp16.cpp"
+              "src/cpu/kernels/scale/sve/fp16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/scale/sve/qasymm8.cpp"
+              "src/cpu/kernels/scale/sve/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/scale/sve/qasymm8_signed.cpp"
+              "src/cpu/kernels/scale/sve/qasymm8_signed.cpp"
             ],
             "integer": [
-              "src/core/cpu/kernels/scale/sve/integer.cpp"
+              "src/cpu/kernels/scale/sve/integer.cpp"
             ]
           },
           "neon": {
             "fp16": [
-              "src/core/cpu/kernels/scale/neon/fp16.cpp"
+              "src/cpu/kernels/scale/neon/fp16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/scale/neon/qasymm8.cpp"
+              "src/cpu/kernels/scale/neon/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/scale/neon/qasymm8_signed.cpp"
+              "src/cpu/kernels/scale/neon/qasymm8_signed.cpp"
             ],
             "integer": [
-              "src/core/cpu/kernels/scale/neon/integer.cpp"
+              "src/cpu/kernels/scale/neon/integer.cpp"
             ]
           }
         }
@@ -1745,14 +1745,14 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuSoftmax.cpp"
+            "src/cpu/operators/CpuSoftmax.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuSoftmaxKernel.cpp"
+            "src/cpu/kernels/CpuSoftmaxKernel.cpp"
           ],
           "sve": {
             "all": [
-              "src/core/cpu/kernels/softmax/impl/sve/impl.cpp"
+              "src/cpu/kernels/softmax/impl/sve/impl.cpp"
             ]
           }
         }
@@ -1760,20 +1760,20 @@
       "Sub": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuSub.cpp"
+            "src/cpu/operators/CpuSub.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuSubKernel.cpp"
+            "src/cpu/kernels/CpuSubKernel.cpp"
           ],
           "neon": {
             "qsymm16": [
-              "src/core/cpu/kernels/sub/neon/qsymm16.cpp"
+              "src/cpu/kernels/sub/neon/qsymm16.cpp"
             ],
             "qasymm8": [
-              "src/core/cpu/kernels/sub/neon/qasymm8.cpp"
+              "src/cpu/kernels/sub/neon/qasymm8.cpp"
             ],
             "qasymm8_signed": [
-              "src/core/cpu/kernels/sub/neon/qasymm8_signed.cpp"
+              "src/cpu/kernels/sub/neon/qasymm8_signed.cpp"
             ]
           }
         }
@@ -1781,10 +1781,10 @@
       "Transpose": {
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuTranspose.cpp"
+            "src/cpu/operators/CpuTranspose.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuTransposeKernel.cpp"
+            "src/cpu/kernels/CpuTransposeKernel.cpp"
           ]
         }
       },
@@ -1802,10 +1802,10 @@
         ],
         "files": {
           "operator": [
-            "src/runtime/cpu/operators/CpuWinogradConv2d.cpp"
+            "src/cpu/operators/CpuWinogradConv2d.cpp"
           ],
           "kernel": [
-            "src/core/cpu/kernels/CpuWinogradConv2dKernel.cpp",
+            "src/cpu/kernels/CpuWinogradConv2dKernel.cpp",
             "src/core/NEON/kernels/convolution/winograd/padding.cpp",
             "src/core/NEON/kernels/convolution/winograd/winograd.cpp",
             "src/core/NEON/kernels/convolution/winograd/winograd_transforms/input_1x8_fp32_fp32_integers.cpp",
diff --git a/scripts/clang_tidy_rules.py b/scripts/clang_tidy_rules.py
index 8ab7c13a7c..d718af0f21 100755
--- a/scripts/clang_tidy_rules.py
+++ b/scripts/clang_tidy_rules.py
@@ -4,7 +4,7 @@ import re
 import sys
 
 def get_list_includes():
-    return "src/core/cpu/kernels/assembly " \
+    return "src/cpu/kernels/assembly " \
            "src/core/NEON/kernels/assembly " \
            "src/core/NEON/kernels/convolution/winograd " \
            "include/linux include " \
diff --git a/src/core/CL/CLHelpers.cpp b/src/core/CL/CLHelpers.cpp
index 5c53455eeb..10ccc4f9a4 100644
--- a/src/core/CL/CLHelpers.cpp
+++ b/src/core/CL/CLHelpers.cpp
@@ -27,9 +27,9 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Log.h"
 #include "arm_compute/core/Types.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/ClCompileContext.h"
 
-#include "src/core/gpu/cl/ClKernelLibrary.h"
+#include "src/gpu/cl/ClKernelLibrary.h"
 
 #include <utility>
 #include <vector>
diff --git a/src/core/CL/CLKernelLibrary.cpp b/src/core/CL/CLKernelLibrary.cpp
index d8983fcae9..c5a0796c3a 100644
--- a/src/core/CL/CLKernelLibrary.cpp
+++ b/src/core/CL/CLKernelLibrary.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/Error.h"
-#include "src/core/gpu/cl/ClKernelLibrary.h"
+#include "src/gpu/cl/ClKernelLibrary.h"
 #include <algorithm>
 #include <array>
 #include <fstream>
diff --git a/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp b/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
index 1437b5bebb..2b74f91a05 100644
--- a/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
+++ b/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
@@ -34,9 +34,9 @@
 #include "src/core/CL/CLUtils.h"
 #include "src/core/CL/CLValidate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/utils/AssemblyUtils.h b/src/core/utils/AssemblyUtils.h
index e682973827..b1aee64d5d 100644
--- a/src/core/utils/AssemblyUtils.h
+++ b/src/core/utils/AssemblyUtils.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "src/core/NEON/kernels/assembly/common.hpp"
-#include "src/core/cpu/kernels/assembly/arm_gemm.hpp"
+#include "src/cpu/kernels/assembly/arm_gemm.hpp"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/ICpuKernel.h b/src/cpu/ICpuKernel.h
index 650b3a7d0b..650b3a7d0b 100644
--- a/src/core/cpu/ICpuKernel.h
+++ b/src/cpu/ICpuKernel.h
diff --git a/src/runtime/cpu/ICpuOperator.h b/src/cpu/ICpuOperator.h
index 70ab4364c7..70ab4364c7 100644
--- a/src/runtime/cpu/ICpuOperator.h
+++ b/src/cpu/ICpuOperator.h
diff --git a/src/core/cpu/kernels/CpuActivationKernel.cpp b/src/cpu/kernels/CpuActivationKernel.cpp
index dad2ecfc5b..8fa7e9525e 100644
--- a/src/core/cpu/kernels/CpuActivationKernel.cpp
+++ b/src/cpu/kernels/CpuActivationKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuActivationKernel.h"
+#include "src/cpu/kernels/CpuActivationKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
@@ -31,7 +31,7 @@
 #include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/activation/list.h"
+#include "src/cpu/kernels/activation/list.h"
 
 #include <array>
 
diff --git a/src/core/cpu/kernels/CpuActivationKernel.h b/src/cpu/kernels/CpuActivationKernel.h
index 37650345fe..43c266529f 100644
--- a/src/core/cpu/kernels/CpuActivationKernel.h
+++ b/src/cpu/kernels/CpuActivationKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_ACTIVATION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuAddKernel.cpp b/src/cpu/kernels/CpuAddKernel.cpp
index 61b7b19443..07c9a65e55 100644
--- a/src/core/cpu/kernels/CpuAddKernel.cpp
+++ b/src/cpu/kernels/CpuAddKernel.cpp
@@ -21,17 +21,17 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuAddKernel.h"
+#include "src/cpu/kernels/CpuAddKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CPP/Validate.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/add/neon/list.h"
-#include "src/core/cpu/kernels/add/sve/list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/add/neon/list.h"
+#include "src/cpu/kernels/add/sve/list.h"
 
 #include <array>
 
diff --git a/src/core/cpu/kernels/CpuAddKernel.h b/src/cpu/kernels/CpuAddKernel.h
index 1205b45dfb..11c0f67132 100644
--- a/src/core/cpu/kernels/CpuAddKernel.h
+++ b/src/cpu/kernels/CpuAddKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_ADD_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuCastKernel.cpp b/src/cpu/kernels/CpuCastKernel.cpp
index 46f3c330ef..db76df9076 100644
--- a/src/core/cpu/kernels/CpuCastKernel.cpp
+++ b/src/cpu/kernels/CpuCastKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuCastKernel.h"
+#include "src/cpu/kernels/CpuCastKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuCastKernel.h b/src/cpu/kernels/CpuCastKernel.h
index 2a75c5850e..a8ce97230e 100644
--- a/src/core/cpu/kernels/CpuCastKernel.h
+++ b/src/cpu/kernels/CpuCastKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_CAST_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuCol2ImKernel.cpp b/src/cpu/kernels/CpuCol2ImKernel.cpp
index f860825de6..bf5a44d78b 100644
--- a/src/core/cpu/kernels/CpuCol2ImKernel.cpp
+++ b/src/cpu/kernels/CpuCol2ImKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuCol2ImKernel.h"
+#include "src/cpu/kernels/CpuCol2ImKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuCol2ImKernel.h b/src/cpu/kernels/CpuCol2ImKernel.h
index 3c1802230b..8e09a2b689 100644
--- a/src/core/cpu/kernels/CpuCol2ImKernel.h
+++ b/src/cpu/kernels/CpuCol2ImKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Size2D.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConcatenateBatchKernel.cpp b/src/cpu/kernels/CpuConcatenateBatchKernel.cpp
index 16c0efc793..29d40f0e52 100644
--- a/src/core/cpu/kernels/CpuConcatenateBatchKernel.cpp
+++ b/src/cpu/kernels/CpuConcatenateBatchKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConcatenateBatchKernel.h"
+#include "src/cpu/kernels/CpuConcatenateBatchKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuConcatenateBatchKernel.h b/src/cpu/kernels/CpuConcatenateBatchKernel.h
index 1706926fa8..91f2808f81 100644
--- a/src/core/cpu/kernels/CpuConcatenateBatchKernel.h
+++ b/src/cpu/kernels/CpuConcatenateBatchKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_CONCATENATE_BATCH_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConcatenateDepthKernel.cpp b/src/cpu/kernels/CpuConcatenateDepthKernel.cpp
index 133499deb6..ebc5322aee 100644
--- a/src/core/cpu/kernels/CpuConcatenateDepthKernel.cpp
+++ b/src/cpu/kernels/CpuConcatenateDepthKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConcatenateDepthKernel.h"
+#include "src/cpu/kernels/CpuConcatenateDepthKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuConcatenateDepthKernel.h b/src/cpu/kernels/CpuConcatenateDepthKernel.h
index 3ec19a86d1..063118b33b 100644
--- a/src/core/cpu/kernels/CpuConcatenateDepthKernel.h
+++ b/src/cpu/kernels/CpuConcatenateDepthKernel.h
@@ -26,7 +26,7 @@
 #define ARM_COMPUTE_CPU_CONCATENATE_DEPTH_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConcatenateHeightKernel.cpp b/src/cpu/kernels/CpuConcatenateHeightKernel.cpp
index dfd442b10a..47a2b44443 100644
--- a/src/core/cpu/kernels/CpuConcatenateHeightKernel.cpp
+++ b/src/cpu/kernels/CpuConcatenateHeightKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConcatenateHeightKernel.h"
+#include "src/cpu/kernels/CpuConcatenateHeightKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuConcatenateHeightKernel.h b/src/cpu/kernels/CpuConcatenateHeightKernel.h
index e5e15e1aee..883c59a206 100644
--- a/src/core/cpu/kernels/CpuConcatenateHeightKernel.h
+++ b/src/cpu/kernels/CpuConcatenateHeightKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_CONCATENATE_HEIGHT_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConcatenateWidthKernel.cpp b/src/cpu/kernels/CpuConcatenateWidthKernel.cpp
index ad33b0c951..90813ff7b4 100644
--- a/src/core/cpu/kernels/CpuConcatenateWidthKernel.cpp
+++ b/src/cpu/kernels/CpuConcatenateWidthKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConcatenateWidthKernel.h"
+#include "src/cpu/kernels/CpuConcatenateWidthKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuConcatenateWidthKernel.h b/src/cpu/kernels/CpuConcatenateWidthKernel.h
index f64191e173..3b4612ab0d 100644
--- a/src/core/cpu/kernels/CpuConcatenateWidthKernel.h
+++ b/src/cpu/kernels/CpuConcatenateWidthKernel.h
@@ -26,7 +26,7 @@
 #define ARM_COMPUTE_CPU_CONCATENATE_WIDTH_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp b/src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp
index 5406356bc9..08b39deef2 100644
--- a/src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp
+++ b/src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h"
+#include "src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
diff --git a/src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h b/src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h
index 7baaf13417..70f0a742f9 100644
--- a/src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h
+++ b/src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_CONVERT_FULLYCONNECTED_WEIGHTS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp b/src/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp
index 26cbb48deb..1005d001ab 100644
--- a/src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp
+++ b/src/cpu/kernels/CpuConvertQuantizedSignednessKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.h"
+#include "src/cpu/kernels/CpuConvertQuantizedSignednessKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.h b/src/cpu/kernels/CpuConvertQuantizedSignednessKernel.h
index 2a8f6c364d..8cce1eaf1d 100644
--- a/src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.h
+++ b/src/cpu/kernels/CpuConvertQuantizedSignednessKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_CONVERTQUANTIZEDSIGNEDNESS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuCopyKernel.cpp b/src/cpu/kernels/CpuCopyKernel.cpp
index 8ec354b2aa..3f0f3fe422 100644
--- a/src/core/cpu/kernels/CpuCopyKernel.cpp
+++ b/src/cpu/kernels/CpuCopyKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuCopyKernel.h"
+#include "src/cpu/kernels/CpuCopyKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuCopyKernel.h b/src/cpu/kernels/CpuCopyKernel.h
index e2f1ed60a6..193f38078b 100644
--- a/src/core/cpu/kernels/CpuCopyKernel.h
+++ b/src/cpu/kernels/CpuCopyKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_COPY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp b/src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp
index 5530eba9f1..d79fe87d1b 100644
--- a/src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp
+++ b/src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
+#include "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/ITensorInfo.h"
diff --git a/src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h b/src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h
index eb7041f7b6..1afb6bed4c 100644
--- a/src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h
+++ b/src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 #include "support/Requires.h"
 
 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
diff --git a/src/core/cpu/kernels/CpuDequantizeKernel.cpp b/src/cpu/kernels/CpuDequantizeKernel.cpp
index 42b5439697..a2d24f9243 100644
--- a/src/core/cpu/kernels/CpuDequantizeKernel.cpp
+++ b/src/cpu/kernels/CpuDequantizeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuDequantizeKernel.h"
+#include "src/cpu/kernels/CpuDequantizeKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuDequantizeKernel.h b/src/cpu/kernels/CpuDequantizeKernel.h
index e80aa3aaad..f515cd36f9 100644
--- a/src/core/cpu/kernels/CpuDequantizeKernel.h
+++ b/src/cpu/kernels/CpuDequantizeKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_DEQUANTIZE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuDirectConv2dKernel.cpp b/src/cpu/kernels/CpuDirectConv2dKernel.cpp
index faff55e905..db1b5f3c54 100644
--- a/src/core/cpu/kernels/CpuDirectConv2dKernel.cpp
+++ b/src/cpu/kernels/CpuDirectConv2dKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuDirectConv2dKernel.h"
+#include "src/cpu/kernels/CpuDirectConv2dKernel.h"
 
 #include "src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h"
 #include "src/core/NEON/wrapper/wrapper.h"
diff --git a/src/core/cpu/kernels/CpuDirectConv2dKernel.h b/src/cpu/kernels/CpuDirectConv2dKernel.h
index 9bef1c484a..3ba7f7ed5f 100644
--- a/src/core/cpu/kernels/CpuDirectConv2dKernel.h
+++ b/src/cpu/kernels/CpuDirectConv2dKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_DIRECT_CONV2D_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp b/src/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp
index 662d052941..93ad5e5eba 100644
--- a/src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp
+++ b/src/cpu/kernels/CpuDirectConv2dOutputStageKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.h"
+#include "src/cpu/kernels/CpuDirectConv2dOutputStageKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.h b/src/cpu/kernels/CpuDirectConv2dOutputStageKernel.h
index 749411c0a7..a68936bbae 100644
--- a/src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.h
+++ b/src/cpu/kernels/CpuDirectConv2dOutputStageKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuElementwiseKernel.cpp b/src/cpu/kernels/CpuElementwiseKernel.cpp
index dc574fce65..91de24b850 100644
--- a/src/core/cpu/kernels/CpuElementwiseKernel.cpp
+++ b/src/cpu/kernels/CpuElementwiseKernel.cpp
@@ -21,17 +21,17 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuElementwiseKernel.h"
+#include "src/cpu/kernels/CpuElementwiseKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "src/core/CPP/Validate.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/elementwise/neon/elementwise_list.h"
-#include "src/core/cpu/kernels/elementwise/neon/elementwise_quantized_list.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_list.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_quantized_list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/elementwise/neon/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/neon/elementwise_quantized_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_quantized_list.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/cpu/kernels/CpuElementwiseKernel.h b/src/cpu/kernels/CpuElementwiseKernel.h
index 75137da65d..f323fe4470 100644
--- a/src/core/cpu/kernels/CpuElementwiseKernel.h
+++ b/src/cpu/kernels/CpuElementwiseKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_ELEMENTWISE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuElementwiseUnaryKernel.cpp b/src/cpu/kernels/CpuElementwiseUnaryKernel.cpp
index b03c32f023..c587e18850 100644
--- a/src/core/cpu/kernels/CpuElementwiseUnaryKernel.cpp
+++ b/src/cpu/kernels/CpuElementwiseUnaryKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuElementwiseUnaryKernel.h"
+#include "src/cpu/kernels/CpuElementwiseUnaryKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -29,10 +29,10 @@
 #include "arm_compute/core/Validate.h"
 #include "src/core/CPP/Validate.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/elementwise/neon/elementwise_unary_list.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_unary_list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/elementwise/neon/elementwise_unary_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_unary_list.h"
 #include "support/ToolchainSupport.h"
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/CpuElementwiseUnaryKernel.h b/src/cpu/kernels/CpuElementwiseUnaryKernel.h
index bda65a35e0..f72eddf737 100644
--- a/src/core/cpu/kernels/CpuElementwiseUnaryKernel.h
+++ b/src/cpu/kernels/CpuElementwiseUnaryKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuFillKernel.cpp b/src/cpu/kernels/CpuFillKernel.cpp
index aab4d715ee..f69de0082d 100644
--- a/src/core/cpu/kernels/CpuFillKernel.cpp
+++ b/src/cpu/kernels/CpuFillKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuFillKernel.h"
+#include "src/cpu/kernels/CpuFillKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
diff --git a/src/core/cpu/kernels/CpuFillKernel.h b/src/cpu/kernels/CpuFillKernel.h
index 9afdee4186..3bc6a40309 100644
--- a/src/core/cpu/kernels/CpuFillKernel.h
+++ b/src/cpu/kernels/CpuFillKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/PixelValue.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuFloorKernel.cpp b/src/cpu/kernels/CpuFloorKernel.cpp
index d41df6a1f5..bcac1a41fc 100644
--- a/src/core/cpu/kernels/CpuFloorKernel.cpp
+++ b/src/cpu/kernels/CpuFloorKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuFloorKernel.h"
+#include "src/cpu/kernels/CpuFloorKernel.h"
 
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Helpers.h"
@@ -32,7 +32,7 @@
 #include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/floor/list.h"
+#include "src/cpu/kernels/floor/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuFloorKernel.h b/src/cpu/kernels/CpuFloorKernel.h
index 78534d2a1d..ffb9658190 100644
--- a/src/core/cpu/kernels/CpuFloorKernel.h
+++ b/src/cpu/kernels/CpuFloorKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_FLOOR_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp b/src/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp
index a6b080c0ab..9fbf2d54c6 100644
--- a/src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp
+++ b/src/cpu/kernels/CpuGemmInterleave4x4Kernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
+#include "src/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
diff --git a/src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h b/src/cpu/kernels/CpuGemmInterleave4x4Kernel.h
index 0c55886d8d..047776bd1e 100644
--- a/src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h
+++ b/src/cpu/kernels/CpuGemmInterleave4x4Kernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMM_INTERLEAVE4x4_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp b/src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp
index 35e542faa4..f8bef64066 100644
--- a/src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h b/src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h
index 77d8741b19..083ee187ef 100644
--- a/src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMMLOWP_MATRIXMULTIPLY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp b/src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp
index 270abc8bbd..534076b97c 100644
--- a/src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/KernelDescriptors.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h b/src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h
index 106980fc0b..918f8c89d9 100644
--- a/src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMMLOWP_REDUCTION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp b/src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp
index 9b1bf08955..a9896772f6 100644
--- a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h b/src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h
index f23a46cde7..1ec969be92 100644
--- a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMMLOWP_OFFSETCONTRIBUTION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp b/src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp
index 332ce6f013..89aa36486c 100644
--- a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h b/src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h
index 404f2c9496..d97727dd09 100644
--- a/src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
@@ -84,7 +84,7 @@ public:
      * @param[in]  output_stage   GEMMLowp output stage info, providing the type of quantization and the necessary parameters.
      */
     void configure(const ITensorInfo *mm_result, const ITensorInfo *vector_sum_col, const ITensorInfo *vector_sum_row, const ITensorInfo *bias, ITensorInfo *dst, int32_t k, int32_t a_offset,
-                   int32_t b_offset,
+                   int32_t                 b_offset,
                    GEMMLowpOutputStageInfo output_stage);
     /** Static function to check if given info will lead to a valid configuration
      *
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp
index f1c797244a..3023d93113 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h
index ca5e1b40fc..ae13e760ff 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
index 390e269cbb..53ca991889 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
index e360e65bae..53a9d34ed1 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
index 318b6a06f8..27214dcb5a 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
index 9c213abdf7..67829e7773 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
index 6631a4fc67..e49fd29115 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
index 13b30f3427..b62cac4818 100644
--- a/src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
+++ b/src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp b/src/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp
index da0f7b135e..81376fb029 100644
--- a/src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp
+++ b/src/cpu/kernels/CpuGemmMatrixAdditionKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.h"
+#include "src/cpu/kernels/CpuGemmMatrixAdditionKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
diff --git a/src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.h b/src/cpu/kernels/CpuGemmMatrixAdditionKernel.h
index f9450b962b..c9798fc24c 100644
--- a/src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.h
+++ b/src/cpu/kernels/CpuGemmMatrixAdditionKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMM_MATRIX_ADDITION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp b/src/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp
index d86ea064de..93ae90436a 100644
--- a/src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp
+++ b/src/cpu/kernels/CpuGemmMatrixMultiplyKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
+#include "src/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
diff --git a/src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.h b/src/cpu/kernels/CpuGemmMatrixMultiplyKernel.h
index 974ff85606..0b4e01579c 100644
--- a/src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.h
+++ b/src/cpu/kernels/CpuGemmMatrixMultiplyKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMM_MATRIX_MULTIPLY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuGemmTranspose1xWKernel.cpp b/src/cpu/kernels/CpuGemmTranspose1xWKernel.cpp
index 4b059f57cb..62d5d5f5e9 100644
--- a/src/core/cpu/kernels/CpuGemmTranspose1xWKernel.cpp
+++ b/src/cpu/kernels/CpuGemmTranspose1xWKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h"
+#include "src/cpu/kernels/CpuGemmTranspose1xWKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
diff --git a/src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h b/src/cpu/kernels/CpuGemmTranspose1xWKernel.h
index 1a9287f7b0..de920b5ed7 100644
--- a/src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h
+++ b/src/cpu/kernels/CpuGemmTranspose1xWKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMM_TRANSPOSE1xW_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuIm2ColKernel.cpp b/src/cpu/kernels/CpuIm2ColKernel.cpp
index ca6c9bfab4..13764c49d1 100644
--- a/src/core/cpu/kernels/CpuIm2ColKernel.cpp
+++ b/src/cpu/kernels/CpuIm2ColKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuIm2ColKernel.h"
+#include "src/cpu/kernels/CpuIm2ColKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuIm2ColKernel.h b/src/cpu/kernels/CpuIm2ColKernel.h
index ffac5077b2..fc8ae056bb 100644
--- a/src/core/cpu/kernels/CpuIm2ColKernel.h
+++ b/src/cpu/kernels/CpuIm2ColKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Size2D.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuMulKernel.cpp b/src/cpu/kernels/CpuMulKernel.cpp
index 82ec322875..da7b6d7d66 100644
--- a/src/core/cpu/kernels/CpuMulKernel.cpp
+++ b/src/cpu/kernels/CpuMulKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuMulKernel.h"
+#include "src/cpu/kernels/CpuMulKernel.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
diff --git a/src/core/cpu/kernels/CpuMulKernel.h b/src/cpu/kernels/CpuMulKernel.h
index 3ea176cc31..b65ec20044 100644
--- a/src/core/cpu/kernels/CpuMulKernel.h
+++ b/src/cpu/kernels/CpuMulKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_MUL_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuPermuteKernel.cpp b/src/cpu/kernels/CpuPermuteKernel.cpp
index 270d6e222e..d65e011032 100644
--- a/src/core/cpu/kernels/CpuPermuteKernel.cpp
+++ b/src/cpu/kernels/CpuPermuteKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuPermuteKernel.h"
+#include "src/cpu/kernels/CpuPermuteKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuPermuteKernel.h b/src/cpu/kernels/CpuPermuteKernel.h
index 2955f38960..1b2672b5b9 100644
--- a/src/core/cpu/kernels/CpuPermuteKernel.h
+++ b/src/cpu/kernels/CpuPermuteKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_PERMUTE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuPool2dKernel.cpp b/src/cpu/kernels/CpuPool2dKernel.cpp
index 27f4b950db..d7fb75ee60 100644
--- a/src/core/cpu/kernels/CpuPool2dKernel.cpp
+++ b/src/cpu/kernels/CpuPool2dKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuPool2dKernel.h"
+#include "src/cpu/kernels/CpuPool2dKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
@@ -34,9 +34,9 @@
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 #include "support/ToolchainSupport.h"
 
 #include "src/core/NEON/wrapper/wrapper.h"
diff --git a/src/core/cpu/kernels/CpuPool2dKernel.h b/src/cpu/kernels/CpuPool2dKernel.h
index 9ed398b907..70fe52d29c 100644
--- a/src/core/cpu/kernels/CpuPool2dKernel.h
+++ b/src/cpu/kernels/CpuPool2dKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuQuantizeKernel.cpp b/src/cpu/kernels/CpuQuantizeKernel.cpp
index 8ca81e8b11..ecae5e7b4e 100644
--- a/src/core/cpu/kernels/CpuQuantizeKernel.cpp
+++ b/src/cpu/kernels/CpuQuantizeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuQuantizeKernel.h"
+#include "src/cpu/kernels/CpuQuantizeKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuQuantizeKernel.h b/src/cpu/kernels/CpuQuantizeKernel.h
index 834a2e03d2..eb0814926d 100644
--- a/src/core/cpu/kernels/CpuQuantizeKernel.h
+++ b/src/cpu/kernels/CpuQuantizeKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_QUANTIZE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuReshapeKernel.cpp b/src/cpu/kernels/CpuReshapeKernel.cpp
index 5b717b9bba..3bbcc09cc5 100644
--- a/src/core/cpu/kernels/CpuReshapeKernel.cpp
+++ b/src/cpu/kernels/CpuReshapeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuReshapeKernel.h"
+#include "src/cpu/kernels/CpuReshapeKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuReshapeKernel.h b/src/cpu/kernels/CpuReshapeKernel.h
index 1425fbe917..9fe4350445 100644
--- a/src/core/cpu/kernels/CpuReshapeKernel.h
+++ b/src/cpu/kernels/CpuReshapeKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_RESHAPE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuScaleKernel.cpp b/src/cpu/kernels/CpuScaleKernel.cpp
index 0c1f08ab79..1108c7a78e 100644
--- a/src/core/cpu/kernels/CpuScaleKernel.cpp
+++ b/src/cpu/kernels/CpuScaleKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuScaleKernel.h"
+#include "src/cpu/kernels/CpuScaleKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Window.h"
@@ -29,12 +29,12 @@
 #include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/scale/neon/list.h"
-#include "src/core/cpu/kernels/scale/sve/list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/helpers/WindowHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
+#include "src/cpu/kernels/scale/neon/list.h"
+#include "src/cpu/kernels/scale/sve/list.h"
 #include "support/Rounding.h"
 
 #include <arm_neon.h>
diff --git a/src/core/cpu/kernels/CpuScaleKernel.h b/src/cpu/kernels/CpuScaleKernel.h
index a2b65370ba..913b5a5593 100644
--- a/src/core/cpu/kernels/CpuScaleKernel.h
+++ b/src/cpu/kernels/CpuScaleKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuSoftmaxKernel.cpp b/src/cpu/kernels/CpuSoftmaxKernel.cpp
index c562699092..cbf3773ddc 100644
--- a/src/core/cpu/kernels/CpuSoftmaxKernel.cpp
+++ b/src/cpu/kernels/CpuSoftmaxKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuSoftmaxKernel.h"
+#include "src/cpu/kernels/CpuSoftmaxKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -34,8 +34,8 @@
 #include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/softmax/impl/neon/list.h"
-#include "src/core/cpu/kernels/softmax/impl/sve/list.h"
+#include "src/cpu/kernels/softmax/impl/neon/list.h"
+#include "src/cpu/kernels/softmax/impl/sve/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuSoftmaxKernel.h b/src/cpu/kernels/CpuSoftmaxKernel.h
index 776c0d6f79..8073a677d9 100644
--- a/src/core/cpu/kernels/CpuSoftmaxKernel.h
+++ b/src/cpu/kernels/CpuSoftmaxKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_SOFTMAX_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuSubKernel.cpp b/src/cpu/kernels/CpuSubKernel.cpp
index fa7a55805e..ec65f12dfc 100644
--- a/src/core/cpu/kernels/CpuSubKernel.cpp
+++ b/src/cpu/kernels/CpuSubKernel.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuSubKernel.h"
+#include "src/cpu/kernels/CpuSubKernel.h"
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CPP/Validate.h"
 #include "src/core/common/Registrars.h"
-#include "src/core/cpu/kernels/sub/neon/list.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/sub/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuSubKernel.h b/src/cpu/kernels/CpuSubKernel.h
index cb64e64cfa..80d6be68b5 100644
--- a/src/core/cpu/kernels/CpuSubKernel.h
+++ b/src/cpu/kernels/CpuSubKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_SUB_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuTransposeKernel.cpp b/src/cpu/kernels/CpuTransposeKernel.cpp
index c7cafe94a8..2f981c15e4 100644
--- a/src/core/cpu/kernels/CpuTransposeKernel.cpp
+++ b/src/cpu/kernels/CpuTransposeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuTransposeKernel.h"
+#include "src/cpu/kernels/CpuTransposeKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuTransposeKernel.h b/src/cpu/kernels/CpuTransposeKernel.h
index 920349d5e7..6805eac642 100644
--- a/src/core/cpu/kernels/CpuTransposeKernel.h
+++ b/src/cpu/kernels/CpuTransposeKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_TRANSPOSE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuWeightsReshapeKernel.cpp b/src/cpu/kernels/CpuWeightsReshapeKernel.cpp
index 79f058944d..2ccc977995 100644
--- a/src/core/cpu/kernels/CpuWeightsReshapeKernel.cpp
+++ b/src/cpu/kernels/CpuWeightsReshapeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuWeightsReshapeKernel.h"
+#include "src/cpu/kernels/CpuWeightsReshapeKernel.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
diff --git a/src/core/cpu/kernels/CpuWeightsReshapeKernel.h b/src/cpu/kernels/CpuWeightsReshapeKernel.h
index eea150a96e..c80bf3b25e 100644
--- a/src/core/cpu/kernels/CpuWeightsReshapeKernel.h
+++ b/src/cpu/kernels/CpuWeightsReshapeKernel.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_WEIGHTSRESHAPE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/CpuWinogradConv2dKernel.cpp b/src/cpu/kernels/CpuWinogradConv2dKernel.cpp
index 9456f96354..803af09a67 100644
--- a/src/core/cpu/kernels/CpuWinogradConv2dKernel.cpp
+++ b/src/cpu/kernels/CpuWinogradConv2dKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuWinogradConv2dKernel.h"
+#include "src/cpu/kernels/CpuWinogradConv2dKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/cpu/kernels/CpuWinogradConv2dKernel.h b/src/cpu/kernels/CpuWinogradConv2dKernel.h
index b5a29ffd02..db2d8acfdb 100644
--- a/src/core/cpu/kernels/CpuWinogradConv2dKernel.h
+++ b/src/cpu/kernels/CpuWinogradConv2dKernel.h
@@ -26,7 +26,7 @@
 
 #include "src/core/NEON/kernels/convolution/common/convolution.hpp"
 #include "src/core/NEON/kernels/convolution/common/tensor.hpp"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 #include "src/core/NEON/kernels/convolution/winograd/winograd_layer.hpp"
 
diff --git a/src/core/cpu/kernels/activation/list.h b/src/cpu/kernels/activation/list.h
index 409d025db0..409d025db0 100644
--- a/src/core/cpu/kernels/activation/list.h
+++ b/src/cpu/kernels/activation/list.h
diff --git a/src/core/cpu/kernels/activation/neon/fp16.cpp b/src/cpu/kernels/activation/neon/fp16.cpp
index 6f2d5d8533..6f2d5d8533 100644
--- a/src/core/cpu/kernels/activation/neon/fp16.cpp
+++ b/src/cpu/kernels/activation/neon/fp16.cpp
diff --git a/src/core/cpu/kernels/activation/neon/fp32.cpp b/src/cpu/kernels/activation/neon/fp32.cpp
index 54301d45ad..54301d45ad 100644
--- a/src/core/cpu/kernels/activation/neon/fp32.cpp
+++ b/src/cpu/kernels/activation/neon/fp32.cpp
diff --git a/src/core/cpu/kernels/activation/neon/qasymm8.cpp b/src/cpu/kernels/activation/neon/qasymm8.cpp
index a1217435b6..a1217435b6 100644
--- a/src/core/cpu/kernels/activation/neon/qasymm8.cpp
+++ b/src/cpu/kernels/activation/neon/qasymm8.cpp
diff --git a/src/core/cpu/kernels/activation/neon/qasymm8_signed.cpp b/src/cpu/kernels/activation/neon/qasymm8_signed.cpp
index 8b40bf8e72..8b40bf8e72 100644
--- a/src/core/cpu/kernels/activation/neon/qasymm8_signed.cpp
+++ b/src/cpu/kernels/activation/neon/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/activation/neon/qsymm16.cpp b/src/cpu/kernels/activation/neon/qsymm16.cpp
index 54b41820f2..54b41820f2 100644
--- a/src/core/cpu/kernels/activation/neon/qsymm16.cpp
+++ b/src/cpu/kernels/activation/neon/qsymm16.cpp
diff --git a/src/core/cpu/kernels/activation/sve/fp16.cpp b/src/cpu/kernels/activation/sve/fp16.cpp
index 5e76e82c52..5e76e82c52 100644
--- a/src/core/cpu/kernels/activation/sve/fp16.cpp
+++ b/src/cpu/kernels/activation/sve/fp16.cpp
diff --git a/src/core/cpu/kernels/activation/sve/fp32.cpp b/src/cpu/kernels/activation/sve/fp32.cpp
index cb9f82eb39..cb9f82eb39 100644
--- a/src/core/cpu/kernels/activation/sve/fp32.cpp
+++ b/src/cpu/kernels/activation/sve/fp32.cpp
diff --git a/src/core/cpu/kernels/activation/sve/qasymm8.cpp b/src/cpu/kernels/activation/sve/qasymm8.cpp
index 69fffd96c5..69fffd96c5 100644
--- a/src/core/cpu/kernels/activation/sve/qasymm8.cpp
+++ b/src/cpu/kernels/activation/sve/qasymm8.cpp
diff --git a/src/core/cpu/kernels/activation/sve/qasymm8_signed.cpp b/src/cpu/kernels/activation/sve/qasymm8_signed.cpp
index 53ee515ff9..53ee515ff9 100644
--- a/src/core/cpu/kernels/activation/sve/qasymm8_signed.cpp
+++ b/src/cpu/kernels/activation/sve/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/activation/sve/qsymm16.cpp b/src/cpu/kernels/activation/sve/qsymm16.cpp
index ac549770a2..ac549770a2 100644
--- a/src/core/cpu/kernels/activation/sve/qsymm16.cpp
+++ b/src/cpu/kernels/activation/sve/qsymm16.cpp
diff --git a/src/core/cpu/kernels/add/neon/list.h b/src/cpu/kernels/add/neon/list.h
index 379bd32fb1..379bd32fb1 100644
--- a/src/core/cpu/kernels/add/neon/list.h
+++ b/src/cpu/kernels/add/neon/list.h
diff --git a/src/core/cpu/kernels/add/neon/qasymm8.cpp b/src/cpu/kernels/add/neon/qasymm8.cpp
index e357a7ef7f..e357a7ef7f 100644
--- a/src/core/cpu/kernels/add/neon/qasymm8.cpp
+++ b/src/cpu/kernels/add/neon/qasymm8.cpp
diff --git a/src/core/cpu/kernels/add/neon/qasymm8_signed.cpp b/src/cpu/kernels/add/neon/qasymm8_signed.cpp
index d62d0739f5..d62d0739f5 100644
--- a/src/core/cpu/kernels/add/neon/qasymm8_signed.cpp
+++ b/src/cpu/kernels/add/neon/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/add/neon/qsymm16.cpp b/src/cpu/kernels/add/neon/qsymm16.cpp
index e76e408d6e..e76e408d6e 100644
--- a/src/core/cpu/kernels/add/neon/qsymm16.cpp
+++ b/src/cpu/kernels/add/neon/qsymm16.cpp
diff --git a/src/core/cpu/kernels/add/sve/impl.cpp b/src/cpu/kernels/add/sve/impl.cpp
index cf9e301c29..f8e16a508c 100644
--- a/src/core/cpu/kernels/add/sve/impl.cpp
+++ b/src/cpu/kernels/add/sve/impl.cpp
@@ -29,7 +29,7 @@
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
 
 #include "src/core/NEON/SVEMath.h"
-#include "src/core/cpu/kernels/add/sve/impl.h"
+#include "src/cpu/kernels/add/sve/impl.h"
 #include <arm_sve.h>
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/add/sve/impl.h b/src/cpu/kernels/add/sve/impl.h
index 32ff5d0496..32ff5d0496 100644
--- a/src/core/cpu/kernels/add/sve/impl.h
+++ b/src/cpu/kernels/add/sve/impl.h
diff --git a/src/core/cpu/kernels/add/sve/list.h b/src/cpu/kernels/add/sve/list.h
index 4d29c2a8f1..4529a9f7c1 100644
--- a/src/core/cpu/kernels/add/sve/list.h
+++ b/src/cpu/kernels/add/sve/list.h
@@ -29,7 +29,7 @@
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/SVEMath.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/add/sve/impl.h"
+#include "src/cpu/kernels/add/sve/impl.h"
 #include <arm_sve.h>
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/add/sve/qasymm8.cpp b/src/cpu/kernels/add/sve/qasymm8.cpp
index 888ad878ca..888ad878ca 100644
--- a/src/core/cpu/kernels/add/sve/qasymm8.cpp
+++ b/src/cpu/kernels/add/sve/qasymm8.cpp
diff --git a/src/core/cpu/kernels/add/sve/qasymm8_signed.cpp b/src/cpu/kernels/add/sve/qasymm8_signed.cpp
index 3b922c6c21..3b922c6c21 100644
--- a/src/core/cpu/kernels/add/sve/qasymm8_signed.cpp
+++ b/src/cpu/kernels/add/sve/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/add/sve/qsymm16.cpp b/src/cpu/kernels/add/sve/qsymm16.cpp
index eef5d245d3..eef5d245d3 100644
--- a/src/core/cpu/kernels/add/sve/qsymm16.cpp
+++ b/src/cpu/kernels/add/sve/qsymm16.cpp
diff --git a/src/core/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h b/src/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h
index 4b7b092d01..3b9a6b4760 100644
--- a/src/core/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h
+++ b/src/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/NEON/INEKernel.h"
-#include "src/core/cpu/kernels/assembly/arm_gemm_compute_iface.hpp"
+#include "src/cpu/kernels/assembly/arm_gemm_compute_iface.hpp"
 
 #include "gemm_common.hpp"
 
diff --git a/src/core/cpu/kernels/assembly/arm_gemm.hpp b/src/cpu/kernels/assembly/arm_gemm.hpp
index e38cc09202..e38cc09202 100644
--- a/src/core/cpu/kernels/assembly/arm_gemm.hpp
+++ b/src/cpu/kernels/assembly/arm_gemm.hpp
diff --git a/src/core/cpu/kernels/assembly/arm_gemm_compute_iface.hpp b/src/cpu/kernels/assembly/arm_gemm_compute_iface.hpp
index 718fcd1fb4..718fcd1fb4 100644
--- a/src/core/cpu/kernels/assembly/arm_gemm_compute_iface.hpp
+++ b/src/cpu/kernels/assembly/arm_gemm_compute_iface.hpp
diff --git a/src/core/cpu/kernels/assembly/arm_gemm_local.hpp b/src/cpu/kernels/assembly/arm_gemm_local.hpp
index 78e0adf31f..78e0adf31f 100644
--- a/src/core/cpu/kernels/assembly/arm_gemm_local.hpp
+++ b/src/cpu/kernels/assembly/arm_gemm_local.hpp
diff --git a/src/core/cpu/kernels/assembly/convolution_parameters.hpp b/src/cpu/kernels/assembly/convolution_parameters.hpp
index 0c1ae58902..0c1ae58902 100644
--- a/src/core/cpu/kernels/assembly/convolution_parameters.hpp
+++ b/src/cpu/kernels/assembly/convolution_parameters.hpp
diff --git a/src/core/cpu/kernels/assembly/gemm_common.hpp b/src/cpu/kernels/assembly/gemm_common.hpp
index 378f1041be..378f1041be 100644
--- a/src/core/cpu/kernels/assembly/gemm_common.hpp
+++ b/src/cpu/kernels/assembly/gemm_common.hpp
diff --git a/src/core/cpu/kernels/assembly/ndrange.hpp b/src/cpu/kernels/assembly/ndrange.hpp
index 1c8261aef7..1c8261aef7 100644
--- a/src/core/cpu/kernels/assembly/ndrange.hpp
+++ b/src/cpu/kernels/assembly/ndrange.hpp
diff --git a/src/core/cpu/kernels/elementwise/neon/elementwise_list.h b/src/cpu/kernels/elementwise/neon/elementwise_list.h
index 43e44be5e2..43e44be5e2 100644
--- a/src/core/cpu/kernels/elementwise/neon/elementwise_list.h
+++ b/src/cpu/kernels/elementwise/neon/elementwise_list.h
diff --git a/src/core/cpu/kernels/elementwise/neon/elementwise_quantized_list.h b/src/cpu/kernels/elementwise/neon/elementwise_quantized_list.h
index 1ff4632f5c..3b4c112770 100644
--- a/src/core/cpu/kernels/elementwise/neon/elementwise_quantized_list.h
+++ b/src/cpu/kernels/elementwise/neon/elementwise_quantized_list.h
@@ -24,7 +24,7 @@
 #ifndef SRC_CORE_NEON_KERNELS_ELEMENTWISE_QUANTIZED_LIST_H
 #define SRC_CORE_NEON_KERNELS_ELEMENTWISE_QUANTIZED_LIST_H
 
-#include "src/core/cpu/kernels/elementwise/neon/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/neon/elementwise_list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/elementwise/neon/elementwise_unary_list.h b/src/cpu/kernels/elementwise/neon/elementwise_unary_list.h
index 307e95fae9..307e95fae9 100644
--- a/src/core/cpu/kernels/elementwise/neon/elementwise_unary_list.h
+++ b/src/cpu/kernels/elementwise/neon/elementwise_unary_list.h
diff --git a/src/core/cpu/kernels/elementwise/sve/elementwise.cpp b/src/cpu/kernels/elementwise/sve/elementwise.cpp
index 58ebb28fe5..2f9a7998df 100644
--- a/src/core/cpu/kernels/elementwise/sve/elementwise.cpp
+++ b/src/cpu/kernels/elementwise/sve/elementwise.cpp
@@ -24,7 +24,7 @@
 #if defined(__ARM_FEATURE_SVE)
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_list.h"
 #include <arm_sve.h>
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/elementwise/sve/elementwise_list.h b/src/cpu/kernels/elementwise/sve/elementwise_list.h
index fea38d2995..f762587ce7 100644
--- a/src/core/cpu/kernels/elementwise/sve/elementwise_list.h
+++ b/src/cpu/kernels/elementwise/sve/elementwise_list.h
@@ -32,7 +32,7 @@
 #include "src/core/NEON/SVEMath.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
 #include "src/core/NEON/wrapper/svtraits.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_list.h"
 #include <arm_sve.h>
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/elementwise/sve/elementwise_quantized_list.h b/src/cpu/kernels/elementwise/sve/elementwise_quantized_list.h
index 5e04128b44..a5d17a86a7 100644
--- a/src/core/cpu/kernels/elementwise/sve/elementwise_quantized_list.h
+++ b/src/cpu/kernels/elementwise/sve/elementwise_quantized_list.h
@@ -27,7 +27,7 @@
 #if defined(ARM_COMPUTE_ENABLE_SVE2)
 
 #include "src/core/NEON/wrapper/svtraits.h"
-#include "src/core/cpu/kernels/elementwise/sve/elementwise_list.h"
+#include "src/cpu/kernels/elementwise/sve/elementwise_list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/elementwise/sve/elementwise_unary.cpp b/src/cpu/kernels/elementwise/sve/elementwise_unary.cpp
index ddf1febd66..ddf1febd66 100644
--- a/src/core/cpu/kernels/elementwise/sve/elementwise_unary.cpp
+++ b/src/cpu/kernels/elementwise/sve/elementwise_unary.cpp
diff --git a/src/core/cpu/kernels/elementwise/sve/elementwise_unary_list.h b/src/cpu/kernels/elementwise/sve/elementwise_unary_list.h
index c2b495f27c..c2b495f27c 100644
--- a/src/core/cpu/kernels/elementwise/sve/elementwise_unary_list.h
+++ b/src/cpu/kernels/elementwise/sve/elementwise_unary_list.h
diff --git a/src/core/cpu/kernels/floor/list.h b/src/cpu/kernels/floor/list.h
index 4367e0ffc9..4367e0ffc9 100644
--- a/src/core/cpu/kernels/floor/list.h
+++ b/src/cpu/kernels/floor/list.h
diff --git a/src/core/cpu/kernels/floor/neon/fp16.cpp b/src/cpu/kernels/floor/neon/fp16.cpp
index f362676a36..f362676a36 100644
--- a/src/core/cpu/kernels/floor/neon/fp16.cpp
+++ b/src/cpu/kernels/floor/neon/fp16.cpp
diff --git a/src/core/cpu/kernels/floor/neon/fp32.cpp b/src/cpu/kernels/floor/neon/fp32.cpp
index f5efb2e849..f5efb2e849 100644
--- a/src/core/cpu/kernels/floor/neon/fp32.cpp
+++ b/src/cpu/kernels/floor/neon/fp32.cpp
diff --git a/src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp b/src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp
index f5c63b763f..eed4bb9dd5 100644
--- a/src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp
+++ b/src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h"
+#include "src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h"
 
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
diff --git a/src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h b/src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h
index 8ff44441e9..8ee24a6613 100644
--- a/src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h
+++ b/src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 namespace arm_conv
 {
diff --git a/src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp b/src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp
index 89dd27a20a..958c04b677 100644
--- a/src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp
+++ b/src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h"
+#include "src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
diff --git a/src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h b/src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h
index 3afa4c16a4..ab3ed25b1e 100644
--- a/src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h
+++ b/src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Types.h"
 #include "src/core/NEON/kernels/assembly/pooling.hpp"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuKernel.h"
 
 #include "pool_common.hpp"
 
diff --git a/src/core/cpu/kernels/pool2d/neon/fp16.cpp b/src/cpu/kernels/pool2d/neon/fp16.cpp
index d21e153f25..534d24ab49 100644
--- a/src/core/cpu/kernels/pool2d/neon/fp16.cpp
+++ b/src/cpu/kernels/pool2d/neon/fp16.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 
 #if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
 
diff --git a/src/core/cpu/kernels/pool2d/neon/fp32.cpp b/src/cpu/kernels/pool2d/neon/fp32.cpp
index c82cad0ffd..26a32ed9d4 100644
--- a/src/core/cpu/kernels/pool2d/neon/fp32.cpp
+++ b/src/cpu/kernels/pool2d/neon/fp32.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/pool2d/neon/list.h b/src/cpu/kernels/pool2d/neon/list.h
index f1e23d43cf..b79323213e 100644
--- a/src/core/cpu/kernels/pool2d/neon/list.h
+++ b/src/cpu/kernels/pool2d/neon/list.h
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/cpu/kernels/pool2d/neon/quantized.h"
+#include "src/cpu/kernels/pool2d/neon/quantized.h"
 #include <arm_neon.h>
 
 namespace arm_compute
diff --git a/src/core/cpu/kernels/pool2d/neon/nchw/all.cpp b/src/cpu/kernels/pool2d/neon/nchw/all.cpp
index bece438989..3ca7701087 100644
--- a/src/core/cpu/kernels/pool2d/neon/nchw/all.cpp
+++ b/src/cpu/kernels/pool2d/neon/nchw/all.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 
 #ifdef ENABLE_NCHW_KERNELS
 namespace arm_compute
diff --git a/src/core/cpu/kernels/pool2d/neon/qasymm8.cpp b/src/cpu/kernels/pool2d/neon/qasymm8.cpp
index 4020e9e3fc..7f8841edd8 100644
--- a/src/core/cpu/kernels/pool2d/neon/qasymm8.cpp
+++ b/src/cpu/kernels/pool2d/neon/qasymm8.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/pool2d/neon/qasymm8_signed.cpp b/src/cpu/kernels/pool2d/neon/qasymm8_signed.cpp
index a899427484..8643651f27 100644
--- a/src/core/cpu/kernels/pool2d/neon/qasymm8_signed.cpp
+++ b/src/cpu/kernels/pool2d/neon/qasymm8_signed.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
-#include "src/core/cpu/kernels/pool2d/neon/list.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/pool2d/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/pool2d/neon/quantized.h b/src/cpu/kernels/pool2d/neon/quantized.h
index a16960a205..a16960a205 100644
--- a/src/core/cpu/kernels/pool2d/neon/quantized.h
+++ b/src/cpu/kernels/pool2d/neon/quantized.h
diff --git a/src/core/cpu/kernels/scale/neon/fp16.cpp b/src/cpu/kernels/scale/neon/fp16.cpp
index 0ad66cab1c..0ad66cab1c 100644
--- a/src/core/cpu/kernels/scale/neon/fp16.cpp
+++ b/src/cpu/kernels/scale/neon/fp16.cpp
diff --git a/src/core/cpu/kernels/scale/neon/integer.cpp b/src/cpu/kernels/scale/neon/integer.cpp
index a2359aac94..a2359aac94 100644
--- a/src/core/cpu/kernels/scale/neon/integer.cpp
+++ b/src/cpu/kernels/scale/neon/integer.cpp
diff --git a/src/core/cpu/kernels/scale/neon/list.h b/src/cpu/kernels/scale/neon/list.h
index c91242f5b2..c91242f5b2 100644
--- a/src/core/cpu/kernels/scale/neon/list.h
+++ b/src/cpu/kernels/scale/neon/list.h
diff --git a/src/core/cpu/kernels/scale/neon/qasymm8.cpp b/src/cpu/kernels/scale/neon/qasymm8.cpp
index 90302ce889..fb52752690 100644
--- a/src/core/cpu/kernels/scale/neon/qasymm8.cpp
+++ b/src/cpu/kernels/scale/neon/qasymm8.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/scale/neon/list.h"
+#include "src/cpu/kernels/scale/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/scale/neon/qasymm8_signed.cpp b/src/cpu/kernels/scale/neon/qasymm8_signed.cpp
index 07d6c6ef03..706bcee954 100644
--- a/src/core/cpu/kernels/scale/neon/qasymm8_signed.cpp
+++ b/src/cpu/kernels/scale/neon/qasymm8_signed.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/scale/neon/list.h"
+#include "src/cpu/kernels/scale/neon/list.h"
 
 namespace arm_compute
 {
diff --git a/src/core/cpu/kernels/scale/sve/fp16.cpp b/src/cpu/kernels/scale/sve/fp16.cpp
index 76e7735b8a..76e7735b8a 100644
--- a/src/core/cpu/kernels/scale/sve/fp16.cpp
+++ b/src/cpu/kernels/scale/sve/fp16.cpp
diff --git a/src/core/cpu/kernels/scale/sve/fp32.cpp b/src/cpu/kernels/scale/sve/fp32.cpp
index 030e109cdf..030e109cdf 100644
--- a/src/core/cpu/kernels/scale/sve/fp32.cpp
+++ b/src/cpu/kernels/scale/sve/fp32.cpp
diff --git a/src/core/cpu/kernels/scale/sve/integer.cpp b/src/cpu/kernels/scale/sve/integer.cpp
index 486c674612..486c674612 100644
--- a/src/core/cpu/kernels/scale/sve/integer.cpp
+++ b/src/cpu/kernels/scale/sve/integer.cpp
diff --git a/src/core/cpu/kernels/scale/sve/list.h b/src/cpu/kernels/scale/sve/list.h
index b9c3a10a78..b9c3a10a78 100644
--- a/src/core/cpu/kernels/scale/sve/list.h
+++ b/src/cpu/kernels/scale/sve/list.h
diff --git a/src/core/cpu/kernels/scale/sve/qasymm8.cpp b/src/cpu/kernels/scale/sve/qasymm8.cpp
index c9122ad40b..c9122ad40b 100644
--- a/src/core/cpu/kernels/scale/sve/qasymm8.cpp
+++ b/src/cpu/kernels/scale/sve/qasymm8.cpp
diff --git a/src/core/cpu/kernels/scale/sve/qasymm8_signed.cpp b/src/cpu/kernels/scale/sve/qasymm8_signed.cpp
index 0843e61fd4..0843e61fd4 100644
--- a/src/core/cpu/kernels/scale/sve/qasymm8_signed.cpp
+++ b/src/cpu/kernels/scale/sve/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/softmax/impl/neon/list.h b/src/cpu/kernels/softmax/impl/neon/list.h
index 5ebee31272..5ebee31272 100644
--- a/src/core/cpu/kernels/softmax/impl/neon/list.h
+++ b/src/cpu/kernels/softmax/impl/neon/list.h
diff --git a/src/core/cpu/kernels/softmax/impl/sve/impl.cpp b/src/cpu/kernels/softmax/impl/sve/impl.cpp
index 7a577fd565..7a577fd565 100644
--- a/src/core/cpu/kernels/softmax/impl/sve/impl.cpp
+++ b/src/cpu/kernels/softmax/impl/sve/impl.cpp
diff --git a/src/core/cpu/kernels/softmax/impl/sve/list.h b/src/cpu/kernels/softmax/impl/sve/list.h
index b4e1e1b186..b4e1e1b186 100644
--- a/src/core/cpu/kernels/softmax/impl/sve/list.h
+++ b/src/cpu/kernels/softmax/impl/sve/list.h
diff --git a/src/core/cpu/kernels/sub/neon/list.h b/src/cpu/kernels/sub/neon/list.h
index ac1346001a..ac1346001a 100644
--- a/src/core/cpu/kernels/sub/neon/list.h
+++ b/src/cpu/kernels/sub/neon/list.h
diff --git a/src/core/cpu/kernels/sub/neon/qasymm8.cpp b/src/cpu/kernels/sub/neon/qasymm8.cpp
index 8f4cd8bdbb..8f4cd8bdbb 100644
--- a/src/core/cpu/kernels/sub/neon/qasymm8.cpp
+++ b/src/cpu/kernels/sub/neon/qasymm8.cpp
diff --git a/src/core/cpu/kernels/sub/neon/qasymm8_signed.cpp b/src/cpu/kernels/sub/neon/qasymm8_signed.cpp
index 2c9e411743..2c9e411743 100644
--- a/src/core/cpu/kernels/sub/neon/qasymm8_signed.cpp
+++ b/src/cpu/kernels/sub/neon/qasymm8_signed.cpp
diff --git a/src/core/cpu/kernels/sub/neon/qsymm16.cpp b/src/cpu/kernels/sub/neon/qsymm16.cpp
index 4dfdc0e78c..4dfdc0e78c 100644
--- a/src/core/cpu/kernels/sub/neon/qsymm16.cpp
+++ b/src/cpu/kernels/sub/neon/qsymm16.cpp
diff --git a/src/runtime/cpu/operators/CpuActivation.cpp b/src/cpu/operators/CpuActivation.cpp
index 0b43b322ad..d9330a8156 100644
--- a/src/runtime/cpu/operators/CpuActivation.cpp
+++ b/src/cpu/operators/CpuActivation.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuActivation.h"
 
 #include "src/common/IOperator.h"
 #include "src/common/utils/LegacySupport.h"
-#include "src/core/cpu/kernels/CpuActivationKernel.h"
 #include "src/cpu/CpuContext.h"
+#include "src/cpu/kernels/CpuActivationKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuActivation.h b/src/cpu/operators/CpuActivation.h
index ded4a37edb..9b97c9d24f 100644
--- a/src/runtime/cpu/operators/CpuActivation.h
+++ b/src/cpu/operators/CpuActivation.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_ACTIVATION_H
 #define ARM_COMPUTE_CPU_ACTIVATION_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuAdd.cpp b/src/cpu/operators/CpuAdd.cpp
index 23b09aca4f..42a7b99ceb 100644
--- a/src/runtime/cpu/operators/CpuAdd.cpp
+++ b/src/cpu/operators/CpuAdd.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuAdd.h"
+#include "src/cpu/operators/CpuAdd.h"
 
-#include "src/core/cpu/kernels/CpuAddKernel.h"
+#include "src/cpu/kernels/CpuAddKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuAdd.h b/src/cpu/operators/CpuAdd.h
index 3ff135fe41..d8ec620aeb 100644
--- a/src/runtime/cpu/operators/CpuAdd.h
+++ b/src/cpu/operators/CpuAdd.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_ADD_H
 #define ARM_COMPUTE_CPU_ADD_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuCast.cpp b/src/cpu/operators/CpuCast.cpp
index 5a4f6c518e..d0980c75b6 100644
--- a/src/runtime/cpu/operators/CpuCast.cpp
+++ b/src/cpu/operators/CpuCast.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuCast.h"
+#include "src/cpu/operators/CpuCast.h"
 
-#include "src/core/cpu/kernels/CpuCastKernel.h"
+#include "src/cpu/kernels/CpuCastKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuCast.h b/src/cpu/operators/CpuCast.h
index 26f5740b86..5e5f3e0ca6 100644
--- a/src/runtime/cpu/operators/CpuCast.h
+++ b/src/cpu/operators/CpuCast.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_CAST_H
 #define ARM_COMPUTE_CPU_CAST_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuConcatenate.cpp b/src/cpu/operators/CpuConcatenate.cpp
index bb475b790e..92c1ef6bdf 100644
--- a/src/runtime/cpu/operators/CpuConcatenate.cpp
+++ b/src/cpu/operators/CpuConcatenate.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuConcatenate.h"
+#include "src/cpu/operators/CpuConcatenate.h"
 
-#include "src/core/cpu/kernels/CpuConcatenateBatchKernel.h"
-#include "src/core/cpu/kernels/CpuConcatenateDepthKernel.h"
-#include "src/core/cpu/kernels/CpuConcatenateHeightKernel.h"
-#include "src/core/cpu/kernels/CpuConcatenateWidthKernel.h"
+#include "src/cpu/kernels/CpuConcatenateBatchKernel.h"
+#include "src/cpu/kernels/CpuConcatenateDepthKernel.h"
+#include "src/cpu/kernels/CpuConcatenateHeightKernel.h"
+#include "src/cpu/kernels/CpuConcatenateWidthKernel.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
diff --git a/src/runtime/cpu/operators/CpuConcatenate.h b/src/cpu/operators/CpuConcatenate.h
index 55eab54996..001ac68162 100644
--- a/src/runtime/cpu/operators/CpuConcatenate.h
+++ b/src/cpu/operators/CpuConcatenate.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CPU_CONCATENATE_H
 #define ARM_COMPUTE_CPU_CONCATENATE_H
 
-#include "src/core/cpu/ICpuKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include <vector>
 
diff --git a/src/runtime/cpu/operators/CpuConv2d.cpp b/src/cpu/operators/CpuConv2d.cpp
index cff9238308..3878e0de58 100644
--- a/src/runtime/cpu/operators/CpuConv2d.cpp
+++ b/src/cpu/operators/CpuConv2d.cpp
@@ -21,14 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuConv2d.h"
+#include "src/cpu/operators/CpuConv2d.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h"
-#include "src/runtime/cpu/operators/CpuDirectConv2d.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
-#include "src/runtime/cpu/operators/CpuGemmConv2d.h"
-#include "src/runtime/cpu/operators/CpuGemmDirectConv2d.h"
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/operators/CpuDirectConv2d.h"
+#include "src/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuGemmConv2d.h"
+#include "src/cpu/operators/CpuGemmDirectConv2d.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuConv2d.h b/src/cpu/operators/CpuConv2d.h
index d7b42deea1..daf068f361 100644
--- a/src/runtime/cpu/operators/CpuConv2d.h
+++ b/src/cpu/operators/CpuConv2d.h
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.cpp b/src/cpu/operators/CpuConvertFullyConnectedWeights.cpp
index 3f2f4e95cf..da744fc100 100644
--- a/src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.cpp
+++ b/src/cpu/operators/CpuConvertFullyConnectedWeights.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.h"
+#include "src/cpu/operators/CpuConvertFullyConnectedWeights.h"
 
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h"
+#include "src/cpu/kernels/CpuConvertFullyConnectedWeightsKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.h b/src/cpu/operators/CpuConvertFullyConnectedWeights.h
index 53ee17f6d1..ea70eee134 100644
--- a/src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.h
+++ b/src/cpu/operators/CpuConvertFullyConnectedWeights.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_CONVERT_FULLYCONNECTED_WEIGHTS_H
 #define ARM_COMPUTE_CPU_CONVERT_FULLYCONNECTED_WEIGHTS_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuCopy.cpp b/src/cpu/operators/CpuCopy.cpp
index 9fbe916163..2eecc2390e 100644
--- a/src/runtime/cpu/operators/CpuCopy.cpp
+++ b/src/cpu/operators/CpuCopy.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuCopy.h"
+#include "src/cpu/operators/CpuCopy.h"
 
-#include "src/core/cpu/kernels/CpuCopyKernel.h"
+#include "src/cpu/kernels/CpuCopyKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuCopy.h b/src/cpu/operators/CpuCopy.h
index 861bbb7849..9ffde4e781 100644
--- a/src/runtime/cpu/operators/CpuCopy.h
+++ b/src/cpu/operators/CpuCopy.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_COPY_H
 #define ARM_COMPUTE_CPU_COPY_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDepthwiseConv2d.cpp b/src/cpu/operators/CpuDepthwiseConv2d.cpp
index 8141487125..071690e7a6 100644
--- a/src/runtime/cpu/operators/CpuDepthwiseConv2d.cpp
+++ b/src/cpu/operators/CpuDepthwiseConv2d.cpp
@@ -21,14 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuDepthwiseConv2d.h"
+#include "src/cpu/operators/CpuDepthwiseConv2d.h"
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/InfoHelpers.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
+#include "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDepthwiseConv2d.h b/src/cpu/operators/CpuDepthwiseConv2d.h
index dd4839b28a..15e52ef515 100644
--- a/src/runtime/cpu/operators/CpuDepthwiseConv2d.h
+++ b/src/cpu/operators/CpuDepthwiseConv2d.h
@@ -26,12 +26,12 @@
 
 #include "arm_compute/core/ITensorInfo.h"
 #include "arm_compute/core/experimental/Types.h"
-#include "src/core/cpu/ICpuKernel.h"
-#include "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
+#include "src/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuOperator.h"
+#include "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h"
+#include "src/cpu/operators/CpuPermute.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp b/src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp
index 660ac0163c..a353a66dc2 100644
--- a/src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp
+++ b/src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.cpp
@@ -22,14 +22,14 @@
  * SOFTWARE.
  */
 
-#include "src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h"
+#include "src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h"
 
 #include "arm_compute/core/ITensorInfo.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "src/core/CPP/Validate.h"
-#include "src/core/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/utils/AssemblyUtils.h"
+#include "src/cpu/kernels/internal/CpuDepthwiseConv2dAssemblyWrapperKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h b/src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h
index f3d3b618c6..af202ced5b 100644
--- a/src/runtime/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h
+++ b/src/cpu/operators/CpuDepthwiseConv2dAssemblyDispatch.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_DEPTHWISE_CONV2D_ASSEMBLY_DISPATCH_H
 
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDequantize.cpp b/src/cpu/operators/CpuDequantize.cpp
index 80a2e28aee..7c03571f40 100644
--- a/src/runtime/cpu/operators/CpuDequantize.cpp
+++ b/src/cpu/operators/CpuDequantize.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuDequantize.h"
+#include "src/cpu/operators/CpuDequantize.h"
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuDequantizeKernel.h"
+#include "src/cpu/kernels/CpuDequantizeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDequantize.h b/src/cpu/operators/CpuDequantize.h
index fdbd6a57c2..dbfc0c612a 100644
--- a/src/runtime/cpu/operators/CpuDequantize.h
+++ b/src/cpu/operators/CpuDequantize.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_DEQUANTIZE_H
 #define ARM_COMPUTE_CPU_DEQUANTIZE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuDirectConv2d.cpp b/src/cpu/operators/CpuDirectConv2d.cpp
index 8812b777a3..ec52dbf153 100644
--- a/src/runtime/cpu/operators/CpuDirectConv2d.cpp
+++ b/src/cpu/operators/CpuDirectConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuDirectConv2d.h"
+#include "src/cpu/operators/CpuDirectConv2d.h"
 
 #include "arm_compute/core/PixelValue.h"
 #include "arm_compute/core/Utils.h"
diff --git a/src/runtime/cpu/operators/CpuDirectConv2d.h b/src/cpu/operators/CpuDirectConv2d.h
index c17b076f85..fa8d61e083 100644
--- a/src/runtime/cpu/operators/CpuDirectConv2d.h
+++ b/src/cpu/operators/CpuDirectConv2d.h
@@ -32,11 +32,11 @@
 #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/NEON/kernels/NEFillBorderKernel.h"
-#include "src/core/cpu/ICpuKernel.h"
-#include "src/core/cpu/kernels/CpuDirectConv2dKernel.h"
-#include "src/core/cpu/kernels/CpuDirectConv2dOutputStageKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
+#include "src/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuOperator.h"
+#include "src/cpu/kernels/CpuDirectConv2dKernel.h"
+#include "src/cpu/kernels/CpuDirectConv2dOutputStageKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuElementwise.cpp b/src/cpu/operators/CpuElementwise.cpp
index 8953d4769c..4f767434f3 100644
--- a/src/runtime/cpu/operators/CpuElementwise.cpp
+++ b/src/cpu/operators/CpuElementwise.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuElementwise.h"
-#include "src/core/cpu/kernels/CpuElementwiseKernel.h"
+#include "src/cpu/operators/CpuElementwise.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/CpuElementwiseKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuElementwise.h b/src/cpu/operators/CpuElementwise.h
index ef5caf2825..b6c61cf245 100644
--- a/src/runtime/cpu/operators/CpuElementwise.h
+++ b/src/cpu/operators/CpuElementwise.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_ELEMENTWISE_H
 #define ARM_COMPUTE_CPU_ELEMENTWISE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuElementwiseUnary.cpp b/src/cpu/operators/CpuElementwiseUnary.cpp
index c79e6e9acf..7cf1488c44 100644
--- a/src/runtime/cpu/operators/CpuElementwiseUnary.cpp
+++ b/src/cpu/operators/CpuElementwiseUnary.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuElementwiseUnary.h"
-#include "src/core/cpu/kernels/CpuElementwiseUnaryKernel.h"
+#include "src/cpu/operators/CpuElementwiseUnary.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/CpuElementwiseUnaryKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuElementwiseUnary.h b/src/cpu/operators/CpuElementwiseUnary.h
index 5ea29e07e9..5e8e98d047 100644
--- a/src/runtime/cpu/operators/CpuElementwiseUnary.h
+++ b/src/cpu/operators/CpuElementwiseUnary.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_ELEMENTWISE_UNARY_H
 
 #include "arm_compute/core/Types.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFill.cpp b/src/cpu/operators/CpuFill.cpp
index 081e30ea17..c0b48f5830 100644
--- a/src/runtime/cpu/operators/CpuFill.cpp
+++ b/src/cpu/operators/CpuFill.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuFill.h"
+#include "src/cpu/operators/CpuFill.h"
 
-#include "src/core/cpu/kernels/CpuFillKernel.h"
+#include "src/cpu/kernels/CpuFillKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFill.h b/src/cpu/operators/CpuFill.h
index b946467da6..1cb99f5662 100644
--- a/src/runtime/cpu/operators/CpuFill.h
+++ b/src/cpu/operators/CpuFill.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_FILL_H
 #define ARM_COMPUTE_CPU_FILL_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFlatten.cpp b/src/cpu/operators/CpuFlatten.cpp
index 58e6e4b671..685e5b9238 100644
--- a/src/runtime/cpu/operators/CpuFlatten.cpp
+++ b/src/cpu/operators/CpuFlatten.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuFlatten.h"
+#include "src/cpu/operators/CpuFlatten.h"
 
-#include "src/core/cpu/kernels/CpuReshapeKernel.h"
+#include "src/cpu/kernels/CpuReshapeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFlatten.h b/src/cpu/operators/CpuFlatten.h
index 3e24a93429..0e9fcbdc35 100644
--- a/src/runtime/cpu/operators/CpuFlatten.h
+++ b/src/cpu/operators/CpuFlatten.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_FLATTEN_H
 #define ARM_COMPUTE_CPU_FLATTEN_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFloor.cpp b/src/cpu/operators/CpuFloor.cpp
index 4e169a04be..55f645847f 100644
--- a/src/runtime/cpu/operators/CpuFloor.cpp
+++ b/src/cpu/operators/CpuFloor.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuFloor.h"
+#include "src/cpu/operators/CpuFloor.h"
 
-#include "src/core/cpu/kernels/CpuFloorKernel.h"
+#include "src/cpu/kernels/CpuFloorKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFloor.h b/src/cpu/operators/CpuFloor.h
index 0cd0cc0b4e..6082f98867 100644
--- a/src/runtime/cpu/operators/CpuFloor.h
+++ b/src/cpu/operators/CpuFloor.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_FLOOR_H
 #define ARM_COMPUTE_CPU_FLOOR_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFullyConnected.cpp b/src/cpu/operators/CpuFullyConnected.cpp
index eeabce0753..cafb3484b6 100644
--- a/src/runtime/cpu/operators/CpuFullyConnected.cpp
+++ b/src/cpu/operators/CpuFullyConnected.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuFullyConnected.h"
+#include "src/cpu/operators/CpuFullyConnected.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensorPack.h"
@@ -29,14 +29,14 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuTransposeKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.h"
-#include "src/runtime/cpu/operators/CpuFlatten.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
-#include "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/CpuTransposeKernel.h"
+#include "src/cpu/operators/CpuConvertFullyConnectedWeights.h"
+#include "src/cpu/operators/CpuFlatten.h"
+#include "src/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuFullyConnected.h b/src/cpu/operators/CpuFullyConnected.h
index 498ceae68d..304ea3c62b 100644
--- a/src/runtime/cpu/operators/CpuFullyConnected.h
+++ b/src/cpu/operators/CpuFullyConnected.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_FULLY_CONNECTED_H
 #define ARM_COMPUTE_CPU_FULLY_CONNECTED_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include "arm_compute/core/TensorInfo.h"
 
diff --git a/src/runtime/cpu/operators/CpuGemm.cpp b/src/cpu/operators/CpuGemm.cpp
index bd3f231001..f7416315e9 100644
--- a/src/runtime/cpu/operators/CpuGemm.cpp
+++ b/src/cpu/operators/CpuGemm.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuGemm.h"
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
@@ -30,7 +30,7 @@
 #include "src/core/CPP/Validate.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 using namespace arm_compute::experimental;
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/runtime/cpu/operators/CpuGemm.h b/src/cpu/operators/CpuGemm.h
index 8d859791f5..334ab6c647 100644
--- a/src/runtime/cpu/operators/CpuGemm.h
+++ b/src/cpu/operators/CpuGemm.h
@@ -24,18 +24,18 @@
 #ifndef ARM_COMPUTE_CPU_GEMM_H
 #define ARM_COMPUTE_CPU_GEMM_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
-#include "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
-#include "src/core/cpu/kernels/CpuGemmMatrixAdditionKernel.h"
-#include "src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
-#include "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/CpuAdd.h"
-#include "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
+#include "src/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
+#include "src/cpu/kernels/CpuGemmMatrixAdditionKernel.h"
+#include "src/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
+#include "src/cpu/kernels/CpuGemmTranspose1xWKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuAdd.h"
+#include "src/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuGemmConv2d.cpp b/src/cpu/operators/CpuGemmConv2d.cpp
index a81dd8a661..5010792a28 100644
--- a/src/runtime/cpu/operators/CpuGemmConv2d.cpp
+++ b/src/cpu/operators/CpuGemmConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuGemmConv2d.h"
+#include "src/cpu/operators/CpuGemmConv2d.h"
 
 #include "arm_compute/core/Size2D.h"
 #include "arm_compute/core/TensorInfo.h"
@@ -31,15 +31,15 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
-#include "src/core/cpu/kernels/CpuCol2ImKernel.h"
-#include "src/core/cpu/kernels/CpuIm2ColKernel.h"
-#include "src/core/cpu/kernels/CpuReshapeKernel.h"
-#include "src/core/cpu/kernels/CpuWeightsReshapeKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
-#include "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
-#include "src/runtime/cpu/operators/CpuGemmLowpOutputStage.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/CpuCol2ImKernel.h"
+#include "src/cpu/kernels/CpuIm2ColKernel.h"
+#include "src/cpu/kernels/CpuReshapeKernel.h"
+#include "src/cpu/kernels/CpuWeightsReshapeKernel.h"
+#include "src/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
+#include "src/cpu/operators/CpuGemmLowpOutputStage.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 #include <set>
 #include <tuple>
diff --git a/src/runtime/cpu/operators/CpuGemmConv2d.h b/src/cpu/operators/CpuGemmConv2d.h
index 529256594f..e63e7169b0 100644
--- a/src/runtime/cpu/operators/CpuGemmConv2d.h
+++ b/src/cpu/operators/CpuGemmConv2d.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuGemmDirectConv2d.cpp b/src/cpu/operators/CpuGemmDirectConv2d.cpp
index 10eece99eb..2e17a21462 100644
--- a/src/runtime/cpu/operators/CpuGemmDirectConv2d.cpp
+++ b/src/cpu/operators/CpuGemmDirectConv2d.cpp
@@ -21,13 +21,13 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuGemmDirectConv2d.h"
+#include "src/cpu/operators/CpuGemmDirectConv2d.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/FunctionDescriptors.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 #include "support/Cast.h"
 
diff --git a/src/runtime/cpu/operators/CpuGemmDirectConv2d.h b/src/cpu/operators/CpuGemmDirectConv2d.h
index 7fb20b3037..e55a461f36 100644
--- a/src/runtime/cpu/operators/CpuGemmDirectConv2d.h
+++ b/src/cpu/operators/CpuGemmDirectConv2d.h
@@ -26,10 +26,10 @@
 
 #include "arm_compute/core/TensorInfo.h"
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
-#include "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
+#include "src/cpu/ICpuOperator.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuPermute.h"
+#include "src/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp b/src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp
index 7affc3f506..2074a89307 100644
--- a/src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp
+++ b/src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
+#include "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -35,16 +35,16 @@
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
 
-#include "src/core/cpu/kernels/CpuConvertQuantizedSignednessKernel.h"
-#include "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h"
-#include "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/CpuConvertQuantizedSignednessKernel.h"
+#include "src/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
+#include "src/cpu/kernels/CpuGemmLowpMatrixMultiplyKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpOffsetContributionKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpOffsetContributionOutputStageKernel.h"
+#include "src/cpu/kernels/CpuGemmTranspose1xWKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 using namespace arm_compute::misc::shape_calculator;
 using namespace arm_compute::experimental;
diff --git a/src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h b/src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h
index 1d0e470559..a7f62aeaa9 100644
--- a/src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h
+++ b/src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/TensorInfo.h"
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuGemmLowpOutputStage.cpp b/src/cpu/operators/CpuGemmLowpOutputStage.cpp
index e17f854a21..ebd3f60280 100644
--- a/src/runtime/cpu/operators/CpuGemmLowpOutputStage.cpp
+++ b/src/cpu/operators/CpuGemmLowpOutputStage.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuGemmLowpOutputStage.h"
+#include "src/cpu/operators/CpuGemmLowpOutputStage.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ScaleKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h"
+#include "src/cpu/kernels/CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuGemmLowpOutputStage.h b/src/cpu/operators/CpuGemmLowpOutputStage.h
index bed88a60d5..39394f6b5f 100644
--- a/src/runtime/cpu/operators/CpuGemmLowpOutputStage.h
+++ b/src/cpu/operators/CpuGemmLowpOutputStage.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_GEMMLOWP_OUTPUT_STAGE_H
 
 #include "arm_compute/core/Types.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 /** This file contains all available output stages for GEMMLowp.
  *
diff --git a/src/runtime/cpu/operators/CpuMul.cpp b/src/cpu/operators/CpuMul.cpp
index 2f3d442a70..06a68d64a8 100644
--- a/src/runtime/cpu/operators/CpuMul.cpp
+++ b/src/cpu/operators/CpuMul.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuMul.h"
+#include "src/cpu/operators/CpuMul.h"
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuMulKernel.h"
+#include "src/cpu/kernels/CpuMulKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuMul.h b/src/cpu/operators/CpuMul.h
index da518c4461..576a357d42 100644
--- a/src/runtime/cpu/operators/CpuMul.h
+++ b/src/cpu/operators/CpuMul.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_MUL_H
 
 #include "arm_compute/core/ITensorInfo.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuPRelu.h b/src/cpu/operators/CpuPRelu.h
index a6859f95d9..084474e2ba 100644
--- a/src/runtime/cpu/operators/CpuPRelu.h
+++ b/src/cpu/operators/CpuPRelu.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_PRELU_H
 #define ARM_COMPUTE_CPU_PRELU_H
 
-#include "src/runtime/cpu/operators/CpuElementwise.h"
+#include "src/cpu/operators/CpuElementwise.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuPermute.cpp b/src/cpu/operators/CpuPermute.cpp
index 7fde1e3767..d730815313 100644
--- a/src/runtime/cpu/operators/CpuPermute.cpp
+++ b/src/cpu/operators/CpuPermute.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuPermute.h"
+#include "src/cpu/operators/CpuPermute.h"
 
-#include "src/core/cpu/kernels/CpuPermuteKernel.h"
+#include "src/cpu/kernels/CpuPermuteKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuPermute.h b/src/cpu/operators/CpuPermute.h
index 2500017c0e..0e0f3ae8db 100644
--- a/src/runtime/cpu/operators/CpuPermute.h
+++ b/src/cpu/operators/CpuPermute.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_PERMUTE_H
 #define ARM_COMPUTE_CPU_PERMUTE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuPool2d.cpp b/src/cpu/operators/CpuPool2d.cpp
index e746c8fb3b..6059c75dd2 100644
--- a/src/runtime/cpu/operators/CpuPool2d.cpp
+++ b/src/cpu/operators/CpuPool2d.cpp
@@ -21,14 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuPool2d.h"
+#include "src/cpu/operators/CpuPool2d.h"
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "src/core/NEON/kernels/NEFillBorderKernel.h"
-#include "src/core/cpu/kernels/CpuPool2dKernel.h"
-#include "src/core/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h"
+#include "src/cpu/kernels/CpuPool2dKernel.h"
+#include "src/cpu/kernels/internal/CpuPool2dAssemblyWrapperKernel.h"
 
 using namespace arm_compute::experimental;
 
diff --git a/src/runtime/cpu/operators/CpuPool2d.h b/src/cpu/operators/CpuPool2d.h
index 7feff91612..471637164f 100644
--- a/src/runtime/cpu/operators/CpuPool2d.h
+++ b/src/cpu/operators/CpuPool2d.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/experimental/Types.h"
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuQuantize.cpp b/src/cpu/operators/CpuQuantize.cpp
index 5af7f6343b..0bfcc21942 100644
--- a/src/runtime/cpu/operators/CpuQuantize.cpp
+++ b/src/cpu/operators/CpuQuantize.cpp
@@ -22,12 +22,12 @@
  * SOFTWARE.
  */
 
-#include "src/runtime/cpu/operators/CpuQuantize.h"
+#include "src/cpu/operators/CpuQuantize.h"
 
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuQuantizeKernel.h"
+#include "src/cpu/kernels/CpuQuantizeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuQuantize.h b/src/cpu/operators/CpuQuantize.h
index 9a34a36bcc..ec1134fee4 100644
--- a/src/runtime/cpu/operators/CpuQuantize.h
+++ b/src/cpu/operators/CpuQuantize.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_QUANTIZE_H
 #define ARM_COMPUTE_CPU_QUANTIZE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuReshape.cpp b/src/cpu/operators/CpuReshape.cpp
index 33c9cb87b6..5d2b052e34 100644
--- a/src/runtime/cpu/operators/CpuReshape.cpp
+++ b/src/cpu/operators/CpuReshape.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuReshape.h"
+#include "src/cpu/operators/CpuReshape.h"
 
-#include "src/core/cpu/kernels/CpuReshapeKernel.h"
+#include "src/cpu/kernels/CpuReshapeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuReshape.h b/src/cpu/operators/CpuReshape.h
index 581b55e0ef..92dcb09aa9 100644
--- a/src/runtime/cpu/operators/CpuReshape.h
+++ b/src/cpu/operators/CpuReshape.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_RESHAPE_H
 #define ARM_COMPUTE_CPU_RESHAPE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuScale.cpp b/src/cpu/operators/CpuScale.cpp
index 475cb2d4e8..9e35bccec5 100644
--- a/src/runtime/cpu/operators/CpuScale.cpp
+++ b/src/cpu/operators/CpuScale.cpp
@@ -21,14 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuScale.h"
+#include "src/cpu/operators/CpuScale.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuScaleKernel.h"
 #include "src/core/utils/ScaleUtils.h"
+#include "src/cpu/kernels/CpuScaleKernel.h"
 #include "support/Rounding.h"
 
 namespace arm_compute
diff --git a/src/runtime/cpu/operators/CpuScale.h b/src/cpu/operators/CpuScale.h
index b83e04bc42..f605af6712 100644
--- a/src/runtime/cpu/operators/CpuScale.h
+++ b/src/cpu/operators/CpuScale.h
@@ -27,8 +27,8 @@
 #include "arm_compute/core/ITensorInfo.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/experimental/Types.h"
-#include "src/core/cpu/ICpuKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/cpu/operators/CpuSoftmax.cpp b/src/cpu/operators/CpuSoftmax.cpp
index abbc539b19..b70ee7e4df 100644
--- a/src/runtime/cpu/operators/CpuSoftmax.cpp
+++ b/src/cpu/operators/CpuSoftmax.cpp
@@ -21,17 +21,17 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuSoftmax.h"
+#include "src/cpu/operators/CpuSoftmax.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/core/cpu/kernels/CpuSoftmaxKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/core/helpers/SoftmaxHelpers.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/CpuSoftmaxKernel.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 using namespace arm_compute::experimental;
 
diff --git a/src/runtime/cpu/operators/CpuSoftmax.h b/src/cpu/operators/CpuSoftmax.h
index a9ac803c09..20f3f006d3 100644
--- a/src/runtime/cpu/operators/CpuSoftmax.h
+++ b/src/cpu/operators/CpuSoftmax.h
@@ -26,9 +26,9 @@
 
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/experimental/Types.h"
-#include "src/core/cpu/ICpuKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
+#include "src/cpu/ICpuKernel.h"
+#include "src/cpu/ICpuOperator.h"
+#include "src/cpu/operators/CpuPermute.h"
 #include <memory>
 
 namespace arm_compute
diff --git a/src/runtime/cpu/operators/CpuSub.cpp b/src/cpu/operators/CpuSub.cpp
index 9baaaa9d67..0485a595c7 100644
--- a/src/runtime/cpu/operators/CpuSub.cpp
+++ b/src/cpu/operators/CpuSub.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuSub.h"
+#include "src/cpu/operators/CpuSub.h"
 
-#include "src/core/cpu/kernels/CpuSubKernel.h"
+#include "src/cpu/kernels/CpuSubKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuSub.h b/src/cpu/operators/CpuSub.h
index 07f5be89cd..025c928d8f 100644
--- a/src/runtime/cpu/operators/CpuSub.h
+++ b/src/cpu/operators/CpuSub.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_SUB_H
 #define ARM_COMPUTE_CPU_SUB_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuTranspose.cpp b/src/cpu/operators/CpuTranspose.cpp
index 51eeb90b8b..518227b464 100644
--- a/src/runtime/cpu/operators/CpuTranspose.cpp
+++ b/src/cpu/operators/CpuTranspose.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuTranspose.h"
+#include "src/cpu/operators/CpuTranspose.h"
 
-#include "src/core/cpu/kernels/CpuTransposeKernel.h"
+#include "src/cpu/kernels/CpuTransposeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuTranspose.h b/src/cpu/operators/CpuTranspose.h
index 0735924839..8934481ef6 100644
--- a/src/runtime/cpu/operators/CpuTranspose.h
+++ b/src/cpu/operators/CpuTranspose.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CPU_TRANSPOSE_H
 #define ARM_COMPUTE_CPU_TRANSPOSE_H
 
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/CpuWinogradConv2d.cpp b/src/cpu/operators/CpuWinogradConv2d.cpp
index 253280a951..8fca836b8e 100644
--- a/src/runtime/cpu/operators/CpuWinogradConv2d.cpp
+++ b/src/cpu/operators/CpuWinogradConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
@@ -32,12 +32,12 @@
 #include "src/core/CPP/Validate.h"
 #include "src/core/NEON/kernels/convolution/common/utils.hpp"
 #include "src/core/NEON/kernels/convolution/winograd/winograd.hpp"
-#include "src/core/cpu/kernels/CpuWinogradConv2dKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/CpuWinogradConv2dKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuPermute.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 #include "support/Cast.h"
 
diff --git a/src/runtime/cpu/operators/CpuWinogradConv2d.h b/src/cpu/operators/CpuWinogradConv2d.h
index b5b9c3f2e3..0abd110f73 100644
--- a/src/runtime/cpu/operators/CpuWinogradConv2d.h
+++ b/src/cpu/operators/CpuWinogradConv2d.h
@@ -27,12 +27,12 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/runtime/FunctionDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/cpu/kernels/CpuWinogradConv2dKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
-#include "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
+#include "src/cpu/ICpuOperator.h"
+#include "src/cpu/kernels/CpuWinogradConv2dKernel.h"
+#include "src/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuPermute.h"
+#include "src/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp b/src/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp
index 9786161dee..97893b0672 100644
--- a/src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp
+++ b/src/cpu/operators/internal/CpuGemmAssemblyDispatch.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
+#include "src/cpu/operators/internal/CpuGemmAssemblyDispatch.h"
 
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "src/core/CPP/Validate.h"
-#include "src/core/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h"
-#include "src/core/cpu/kernels/assembly/arm_gemm.hpp"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/core/utils/AssemblyUtils.h"
-#include "src/runtime/cpu/utils/CpuAuxTensorHandler.h"
+#include "src/cpu/kernels/assembly/CpuGemmAssemblyWrapperKernel.h"
+#include "src/cpu/kernels/assembly/arm_gemm.hpp"
+#include "src/cpu/utils/CpuAuxTensorHandler.h"
 
 #include <arm_neon.h>
 
diff --git a/src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h b/src/cpu/operators/internal/CpuGemmAssemblyDispatch.h
index 88cfed002a..a50f3634c2 100644
--- a/src/runtime/cpu/operators/internal/CpuGemmAssemblyDispatch.h
+++ b/src/cpu/operators/internal/CpuGemmAssemblyDispatch.h
@@ -25,7 +25,7 @@
 #define ARM_COMPUTE_CPU_INTERNAL_CPU_GEMM_ASSEMBLY_DISPATCH_H
 
 #include "src/core/common/Macros.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/cpu/utils/CpuAuxTensorHandler.h b/src/cpu/utils/CpuAuxTensorHandler.h
index ae1cffb659..ae1cffb659 100644
--- a/src/runtime/cpu/utils/CpuAuxTensorHandler.h
+++ b/src/cpu/utils/CpuAuxTensorHandler.h
diff --git a/src/core/gpu/cl/ClCompileContext.h b/src/gpu/cl/ClCompileContext.h
index e69cc0200f..e69cc0200f 100644
--- a/src/core/gpu/cl/ClCompileContext.h
+++ b/src/gpu/cl/ClCompileContext.h
diff --git a/src/core/gpu/cl/ClKernelLibrary.cpp b/src/gpu/cl/ClKernelLibrary.cpp
index 4a9ba874b1..5cd969e7f2 100644
--- a/src/core/gpu/cl/ClKernelLibrary.cpp
+++ b/src/gpu/cl/ClKernelLibrary.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/ClKernelLibrary.h"
+#include "src/gpu/cl/ClKernelLibrary.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Utils.h"
diff --git a/src/core/gpu/cl/ClKernelLibrary.h b/src/gpu/cl/ClKernelLibrary.h
index 42bec95032..42bec95032 100644
--- a/src/core/gpu/cl/ClKernelLibrary.h
+++ b/src/gpu/cl/ClKernelLibrary.h
diff --git a/src/core/gpu/cl/IClKernel.h b/src/gpu/cl/IClKernel.h
index 52ea3c9183..52ea3c9183 100644
--- a/src/core/gpu/cl/IClKernel.h
+++ b/src/gpu/cl/IClKernel.h
diff --git a/src/runtime/gpu/cl/IClOperator.h b/src/gpu/cl/IClOperator.h
index 049bf05dc1..049bf05dc1 100644
--- a/src/runtime/gpu/cl/IClOperator.h
+++ b/src/gpu/cl/IClOperator.h
diff --git a/src/core/gpu/cl/kernels/ClActivationKernel.cpp b/src/gpu/cl/kernels/ClActivationKernel.cpp
index 21c05632f9..13d55b3f5a 100644
--- a/src/core/gpu/cl/kernels/ClActivationKernel.cpp
+++ b/src/gpu/cl/kernels/ClActivationKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClActivationKernel.h"
+#include "src/gpu/cl/kernels/ClActivationKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClActivationKernel.h b/src/gpu/cl/kernels/ClActivationKernel.h
index 720b16a691..95e010889e 100644
--- a/src/core/gpu/cl/kernels/ClActivationKernel.h
+++ b/src/gpu/cl/kernels/ClActivationKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_ACTIVATION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClBatchConcatenateKernel.cpp b/src/gpu/cl/kernels/ClBatchConcatenateKernel.cpp
index fba1b0e087..8c2af5ffb6 100644
--- a/src/core/gpu/cl/kernels/ClBatchConcatenateKernel.cpp
+++ b/src/gpu/cl/kernels/ClBatchConcatenateKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClBatchConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClBatchConcatenateKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClBatchConcatenateKernel.h b/src/gpu/cl/kernels/ClBatchConcatenateKernel.h
index 2963d7cdfd..f6b7c0ed09 100644
--- a/src/core/gpu/cl/kernels/ClBatchConcatenateKernel.h
+++ b/src/gpu/cl/kernels/ClBatchConcatenateKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_BATCH_CONCATENATE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClCastKernel.cpp b/src/gpu/cl/kernels/ClCastKernel.cpp
index fac9ebe5cf..48caf21d16 100644
--- a/src/core/gpu/cl/kernels/ClCastKernel.cpp
+++ b/src/gpu/cl/kernels/ClCastKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClCastKernel.h"
+#include "src/gpu/cl/kernels/ClCastKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClCastKernel.h b/src/gpu/cl/kernels/ClCastKernel.h
index 6bf3cd9e50..5c223fc5fa 100644
--- a/src/core/gpu/cl/kernels/ClCastKernel.h
+++ b/src/gpu/cl/kernels/ClCastKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_CAST_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClCol2ImKernel.cpp b/src/gpu/cl/kernels/ClCol2ImKernel.cpp
index a3d57115f9..ac7c4a43f9 100644
--- a/src/core/gpu/cl/kernels/ClCol2ImKernel.cpp
+++ b/src/gpu/cl/kernels/ClCol2ImKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClCol2ImKernel.h"
+#include "src/gpu/cl/kernels/ClCol2ImKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClCol2ImKernel.h b/src/gpu/cl/kernels/ClCol2ImKernel.h
index 74a9027628..e19b7c8e16 100644
--- a/src/core/gpu/cl/kernels/ClCol2ImKernel.h
+++ b/src/gpu/cl/kernels/ClCol2ImKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/Size2D.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp b/src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp
index d1abd274d6..716dec1f30 100644
--- a/src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp
+++ b/src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h"
+#include "src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h b/src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h
index 3976fd45db..16000e82f6 100644
--- a/src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h
+++ b/src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_CONVERT_FULLYCONNECTED_WEIGHTS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClCopyKernel.cpp b/src/gpu/cl/kernels/ClCopyKernel.cpp
index 98c6f34e60..eefdc925a4 100644
--- a/src/core/gpu/cl/kernels/ClCopyKernel.cpp
+++ b/src/gpu/cl/kernels/ClCopyKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClCopyKernel.h"
+#include "src/gpu/cl/kernels/ClCopyKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClCopyKernel.h b/src/gpu/cl/kernels/ClCopyKernel.h
index d2732c4e59..63fd806586 100644
--- a/src/core/gpu/cl/kernels/ClCopyKernel.h
+++ b/src/gpu/cl/kernels/ClCopyKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_COPY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClCropKernel.cpp b/src/gpu/cl/kernels/ClCropKernel.cpp
index a052ef53f9..c7e5537977 100644
--- a/src/core/gpu/cl/kernels/ClCropKernel.cpp
+++ b/src/gpu/cl/kernels/ClCropKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClCropKernel.h"
+#include "src/gpu/cl/kernels/ClCropKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClCropKernel.h b/src/gpu/cl/kernels/ClCropKernel.h
index d81912284e..2f166e184c 100644
--- a/src/core/gpu/cl/kernels/ClCropKernel.h
+++ b/src/gpu/cl/kernels/ClCropKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_CROP_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClDepthConcatenateKernel.cpp b/src/gpu/cl/kernels/ClDepthConcatenateKernel.cpp
index e3e384f748..d716f1e430 100644
--- a/src/core/gpu/cl/kernels/ClDepthConcatenateKernel.cpp
+++ b/src/gpu/cl/kernels/ClDepthConcatenateKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClDepthConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClDepthConcatenateKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClDepthConcatenateKernel.h b/src/gpu/cl/kernels/ClDepthConcatenateKernel.h
index 0f408477b1..4739677f3b 100644
--- a/src/core/gpu/cl/kernels/ClDepthConcatenateKernel.h
+++ b/src/gpu/cl/kernels/ClDepthConcatenateKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_DEPTH_CONCATENATE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClDequantizeKernel.cpp b/src/gpu/cl/kernels/ClDequantizeKernel.cpp
index d69da8716c..c2e8f2f91f 100644
--- a/src/core/gpu/cl/kernels/ClDequantizeKernel.cpp
+++ b/src/gpu/cl/kernels/ClDequantizeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClDequantizeKernel.h"
+#include "src/gpu/cl/kernels/ClDequantizeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClDequantizeKernel.h b/src/gpu/cl/kernels/ClDequantizeKernel.h
index 0912e1b228..a32f506c9a 100644
--- a/src/core/gpu/cl/kernels/ClDequantizeKernel.h
+++ b/src/gpu/cl/kernels/ClDequantizeKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_DEQUANTIZE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClDirectConv2dKernel.cpp b/src/gpu/cl/kernels/ClDirectConv2dKernel.cpp
index 7b98671da2..cbeb9c43e9 100644
--- a/src/core/gpu/cl/kernels/ClDirectConv2dKernel.cpp
+++ b/src/gpu/cl/kernels/ClDirectConv2dKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClDirectConv2dKernel.h"
+#include "src/gpu/cl/kernels/ClDirectConv2dKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -35,9 +35,9 @@
 #include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLUtils.h"
 #include "src/core/CL/CLValidate.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "support/Cast.h"
 #include "support/StringSupport.h"
 namespace arm_compute
diff --git a/src/core/gpu/cl/kernels/ClDirectConv2dKernel.h b/src/gpu/cl/kernels/ClDirectConv2dKernel.h
index 4880d4a668..4041c7bf27 100644
--- a/src/core/gpu/cl/kernels/ClDirectConv2dKernel.h
+++ b/src/gpu/cl/kernels/ClDirectConv2dKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_DIRECT_CONV2D_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp b/src/gpu/cl/kernels/ClElementwiseKernel.cpp
index 3d9f0b6fcf..65cd052995 100644
--- a/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp
+++ b/src/gpu/cl/kernels/ClElementwiseKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClElementwiseKernel.h b/src/gpu/cl/kernels/ClElementwiseKernel.h
index 4525cec55b..3783e1571c 100644
--- a/src/core/gpu/cl/kernels/ClElementwiseKernel.h
+++ b/src/gpu/cl/kernels/ClElementwiseKernel.h
@@ -26,8 +26,8 @@
 
 #include "src/core/KernelTypes.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp b/src/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp
index 1525c0fe54..1f09515b86 100644
--- a/src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp
+++ b/src/gpu/cl/kernels/ClElementwiseUnaryKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.h b/src/gpu/cl/kernels/ClElementwiseUnaryKernel.h
index 64cc2f7afc..0f270f25e8 100644
--- a/src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.h
+++ b/src/gpu/cl/kernels/ClElementwiseUnaryKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_ELEMENTWISE_UNARY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClFillKernel.cpp b/src/gpu/cl/kernels/ClFillKernel.cpp
index f213bf8e6a..a9345ee334 100644
--- a/src/core/gpu/cl/kernels/ClFillKernel.cpp
+++ b/src/gpu/cl/kernels/ClFillKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClFillKernel.h"
+#include "src/gpu/cl/kernels/ClFillKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClFillKernel.h b/src/gpu/cl/kernels/ClFillKernel.h
index ecc2546e4a..f25cf928ad 100644
--- a/src/core/gpu/cl/kernels/ClFillKernel.h
+++ b/src/gpu/cl/kernels/ClFillKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_FILL_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClFloorKernel.cpp b/src/gpu/cl/kernels/ClFloorKernel.cpp
index 2047128963..1cb78242cb 100644
--- a/src/core/gpu/cl/kernels/ClFloorKernel.cpp
+++ b/src/gpu/cl/kernels/ClFloorKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClFloorKernel.h"
+#include "src/gpu/cl/kernels/ClFloorKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClFloorKernel.h b/src/gpu/cl/kernels/ClFloorKernel.h
index 57c9906f2c..6e413340ba 100644
--- a/src/core/gpu/cl/kernels/ClFloorKernel.h
+++ b/src/gpu/cl/kernels/ClFloorKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_FLOOR_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp
index ec0a3bf8e0..cb03c6255f 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h
index eaa125fbf2..4b328e0ab8 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp
index 44fda01ded..6446b4ce38 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h
index 99cff011d1..a16f500f11 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp
index 9d626936ff..bacf07fb4b 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h
index 9e52b38249..a77604db7c 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp
index e491cca914..5d2561d0dc 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h
index d1712f4f4b..48926e280b 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMMLOWP_OFFSET_CONTRIBUTION_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp
index 1e2d7d7efe..a8a8207504 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h
index 977f2eac53..cee04473c4 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMMLOWP_OFFSET_CONTRIBUTION_OUTPUT_STAGE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
index 8aec1654d9..c50023c3dd 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h
index c935aa7ec4..69b5fc5018 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMMLOWP_QUANTIZEDOWN_INT32_SCALE_BY_FIXEDPOINT_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp
index 9b488ff329..c5cea3d17d 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h
index eff8c4b2be..8eda24d25f 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMMLOWP_QUANTIZEDOWN_INT32_SCALE_BY_FLOAT_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
index 9a25973a93..5469ea9602 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h
index c5374755c8..84c5060362 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMMLOWP_QUANTIZEDOWN_INT32_SCALE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp
index b4886805fb..7f6f5731d8 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpReductionKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h b/src/gpu/cl/kernels/ClGemmLowpReductionKernel.h
index 11188ed062..7119b5fee0 100644
--- a/src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h
+++ b/src/gpu/cl/kernels/ClGemmLowpReductionKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp b/src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp
index 6079644935..4e934f0f33 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h b/src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h
index c303f78b07..c16e3279f5 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMM_MATRIXMULTIPLY_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp b/src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp
index 5ae55ab04a..448d35353b 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h b/src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h
index cd7bf278c2..26dec918cd 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp
index 591834f762..959b3ab21d 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -35,10 +35,10 @@
 #include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLUtils.h"
 #include "src/core/CL/CLValidate.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
 #include "src/core/utils/helpers/float_ops.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "support/Cast.h"
 #include "support/StringSupport.h"
 
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h
index b8ae4b9ae3..435a3a67f6 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMM_MATRIXMULTIPLY_RESHAPED_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 #include "arm_compute/core/KernelDescriptors.h"
 
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp
index 32ee0f9705..149c92b7a9 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
@@ -31,10 +31,10 @@
 #include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLUtils.h"
 #include "src/core/CL/CLValidate.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
 #include "src/core/utils/helpers/float_ops.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "support/Cast.h"
 #include "support/StringSupport.h"
 
diff --git a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h
index 3d6164eca9..3be96d3add 100644
--- a/src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h
+++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMM_MATRIXMULTIPLY_RESHAPED_ONLY_RHS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 #include "arm_compute/core/KernelDescriptors.h"
 
diff --git a/src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp b/src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp
index f92945e2a4..4a01c77d0a 100644
--- a/src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h b/src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h
index 73d811f3c3..69ec8f04f0 100644
--- a/src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h
+++ b/src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMM_RESHAPE_LHS_MATRIX_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp b/src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp
index 3a6f3c7e8f..778b9b9fa2 100644
--- a/src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -33,9 +33,9 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLValidate.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include "support/Cast.h"
 #include "support/StringSupport.h"
 
diff --git a/src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h b/src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h
index 27f80d3428..31eaa46e02 100644
--- a/src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h
+++ b/src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_GEMM_RESHAPE_RHS_MATRIX_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClHeightConcatenateKernel.cpp b/src/gpu/cl/kernels/ClHeightConcatenateKernel.cpp
index 9ff30eedcd..688308098a 100644
--- a/src/core/gpu/cl/kernels/ClHeightConcatenateKernel.cpp
+++ b/src/gpu/cl/kernels/ClHeightConcatenateKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClHeightConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClHeightConcatenateKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClHeightConcatenateKernel.h b/src/gpu/cl/kernels/ClHeightConcatenateKernel.h
index 0733078fc2..1e544d3025 100644
--- a/src/core/gpu/cl/kernels/ClHeightConcatenateKernel.h
+++ b/src/gpu/cl/kernels/ClHeightConcatenateKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_HEIGHT_CONCATENATE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClIm2ColKernel.cpp b/src/gpu/cl/kernels/ClIm2ColKernel.cpp
index 61ee443aa5..c42762b99c 100644
--- a/src/core/gpu/cl/kernels/ClIm2ColKernel.cpp
+++ b/src/gpu/cl/kernels/ClIm2ColKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClIm2ColKernel.h"
+#include "src/gpu/cl/kernels/ClIm2ColKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClIm2ColKernel.h b/src/gpu/cl/kernels/ClIm2ColKernel.h
index d1443f0434..a637ad215d 100644
--- a/src/core/gpu/cl/kernels/ClIm2ColKernel.h
+++ b/src/gpu/cl/kernels/ClIm2ColKernel.h
@@ -27,8 +27,8 @@
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/Size2D.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClMulKernel.cpp b/src/gpu/cl/kernels/ClMulKernel.cpp
index 7c4dddc20e..0bf1932085 100644
--- a/src/core/gpu/cl/kernels/ClMulKernel.cpp
+++ b/src/gpu/cl/kernels/ClMulKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClMulKernel.h"
+#include "src/gpu/cl/kernels/ClMulKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClMulKernel.h b/src/gpu/cl/kernels/ClMulKernel.h
index 2ee182b932..41c862eb03 100644
--- a/src/core/gpu/cl/kernels/ClMulKernel.h
+++ b/src/gpu/cl/kernels/ClMulKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_MUL_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClPermuteKernel.cpp b/src/gpu/cl/kernels/ClPermuteKernel.cpp
index 722bf454f2..8d4655114b 100644
--- a/src/core/gpu/cl/kernels/ClPermuteKernel.cpp
+++ b/src/gpu/cl/kernels/ClPermuteKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClPermuteKernel.h"
+#include "src/gpu/cl/kernels/ClPermuteKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClPermuteKernel.h b/src/gpu/cl/kernels/ClPermuteKernel.h
index 839e224ee4..0d349e739b 100644
--- a/src/core/gpu/cl/kernels/ClPermuteKernel.h
+++ b/src/gpu/cl/kernels/ClPermuteKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_PERMUTE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClPool2dKernel.cpp b/src/gpu/cl/kernels/ClPool2dKernel.cpp
index e522814b6d..04f2b142bd 100644
--- a/src/core/gpu/cl/kernels/ClPool2dKernel.cpp
+++ b/src/gpu/cl/kernels/ClPool2dKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClPool2dKernel.h"
+#include "src/gpu/cl/kernels/ClPool2dKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClPool2dKernel.h b/src/gpu/cl/kernels/ClPool2dKernel.h
index ab8c56a857..61d204dc68 100644
--- a/src/core/gpu/cl/kernels/ClPool2dKernel.h
+++ b/src/gpu/cl/kernels/ClPool2dKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_POOL2D_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClQuantizeKernel.cpp b/src/gpu/cl/kernels/ClQuantizeKernel.cpp
index 7900489db7..de222a99b2 100644
--- a/src/core/gpu/cl/kernels/ClQuantizeKernel.cpp
+++ b/src/gpu/cl/kernels/ClQuantizeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClQuantizeKernel.h"
+#include "src/gpu/cl/kernels/ClQuantizeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClQuantizeKernel.h b/src/gpu/cl/kernels/ClQuantizeKernel.h
index 1991a2fba8..aeab28febe 100644
--- a/src/core/gpu/cl/kernels/ClQuantizeKernel.h
+++ b/src/gpu/cl/kernels/ClQuantizeKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_QUANTIZE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClReshapeKernel.cpp b/src/gpu/cl/kernels/ClReshapeKernel.cpp
index fcda061930..246bd9c838 100644
--- a/src/core/gpu/cl/kernels/ClReshapeKernel.cpp
+++ b/src/gpu/cl/kernels/ClReshapeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClReshapeKernel.h"
+#include "src/gpu/cl/kernels/ClReshapeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClReshapeKernel.h b/src/gpu/cl/kernels/ClReshapeKernel.h
index 01e1ee84b9..db6ab5da58 100644
--- a/src/core/gpu/cl/kernels/ClReshapeKernel.h
+++ b/src/gpu/cl/kernels/ClReshapeKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_RESHAPE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClScaleKernel.cpp b/src/gpu/cl/kernels/ClScaleKernel.cpp
index ee4ee22aa0..9307f7d4fb 100644
--- a/src/core/gpu/cl/kernels/ClScaleKernel.cpp
+++ b/src/gpu/cl/kernels/ClScaleKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClScaleKernel.h"
+#include "src/gpu/cl/kernels/ClScaleKernel.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
diff --git a/src/core/gpu/cl/kernels/ClScaleKernel.h b/src/gpu/cl/kernels/ClScaleKernel.h
index 6674931296..dd09e92ee2 100644
--- a/src/core/gpu/cl/kernels/ClScaleKernel.h
+++ b/src/gpu/cl/kernels/ClScaleKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClSoftmaxKernel.cpp b/src/gpu/cl/kernels/ClSoftmaxKernel.cpp
index 1dd905d66e..4c00413469 100644
--- a/src/core/gpu/cl/kernels/ClSoftmaxKernel.cpp
+++ b/src/gpu/cl/kernels/ClSoftmaxKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClSoftmaxKernel.h"
+#include "src/gpu/cl/kernels/ClSoftmaxKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/experimental/Types.h"
diff --git a/src/core/gpu/cl/kernels/ClSoftmaxKernel.h b/src/gpu/cl/kernels/ClSoftmaxKernel.h
index a2ad02d6b7..a221e12132 100644
--- a/src/core/gpu/cl/kernels/ClSoftmaxKernel.h
+++ b/src/gpu/cl/kernels/ClSoftmaxKernel.h
@@ -27,8 +27,8 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClTransposeKernel.cpp b/src/gpu/cl/kernels/ClTransposeKernel.cpp
index 40bd4b034a..bdce2a2f5c 100644
--- a/src/core/gpu/cl/kernels/ClTransposeKernel.cpp
+++ b/src/gpu/cl/kernels/ClTransposeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClTransposeKernel.h"
+#include "src/gpu/cl/kernels/ClTransposeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClTransposeKernel.h b/src/gpu/cl/kernels/ClTransposeKernel.h
index c8379d44c7..b30d6f0281 100644
--- a/src/core/gpu/cl/kernels/ClTransposeKernel.h
+++ b/src/gpu/cl/kernels/ClTransposeKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_TRANSPOSE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWeightsReshapeKernel.cpp b/src/gpu/cl/kernels/ClWeightsReshapeKernel.cpp
index e3629f7706..8f36345076 100644
--- a/src/core/gpu/cl/kernels/ClWeightsReshapeKernel.cpp
+++ b/src/gpu/cl/kernels/ClWeightsReshapeKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h"
+#include "src/gpu/cl/kernels/ClWeightsReshapeKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
diff --git a/src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h b/src/gpu/cl/kernels/ClWeightsReshapeKernel.h
index de2f2d10cc..7364eb97ae 100644
--- a/src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h
+++ b/src/gpu/cl/kernels/ClWeightsReshapeKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_WEIGHTSRESHAPE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp b/src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp
index 8607620e92..6e7b7f6e14 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp
+++ b/src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h b/src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h
index 15e0757aec..8065fb9f75 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h
+++ b/src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_WIDTHCONCATENATE_2TENSORS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp b/src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp
index edbc23c1d3..a08490c565 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp
+++ b/src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h b/src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h
index 1e3f47f7fb..80afb3b85d 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h
+++ b/src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h
@@ -26,8 +26,8 @@
 #define ARM_COMPUTE_CL_WIDTH_CONCATENATE_4TENSORS_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenateKernel.cpp b/src/gpu/cl/kernels/ClWidthConcatenateKernel.cpp
index 5510c746f8..88b5a5e334 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenateKernel.cpp
+++ b/src/gpu/cl/kernels/ClWidthConcatenateKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWidthConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenateKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWidthConcatenateKernel.h b/src/gpu/cl/kernels/ClWidthConcatenateKernel.h
index 300c4beb30..71df077ada 100644
--- a/src/core/gpu/cl/kernels/ClWidthConcatenateKernel.h
+++ b/src/gpu/cl/kernels/ClWidthConcatenateKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_WIDTH_CONCATENATE_KERNEL_H
 
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp
index ae43fed12d..4ba6ba8a9a 100644
--- a/src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp
+++ b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.h"
+#include "src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.h b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h
index 145954fbb1..fe0c3da174 100644
--- a/src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.h
+++ b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp b/src/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp
index 538d8ae602..58874216bb 100644
--- a/src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp
+++ b/src/gpu/cl/kernels/ClWinogradInputTransformKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.h"
+#include "src/gpu/cl/kernels/ClWinogradInputTransformKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.h b/src/gpu/cl/kernels/ClWinogradInputTransformKernel.h
index 40fc2f387a..631f427b82 100644
--- a/src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.h
+++ b/src/gpu/cl/kernels/ClWinogradInputTransformKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp b/src/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp
index f6ade57e5d..a8cf8234ad 100644
--- a/src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp
+++ b/src/gpu/cl/kernels/ClWinogradOutputTransformKernel.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.h"
+#include "src/gpu/cl/kernels/ClWinogradOutputTransformKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.h b/src/gpu/cl/kernels/ClWinogradOutputTransformKernel.h
index 22b7f079c1..674d52c904 100644
--- a/src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.h
+++ b/src/gpu/cl/kernels/ClWinogradOutputTransformKernel.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/common/Macros.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/ClGemmHelpers.cpp b/src/gpu/cl/kernels/gemm/ClGemmHelpers.cpp
index 7866ccb679..1bf27ba277 100644
--- a/src/core/gpu/cl/kernels/gemm/ClGemmHelpers.cpp
+++ b/src/gpu/cl/kernels/gemm/ClGemmHelpers.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h b/src/gpu/cl/kernels/gemm/ClGemmHelpers.h
index 3fce8c9173..3fce8c9173 100644
--- a/src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h
+++ b/src/gpu/cl/kernels/gemm/ClGemmHelpers.h
diff --git a/src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h b/src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h
index a49836cfda..a49836cfda 100644
--- a/src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h
+++ b/src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp
index 9d11006703..b9eac2412e 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/GPUTarget.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h
index 385b96e40e..9af5dc4135 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_BIFROST_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_BIFROST_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp
index e3c129e3be..b9f36c7210 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/GPUTarget.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h
index 0ff5471f7c..c055753c48 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_MIDGARD_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_MIDGARD_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp
index 92767aca52..95a4d2bd69 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/GPUTarget.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h
index 17e4c9d339..f0f812fd46 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_VALHALL_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_NATIVE_VALHALL_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h b/src/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h
index ff6a0128af..cf8412830b 100644
--- a/src/core/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h
+++ b/src/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h
@@ -24,10 +24,10 @@
 #ifndef ARM_COMPUTE_CL_GEMM_NATIVE_KERNEL_CONFIGURATION_H
 #define ARM_COMPUTE_CL_GEMM_NATIVE_KERNEL_CONFIGURATION_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h"
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h"
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeBifrost.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeMidgard.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmDefaultConfigNativeValhall.h"
 
 #include <memory>
 
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp
index b030913a87..657018eb53 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp
+++ b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h"
+#include "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -29,7 +29,7 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h
index 52e6ce3f48..d86d1ba0a7 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h
+++ b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_BIFROST_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_BIFROST_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp
index 57e42c92b3..58d0873b86 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp
+++ b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h"
+#include "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/GPUTarget.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h
index 588cd64e0e..466eda00a6 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h
+++ b/src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_VALHALL_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_VALHALL_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h b/src/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h
index c990c89a91..1c32f1358b 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h
+++ b/src/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h
@@ -24,9 +24,9 @@
 #ifndef ARM_COMPUTE_CL_GEMM_RESHAPED_KERNEL_CONFIGURATION_H
 #define ARM_COMPUTE_CL_GEMM_RESHAPED_KERNEL_CONFIGURATION_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedBifrost.h"
+#include "src/gpu/cl/kernels/gemm/reshaped/ClGemmDefaultConfigReshapedValhall.h"
 
 #include <memory>
 
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp
index 417d540468..9c23d9c998 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp
+++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h"
+#include "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -29,7 +29,7 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 #include <utility>
 
 namespace arm_compute
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h
index 98c8e53569..321cbb5250 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h
+++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_RHS_ONLY_BIFROST_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_RHS_ONLY_BIFROST_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp
index 4c6e633896..a82084a8df 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp
+++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h"
+#include "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
@@ -29,7 +29,7 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <utility>
 
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h
index 6a11ddb748..c5e80a7ddc 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h
+++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_RHS_ONLY_VALHALL_H
 #define ARM_COMPUTE_CL_GEMM_DEFAULT_CONFIG_RESHAPED_RHS_ONLY_VALHALL_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
 
 namespace arm_compute
 {
diff --git a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h
index 8fd71276a0..1503e74eb6 100644
--- a/src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h
+++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h
@@ -24,9 +24,9 @@
 #ifndef ARM_COMPUTE_CL_GEMM_RESHAPED_ONLY_RHS_KERNEL_CONFIGURATION_H
 #define ARM_COMPUTE_CL_GEMM_RESHAPED_ONLY_RHS_KERNEL_CONFIGURATION_H
 
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyBifrost.h"
+#include "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClActivation.cpp b/src/gpu/cl/operators/ClActivation.cpp
index 34a2f94fdc..6b36cc34b4 100644
--- a/src/runtime/gpu/cl/operators/ClActivation.cpp
+++ b/src/gpu/cl/operators/ClActivation.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClActivation.h"
+#include "src/gpu/cl/operators/ClActivation.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClActivationKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClActivationKernel.h"
 
 #include "src/common/IOperator.h"
 #include "src/common/utils/LegacySupport.h"
diff --git a/src/runtime/gpu/cl/operators/ClActivation.h b/src/gpu/cl/operators/ClActivation.h
index 82ef8ac63a..75b38e8a00 100644
--- a/src/runtime/gpu/cl/operators/ClActivation.h
+++ b/src/gpu/cl/operators/ClActivation.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_ACTIVATION_H
 #define ARM_COMPUTE_CL_ACTIVATION_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClAdd.cpp b/src/gpu/cl/operators/ClAdd.cpp
index 01f550f819..e1a013a6b5 100644
--- a/src/runtime/gpu/cl/operators/ClAdd.cpp
+++ b/src/gpu/cl/operators/ClAdd.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClAdd.h"
+#include "src/gpu/cl/operators/ClAdd.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClAdd.h b/src/gpu/cl/operators/ClAdd.h
index 7b84a767d6..d99f983ed0 100644
--- a/src/runtime/gpu/cl/operators/ClAdd.h
+++ b/src/gpu/cl/operators/ClAdd.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_ADD_H
 #define ARM_COMPUTE_CL_ADD_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCast.cpp b/src/gpu/cl/operators/ClCast.cpp
index 3f54004aa7..8911d208a7 100644
--- a/src/runtime/gpu/cl/operators/ClCast.cpp
+++ b/src/gpu/cl/operators/ClCast.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClCast.h"
+#include "src/gpu/cl/operators/ClCast.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClCastKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClCastKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCast.h b/src/gpu/cl/operators/ClCast.h
index 107eb2bfe9..1b67ff7c8e 100644
--- a/src/runtime/gpu/cl/operators/ClCast.h
+++ b/src/gpu/cl/operators/ClCast.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_CAST_H
 #define ARM_COMPUTE_CL_CAST_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClConcatenate.cpp b/src/gpu/cl/operators/ClConcatenate.cpp
index d3c05eae78..731d9b5054 100644
--- a/src/runtime/gpu/cl/operators/ClConcatenate.cpp
+++ b/src/gpu/cl/operators/ClConcatenate.cpp
@@ -21,17 +21,17 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClConcatenate.h"
+#include "src/gpu/cl/operators/ClConcatenate.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
-#include "src/core/gpu/cl/kernels/ClBatchConcatenateKernel.h"
-#include "src/core/gpu/cl/kernels/ClDepthConcatenateKernel.h"
-#include "src/core/gpu/cl/kernels/ClHeightConcatenateKernel.h"
-#include "src/core/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h"
-#include "src/core/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h"
-#include "src/core/gpu/cl/kernels/ClWidthConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClBatchConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClDepthConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClHeightConcatenateKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenate4TensorsKernel.h"
+#include "src/gpu/cl/kernels/ClWidthConcatenateKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
diff --git a/src/runtime/gpu/cl/operators/ClConcatenate.h b/src/gpu/cl/operators/ClConcatenate.h
index 153400bd73..de0cf84d2c 100644
--- a/src/runtime/gpu/cl/operators/ClConcatenate.h
+++ b/src/gpu/cl/operators/ClConcatenate.h
@@ -24,9 +24,9 @@
 #ifndef ARM_COMPUTE_CLCONCATENATE_H
 #define ARM_COMPUTE_CLCONCATENATE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/IClOperator.h"
 
 #include <vector>
 
diff --git a/src/runtime/gpu/cl/operators/ClConv2d.cpp b/src/gpu/cl/operators/ClConv2d.cpp
index 0cb3a968e6..c91a4831a8 100644
--- a/src/runtime/gpu/cl/operators/ClConv2d.cpp
+++ b/src/gpu/cl/operators/ClConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClConv2d.h"
+#include "src/gpu/cl/operators/ClConv2d.h"
 
 #include "arm_compute/core/PixelValue.h"
 #include "arm_compute/core/Utils.h"
@@ -30,9 +30,9 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h"
-#include "src/runtime/gpu/cl/operators/ClDirectConv2d.h"
-#include "src/runtime/gpu/cl/operators/ClGemmConv2d.h"
-#include "src/runtime/gpu/cl/operators/ClWinogradConv2d.h"
+#include "src/gpu/cl/operators/ClDirectConv2d.h"
+#include "src/gpu/cl/operators/ClGemmConv2d.h"
+#include "src/gpu/cl/operators/ClWinogradConv2d.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClConv2d.h b/src/gpu/cl/operators/ClConv2d.h
index cdf3b7df32..1c3a81c77a 100644
--- a/src/runtime/gpu/cl/operators/ClConv2d.h
+++ b/src/gpu/cl/operators/ClConv2d.h
@@ -26,9 +26,9 @@
 
 #include "arm_compute/core/Types.h"
 #include "arm_compute/runtime/FunctionDescriptors.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp b/src/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp
index 0d2f2925d3..61e33f2fdb 100644
--- a/src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp
+++ b/src/gpu/cl/operators/ClConvertFullyConnectedWeights.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
+#include "src/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClConvertFullyConnectedWeightsKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.h b/src/gpu/cl/operators/ClConvertFullyConnectedWeights.h
index 7ea35c5a8a..2794eb17b0 100644
--- a/src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.h
+++ b/src/gpu/cl/operators/ClConvertFullyConnectedWeights.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_CONVERT_FULLYCONNECTED_WEIGHTS_H
 #define ARM_COMPUTE_CL_CONVERT_FULLYCONNECTED_WEIGHTS_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCopy.cpp b/src/gpu/cl/operators/ClCopy.cpp
index 2bdb1f5ba1..c1a9f264b6 100644
--- a/src/runtime/gpu/cl/operators/ClCopy.cpp
+++ b/src/gpu/cl/operators/ClCopy.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClCopy.h"
+#include "src/gpu/cl/operators/ClCopy.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClCopyKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClCopyKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCopy.h b/src/gpu/cl/operators/ClCopy.h
index e8ea8125eb..9b427f9675 100644
--- a/src/runtime/gpu/cl/operators/ClCopy.h
+++ b/src/gpu/cl/operators/ClCopy.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_COPY_H
 
 #include "arm_compute/core/Window.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCrop.cpp b/src/gpu/cl/operators/ClCrop.cpp
index 17bb11912f..a6a1c8b103 100644
--- a/src/runtime/gpu/cl/operators/ClCrop.cpp
+++ b/src/gpu/cl/operators/ClCrop.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClCrop.h"
+#include "src/gpu/cl/operators/ClCrop.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClCropKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClCropKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClCrop.h b/src/gpu/cl/operators/ClCrop.h
index cca69d6d77..1cf1c9bff4 100644
--- a/src/runtime/gpu/cl/operators/ClCrop.h
+++ b/src/gpu/cl/operators/ClCrop.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_CROP_H
 
 #include "arm_compute/core/Window.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClDequantize.cpp b/src/gpu/cl/operators/ClDequantize.cpp
index 0c1391bb45..dbaa5f67df 100644
--- a/src/runtime/gpu/cl/operators/ClDequantize.cpp
+++ b/src/gpu/cl/operators/ClDequantize.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClDequantize.h"
+#include "src/gpu/cl/operators/ClDequantize.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClDequantizeKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClDequantizeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClDequantize.h b/src/gpu/cl/operators/ClDequantize.h
index 5bcdcb2113..ccaac2cd49 100644
--- a/src/runtime/gpu/cl/operators/ClDequantize.h
+++ b/src/gpu/cl/operators/ClDequantize.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_DEQUANTIZE_H
 #define ARM_COMPUTE_CL_DEQUANTIZE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClDirectConv2d.cpp b/src/gpu/cl/operators/ClDirectConv2d.cpp
index 13ef42a640..50e63beedc 100644
--- a/src/runtime/gpu/cl/operators/ClDirectConv2d.cpp
+++ b/src/gpu/cl/operators/ClDirectConv2d.cpp
@@ -21,13 +21,13 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClDirectConv2d.h"
+#include "src/gpu/cl/operators/ClDirectConv2d.h"
 
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClActivationKernel.h"
-#include "src/core/gpu/cl/kernels/ClDirectConv2dKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClActivationKernel.h"
+#include "src/gpu/cl/kernels/ClDirectConv2dKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClDirectConv2d.h b/src/gpu/cl/operators/ClDirectConv2d.h
index a2785b52e3..85365b76ff 100644
--- a/src/runtime/gpu/cl/operators/ClDirectConv2d.h
+++ b/src/gpu/cl/operators/ClDirectConv2d.h
@@ -24,9 +24,9 @@
 #ifndef ARM_COMPUTE_CL_DIRECT_CONV2D_H
 #define ARM_COMPUTE_CL_DIRECT_CONV2D_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/IClOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClElementwiseOperations.cpp b/src/gpu/cl/operators/ClElementwiseOperations.cpp
index e5b836a0d8..4e4cd5ae9d 100644
--- a/src/runtime/gpu/cl/operators/ClElementwiseOperations.cpp
+++ b/src/gpu/cl/operators/ClElementwiseOperations.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClElementwiseOperations.h"
+#include "src/gpu/cl/operators/ClElementwiseOperations.h"
 
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClElementwiseOperations.h b/src/gpu/cl/operators/ClElementwiseOperations.h
index c01b107d97..304b250d66 100644
--- a/src/runtime/gpu/cl/operators/ClElementwiseOperations.h
+++ b/src/gpu/cl/operators/ClElementwiseOperations.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_ELEMENTWISE_OPERATIONS_H
 #define ARM_COMPUTE_CL_ELEMENTWISE_OPERATIONS_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClElementwiseUnary.cpp b/src/gpu/cl/operators/ClElementwiseUnary.cpp
index 7b830a077f..24a603e8c3 100644
--- a/src/runtime/gpu/cl/operators/ClElementwiseUnary.cpp
+++ b/src/gpu/cl/operators/ClElementwiseUnary.cpp
@@ -21,9 +21,9 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClElementwiseUnary.h"
+#include "src/gpu/cl/operators/ClElementwiseUnary.h"
 
-#include "src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClElementwiseUnary.h b/src/gpu/cl/operators/ClElementwiseUnary.h
index b9acf6f5b8..a23b789ab5 100644
--- a/src/runtime/gpu/cl/operators/ClElementwiseUnary.h
+++ b/src/gpu/cl/operators/ClElementwiseUnary.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_ELEMENTWISE_UNARY_H
 #define ARM_COMPUTE_CL_ELEMENTWISE_UNARY_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFill.cpp b/src/gpu/cl/operators/ClFill.cpp
index 4d0afaef24..9e006c1649 100644
--- a/src/runtime/gpu/cl/operators/ClFill.cpp
+++ b/src/gpu/cl/operators/ClFill.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClFill.h"
+#include "src/gpu/cl/operators/ClFill.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClFillKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClFillKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFill.h b/src/gpu/cl/operators/ClFill.h
index cc79b915a7..c9289b2b95 100644
--- a/src/runtime/gpu/cl/operators/ClFill.h
+++ b/src/gpu/cl/operators/ClFill.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_FILL_H
 
 #include "arm_compute/core/Window.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFlatten.cpp b/src/gpu/cl/operators/ClFlatten.cpp
index 060b653dee..3283454fd6 100644
--- a/src/runtime/gpu/cl/operators/ClFlatten.cpp
+++ b/src/gpu/cl/operators/ClFlatten.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClFlatten.h"
+#include "src/gpu/cl/operators/ClFlatten.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClReshapeKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClReshapeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFlatten.h b/src/gpu/cl/operators/ClFlatten.h
index 8bd619b518..d2ce3b701d 100644
--- a/src/runtime/gpu/cl/operators/ClFlatten.h
+++ b/src/gpu/cl/operators/ClFlatten.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_FLATTEN_H
 #define ARM_COMPUTE_CL_FLATTEN_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFloor.cpp b/src/gpu/cl/operators/ClFloor.cpp
index 94e77c0c54..866bff2fad 100644
--- a/src/runtime/gpu/cl/operators/ClFloor.cpp
+++ b/src/gpu/cl/operators/ClFloor.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClFloor.h"
+#include "src/gpu/cl/operators/ClFloor.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClFloorKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClFloorKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFloor.h b/src/gpu/cl/operators/ClFloor.h
index 90bdee6c7e..746147335e 100644
--- a/src/runtime/gpu/cl/operators/ClFloor.h
+++ b/src/gpu/cl/operators/ClFloor.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_FLOOR_H
 #define ARM_COMPUTE_CL_FLOOR_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClFullyConnected.cpp b/src/gpu/cl/operators/ClFullyConnected.cpp
index 377168d864..8b7e336c9f 100644
--- a/src/runtime/gpu/cl/operators/ClFullyConnected.cpp
+++ b/src/gpu/cl/operators/ClFullyConnected.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClFullyConnected.h"
+#include "src/gpu/cl/operators/ClFullyConnected.h"
 
 #include "arm_compute/core/Size2D.h"
 #include "arm_compute/core/Validate.h"
@@ -31,12 +31,12 @@
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
 
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
-#include "src/runtime/gpu/cl/operators/ClFlatten.h"
-#include "src/runtime/gpu/cl/operators/ClGemm.h"
-#include "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
-#include "src/runtime/gpu/cl/operators/ClTranspose.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
+#include "src/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
+#include "src/gpu/cl/operators/ClFlatten.h"
+#include "src/gpu/cl/operators/ClGemm.h"
+#include "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
+#include "src/gpu/cl/operators/ClTranspose.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 
 #include "support/Cast.h"
 
diff --git a/src/runtime/gpu/cl/operators/ClFullyConnected.h b/src/gpu/cl/operators/ClFullyConnected.h
index 86f95756d5..dc5f9e5c9b 100644
--- a/src/runtime/gpu/cl/operators/ClFullyConnected.h
+++ b/src/gpu/cl/operators/ClFullyConnected.h
@@ -26,8 +26,8 @@
 
 #include "arm_compute/core/TensorInfo.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClGemm.cpp b/src/gpu/cl/operators/ClGemm.cpp
index 59bbabba26..625c057cf4 100644
--- a/src/runtime/gpu/cl/operators/ClGemm.cpp
+++ b/src/gpu/cl/operators/ClGemm.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClGemm.h"
+#include "src/gpu/cl/operators/ClGemm.h"
 
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
@@ -39,13 +39,13 @@
 #include "arm_compute/runtime/ITensorAllocator.h"
 
 #include "src/common/utils/Log.h"
-#include "src/core/gpu/cl/IClKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/core/utils/helpers/float_ops.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 #include "src/runtime/CL/gemm/CLGEMMKernelSelection.h"
 #include "src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
 
 #include "support/Cast.h"
 #include "utils/TypePrinter.h"
diff --git a/src/runtime/gpu/cl/operators/ClGemm.h b/src/gpu/cl/operators/ClGemm.h
index 254344e862..60bb78c371 100644
--- a/src/runtime/gpu/cl/operators/ClGemm.h
+++ b/src/gpu/cl/operators/ClGemm.h
@@ -28,15 +28,15 @@
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTypes.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/IClKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClGemmConv2d.cpp b/src/gpu/cl/operators/ClGemmConv2d.cpp
index 8c796e0712..0f625bc56a 100644
--- a/src/runtime/gpu/cl/operators/ClGemmConv2d.cpp
+++ b/src/gpu/cl/operators/ClGemmConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClGemmConv2d.h"
+#include "src/gpu/cl/operators/ClGemmConv2d.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/PixelValue.h"
@@ -32,15 +32,15 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "src/core/gpu/cl/kernels/ClActivationKernel.h"
-#include "src/core/gpu/cl/kernels/ClCol2ImKernel.h"
-#include "src/core/gpu/cl/kernels/ClIm2ColKernel.h"
-#include "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClGemm.h"
-#include "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
+#include "src/gpu/cl/kernels/ClActivationKernel.h"
+#include "src/gpu/cl/kernels/ClCol2ImKernel.h"
+#include "src/gpu/cl/kernels/ClIm2ColKernel.h"
+#include "src/gpu/cl/kernels/ClWeightsReshapeKernel.h"
+#include "src/gpu/cl/operators/ClGemm.h"
+#include "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 #include "support/Cast.h"
 
 namespace arm_compute
diff --git a/src/runtime/gpu/cl/operators/ClGemmConv2d.h b/src/gpu/cl/operators/ClGemmConv2d.h
index e16d029e71..9a5e381dd7 100644
--- a/src/runtime/gpu/cl/operators/ClGemmConv2d.h
+++ b/src/gpu/cl/operators/ClGemmConv2d.h
@@ -27,8 +27,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/runtime/FunctionDescriptors.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp b/src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp
index 0c72912642..f3c0ee1c8f 100644
--- a/src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp
+++ b/src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
+#include "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Error.h"
@@ -35,17 +35,17 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
-#include "src/core/gpu/cl/kernels/ClCastKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/MemoryHelpers.h"
+#include "src/gpu/cl/kernels/ClCastKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpOffsetContributionKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpOffsetContributionOutputStageKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 #include "src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
 
 #include "utils/TypePrinter.h"
 
diff --git a/src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h b/src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h
index 36a4257b86..1965e3f97b 100644
--- a/src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h
+++ b/src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h
@@ -27,8 +27,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/runtime/CL/CLTypes.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.cpp b/src/gpu/cl/operators/ClGemmLowpOutputStage.cpp
index 3477583c76..27fb89217c 100644
--- a/src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.cpp
+++ b/src/gpu/cl/operators/ClGemmLowpOutputStage.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.h"
+#include "src/gpu/cl/operators/ClGemmLowpOutputStage.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleByFloatKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.h b/src/gpu/cl/operators/ClGemmLowpOutputStage.h
index 33b82fcafa..3f1b04dcce 100644
--- a/src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.h
+++ b/src/gpu/cl/operators/ClGemmLowpOutputStage.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_GEMMLOWP_OUTPUT_STAGE_H
 #define ARM_COMPUTE_CL_GEMMLOWP_OUTPUT_STAGE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 /** This file contains all available output stages for GEMMLowp on OpenCL.
  *
diff --git a/src/runtime/gpu/cl/operators/ClLogicalNot.cpp b/src/gpu/cl/operators/ClLogicalNot.cpp
index 400efe450d..b909066e4c 100644
--- a/src/runtime/gpu/cl/operators/ClLogicalNot.cpp
+++ b/src/gpu/cl/operators/ClLogicalNot.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClLogicalNot.h"
+#include "src/gpu/cl/operators/ClLogicalNot.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClElementwiseUnaryKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClLogicalNot.h b/src/gpu/cl/operators/ClLogicalNot.h
index 782ac0848f..31d4a99be6 100644
--- a/src/runtime/gpu/cl/operators/ClLogicalNot.h
+++ b/src/gpu/cl/operators/ClLogicalNot.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_LOGICAL_NOT_H
 #define ARM_COMPUTE_CL_LOGICAL_NOT_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClMul.cpp b/src/gpu/cl/operators/ClMul.cpp
index d1e2bc806f..59d2b96bee 100644
--- a/src/runtime/gpu/cl/operators/ClMul.cpp
+++ b/src/gpu/cl/operators/ClMul.cpp
@@ -21,11 +21,11 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClMul.h"
+#include "src/gpu/cl/operators/ClMul.h"
 
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClMulKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClMulKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClMul.h b/src/gpu/cl/operators/ClMul.h
index 29d5885a1c..6a158c910d 100644
--- a/src/runtime/gpu/cl/operators/ClMul.h
+++ b/src/gpu/cl/operators/ClMul.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_MUL_H
 #define ARM_COMPUTE_CL_MUL_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPRelu.cpp b/src/gpu/cl/operators/ClPRelu.cpp
index d1ce14cc87..05717d5bb7 100644
--- a/src/runtime/gpu/cl/operators/ClPRelu.cpp
+++ b/src/gpu/cl/operators/ClPRelu.cpp
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClPRelu.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/operators/ClPRelu.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPRelu.h b/src/gpu/cl/operators/ClPRelu.h
index 3a02030635..8084ab86cd 100644
--- a/src/runtime/gpu/cl/operators/ClPRelu.h
+++ b/src/gpu/cl/operators/ClPRelu.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_PRELU_H
 #define ARM_COMPUTE_CL_PRELU_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPermute.cpp b/src/gpu/cl/operators/ClPermute.cpp
index 719bb6dac6..ed74e22b6c 100644
--- a/src/runtime/gpu/cl/operators/ClPermute.cpp
+++ b/src/gpu/cl/operators/ClPermute.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClPermute.h"
+#include "src/gpu/cl/operators/ClPermute.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClPermuteKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClPermuteKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPermute.h b/src/gpu/cl/operators/ClPermute.h
index 867aba010d..3e87329f9b 100644
--- a/src/runtime/gpu/cl/operators/ClPermute.h
+++ b/src/gpu/cl/operators/ClPermute.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_PERMUTE_H
 #define ARM_COMPUTE_CL_PERMUTE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPool2d.cpp b/src/gpu/cl/operators/ClPool2d.cpp
index 40c2b0a8ba..fdadd199fc 100644
--- a/src/runtime/gpu/cl/operators/ClPool2d.cpp
+++ b/src/gpu/cl/operators/ClPool2d.cpp
@@ -21,13 +21,13 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClPool2d.h"
+#include "src/gpu/cl/operators/ClPool2d.h"
 
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClPool2dKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClPool2dKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClPool2d.h b/src/gpu/cl/operators/ClPool2d.h
index 8ac386a64b..a041053bb3 100644
--- a/src/runtime/gpu/cl/operators/ClPool2d.h
+++ b/src/gpu/cl/operators/ClPool2d.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_POOL2D_H
 #define ARM_COMPUTE_CL_POOL2D_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 #include <memory>
 
diff --git a/src/runtime/gpu/cl/operators/ClQuantize.cpp b/src/gpu/cl/operators/ClQuantize.cpp
index 92bbb62ba5..915e0fdef0 100644
--- a/src/runtime/gpu/cl/operators/ClQuantize.cpp
+++ b/src/gpu/cl/operators/ClQuantize.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClQuantize.h"
+#include "src/gpu/cl/operators/ClQuantize.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClQuantizeKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClQuantizeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClQuantize.h b/src/gpu/cl/operators/ClQuantize.h
index b15d389cca..3e50fcefb3 100644
--- a/src/runtime/gpu/cl/operators/ClQuantize.h
+++ b/src/gpu/cl/operators/ClQuantize.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_QUANTIZE_H
 #define ARM_COMPUTE_CL_QUANTIZE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClReshape.cpp b/src/gpu/cl/operators/ClReshape.cpp
index d3fa9f10ab..2c1d1817d1 100644
--- a/src/runtime/gpu/cl/operators/ClReshape.cpp
+++ b/src/gpu/cl/operators/ClReshape.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClReshape.h"
+#include "src/gpu/cl/operators/ClReshape.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClReshapeKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClReshapeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClReshape.h b/src/gpu/cl/operators/ClReshape.h
index b3d9267be4..fee69a1c24 100644
--- a/src/runtime/gpu/cl/operators/ClReshape.h
+++ b/src/gpu/cl/operators/ClReshape.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_RESHAPE_H
 #define ARM_COMPUTE_CL_RESHAPE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClScale.cpp b/src/gpu/cl/operators/ClScale.cpp
index 5c8d754c7e..6dab66786a 100644
--- a/src/runtime/gpu/cl/operators/ClScale.cpp
+++ b/src/gpu/cl/operators/ClScale.cpp
@@ -21,13 +21,13 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClScale.h"
+#include "src/gpu/cl/operators/ClScale.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClScaleKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClScaleKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClScale.h b/src/gpu/cl/operators/ClScale.h
index 0ff78640f7..af97cf23e7 100644
--- a/src/runtime/gpu/cl/operators/ClScale.h
+++ b/src/gpu/cl/operators/ClScale.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_SCALE_H
 
 #include "arm_compute/core/KernelDescriptors.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClSoftmax.cpp b/src/gpu/cl/operators/ClSoftmax.cpp
index 975bb0b932..6b728f5354 100644
--- a/src/runtime/gpu/cl/operators/ClSoftmax.cpp
+++ b/src/gpu/cl/operators/ClSoftmax.cpp
@@ -21,13 +21,13 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClSoftmax.h"
+#include "src/gpu/cl/operators/ClSoftmax.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/ClSoftmaxKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/core/helpers/SoftmaxHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClPermute.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
+#include "src/gpu/cl/kernels/ClSoftmaxKernel.h"
+#include "src/gpu/cl/operators/ClPermute.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 #include "support/Cast.h"
 
 using namespace arm_compute::experimental;
diff --git a/src/runtime/gpu/cl/operators/ClSoftmax.h b/src/gpu/cl/operators/ClSoftmax.h
index c85b193d9d..6c9af585d6 100644
--- a/src/runtime/gpu/cl/operators/ClSoftmax.h
+++ b/src/gpu/cl/operators/ClSoftmax.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_CL_SOFTMAX_H
 
 #include "arm_compute/runtime/CL/CLTensor.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClSub.cpp b/src/gpu/cl/operators/ClSub.cpp
index 429f23a837..b94fef3cf9 100644
--- a/src/runtime/gpu/cl/operators/ClSub.cpp
+++ b/src/gpu/cl/operators/ClSub.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClSub.h"
+#include "src/gpu/cl/operators/ClSub.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClSub.h b/src/gpu/cl/operators/ClSub.h
index 2dac11c00e..902adbf39d 100644
--- a/src/runtime/gpu/cl/operators/ClSub.h
+++ b/src/gpu/cl/operators/ClSub.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_SUB_H
 #define ARM_COMPUTE_CL_SUB_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClTranspose.cpp b/src/gpu/cl/operators/ClTranspose.cpp
index 48f44282e8..6429451a42 100644
--- a/src/runtime/gpu/cl/operators/ClTranspose.cpp
+++ b/src/gpu/cl/operators/ClTranspose.cpp
@@ -21,10 +21,10 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClTranspose.h"
+#include "src/gpu/cl/operators/ClTranspose.h"
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/core/gpu/cl/kernels/ClTransposeKernel.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/kernels/ClTransposeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClTranspose.h b/src/gpu/cl/operators/ClTranspose.h
index dcd80820bb..3642fc23f9 100644
--- a/src/runtime/gpu/cl/operators/ClTranspose.h
+++ b/src/gpu/cl/operators/ClTranspose.h
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_CL_TRANSPOSE_H
 #define ARM_COMPUTE_CL_TRANSPOSE_H
 
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/operators/ClWinogradConv2d.cpp b/src/gpu/cl/operators/ClWinogradConv2d.cpp
index 07f90ddaef..fbf6442a80 100644
--- a/src/runtime/gpu/cl/operators/ClWinogradConv2d.cpp
+++ b/src/gpu/cl/operators/ClWinogradConv2d.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/runtime/gpu/cl/operators/ClWinogradConv2d.h"
+#include "src/gpu/cl/operators/ClWinogradConv2d.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Utils.h"
@@ -31,11 +31,11 @@
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/kernels/ClWinogradFilterTransformKernel.h"
-#include "src/core/gpu/cl/kernels/ClWinogradInputTransformKernel.h"
-#include "src/core/gpu/cl/kernels/ClWinogradOutputTransformKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
+#include "src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h"
+#include "src/gpu/cl/kernels/ClWinogradInputTransformKernel.h"
+#include "src/gpu/cl/kernels/ClWinogradOutputTransformKernel.h"
+#include "src/gpu/cl/utils/ClAuxTensorHandler.h"
 #include "support/Cast.h"
 
 using namespace arm_compute::experimental;
diff --git a/src/runtime/gpu/cl/operators/ClWinogradConv2d.h b/src/gpu/cl/operators/ClWinogradConv2d.h
index 83b31f1c99..eb2f7a72b2 100644
--- a/src/runtime/gpu/cl/operators/ClWinogradConv2d.h
+++ b/src/gpu/cl/operators/ClWinogradConv2d.h
@@ -26,9 +26,9 @@
 
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/ClCompileContext.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
-#include "src/runtime/gpu/cl/operators/ClGemm.h"
+#include "src/gpu/cl/ClCompileContext.h"
+#include "src/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/operators/ClGemm.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/gpu/cl/utils/ClAuxTensorHandler.h b/src/gpu/cl/utils/ClAuxTensorHandler.h
index af383489a1..af383489a1 100644
--- a/src/runtime/gpu/cl/utils/ClAuxTensorHandler.h
+++ b/src/gpu/cl/utils/ClAuxTensorHandler.h
diff --git a/src/runtime/CL/functions/CLActivationLayer.cpp b/src/runtime/CL/functions/CLActivationLayer.cpp
index 9c71b2aa7d..bf69868663 100644
--- a/src/runtime/CL/functions/CLActivationLayer.cpp
+++ b/src/runtime/CL/functions/CLActivationLayer.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/CL/CLRuntimeContext.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClActivation.h"
+#include "src/gpu/cl/operators/ClActivation.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLCast.cpp b/src/runtime/CL/functions/CLCast.cpp
index 53256ebed4..f9403afcb8 100644
--- a/src/runtime/CL/functions/CLCast.cpp
+++ b/src/runtime/CL/functions/CLCast.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClCast.h"
+#include "src/gpu/cl/operators/ClCast.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLConcatenateLayer.cpp b/src/runtime/CL/functions/CLConcatenateLayer.cpp
index ea96e45bf8..8ab50beacd 100644
--- a/src/runtime/CL/functions/CLConcatenateLayer.cpp
+++ b/src/runtime/CL/functions/CLConcatenateLayer.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClConcatenate.h"
+#include "src/gpu/cl/operators/ClConcatenate.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp b/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
index 8189eee402..7780c0a444 100644
--- a/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
+++ b/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
+#include "src/gpu/cl/operators/ClConvertFullyConnectedWeights.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLConvolutionLayer.cpp b/src/runtime/CL/functions/CLConvolutionLayer.cpp
index b295a274bd..1f715d246d 100644
--- a/src/runtime/CL/functions/CLConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLConvolutionLayer.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h"
 #include "src/core/CL/ICLKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClConv2d.h"
+#include "src/gpu/cl/operators/ClConv2d.h"
 #include "support/Cast.h"
 
 namespace arm_compute
diff --git a/src/runtime/CL/functions/CLCopy.cpp b/src/runtime/CL/functions/CLCopy.cpp
index 98916bf38a..e8aaf85876 100644
--- a/src/runtime/CL/functions/CLCopy.cpp
+++ b/src/runtime/CL/functions/CLCopy.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClCopy.h"
+#include "src/gpu/cl/operators/ClCopy.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLCrop.cpp b/src/runtime/CL/functions/CLCrop.cpp
index 20cab4df5f..ff30837506 100644
--- a/src/runtime/CL/functions/CLCrop.cpp
+++ b/src/runtime/CL/functions/CLCrop.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClCrop.h"
+#include "src/gpu/cl/operators/ClCrop.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLDepthConvertLayer.cpp b/src/runtime/CL/functions/CLDepthConvertLayer.cpp
index 6aa370b23c..5930ff19f8 100644
--- a/src/runtime/CL/functions/CLDepthConvertLayer.cpp
+++ b/src/runtime/CL/functions/CLDepthConvertLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClCast.h"
+#include "src/gpu/cl/operators/ClCast.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLDequantizationLayer.cpp b/src/runtime/CL/functions/CLDequantizationLayer.cpp
index 3b104017e7..e11802ed1b 100644
--- a/src/runtime/CL/functions/CLDequantizationLayer.cpp
+++ b/src/runtime/CL/functions/CLDequantizationLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClDequantize.h"
+#include "src/gpu/cl/operators/ClDequantize.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
index 907e69d8d7..7bbb7e8e4a 100644
--- a/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
@@ -28,8 +28,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "src/runtime/gpu/cl/operators/ClActivation.h"
-#include "src/runtime/gpu/cl/operators/ClDirectConv2d.h"
+#include "src/gpu/cl/operators/ClActivation.h"
+#include "src/gpu/cl/operators/ClDirectConv2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLElementwiseOperations.cpp b/src/runtime/CL/functions/CLElementwiseOperations.cpp
index 60c699cbb8..936b37fb31 100644
--- a/src/runtime/CL/functions/CLElementwiseOperations.cpp
+++ b/src/runtime/CL/functions/CLElementwiseOperations.cpp
@@ -28,9 +28,9 @@
 #include "arm_compute/core/Types.h"
 #include "src/core/CL/ICLKernel.h"
 
-#include "src/runtime/gpu/cl/operators/ClAdd.h"
-#include "src/runtime/gpu/cl/operators/ClElementwiseOperations.h"
-#include "src/runtime/gpu/cl/operators/ClSub.h"
+#include "src/gpu/cl/operators/ClAdd.h"
+#include "src/gpu/cl/operators/ClElementwiseOperations.h"
+#include "src/gpu/cl/operators/ClSub.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLElementwiseUnaryLayer.cpp b/src/runtime/CL/functions/CLElementwiseUnaryLayer.cpp
index a45dd6f9a6..9dcd2d1891 100644
--- a/src/runtime/CL/functions/CLElementwiseUnaryLayer.cpp
+++ b/src/runtime/CL/functions/CLElementwiseUnaryLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClElementwiseUnary.h"
+#include "src/gpu/cl/operators/ClElementwiseUnary.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLFill.cpp b/src/runtime/CL/functions/CLFill.cpp
index b22d79fea4..6019a84aba 100644
--- a/src/runtime/CL/functions/CLFill.cpp
+++ b/src/runtime/CL/functions/CLFill.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClFill.h"
+#include "src/gpu/cl/operators/ClFill.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLFlattenLayer.cpp b/src/runtime/CL/functions/CLFlattenLayer.cpp
index 9563055276..32fc37552c 100644
--- a/src/runtime/CL/functions/CLFlattenLayer.cpp
+++ b/src/runtime/CL/functions/CLFlattenLayer.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/CL/ICLKernel.h"
 #include "src/core/helpers/AutoConfiguration.h"
-#include "src/runtime/gpu/cl/operators/ClFlatten.h"
+#include "src/gpu/cl/operators/ClFlatten.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLFloor.cpp b/src/runtime/CL/functions/CLFloor.cpp
index 4c5e482b10..8739e1803e 100644
--- a/src/runtime/CL/functions/CLFloor.cpp
+++ b/src/runtime/CL/functions/CLFloor.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClFloor.h"
+#include "src/gpu/cl/operators/ClFloor.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLFullyConnectedLayer.cpp b/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
index 4f9759c590..02b2042a6c 100644
--- a/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
+++ b/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClFullyConnected.h"
+#include "src/gpu/cl/operators/ClFullyConnected.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLGEMM.cpp b/src/runtime/CL/functions/CLGEMM.cpp
index 14b0633e09..cc6689c504 100644
--- a/src/runtime/CL/functions/CLGEMM.cpp
+++ b/src/runtime/CL/functions/CLGEMM.cpp
@@ -32,7 +32,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/runtime/CL/functions/CLGEMM.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClGemm.h"
+#include "src/gpu/cl/operators/ClGemm.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp b/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
index 563dbd414f..837527bac3 100644
--- a/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
@@ -32,7 +32,7 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClGemmConv2d.h"
+#include "src/gpu/cl/operators/ClGemmConv2d.h"
 #include "support/Cast.h"
 
 #include <cmath>
diff --git a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
index 1ae2dfbad6..d9029478a1 100644
--- a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
+++ b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
@@ -37,7 +37,7 @@
 #include "arm_compute/runtime/IMemoryManager.h"
 #include "src/core/helpers/MemoryHelpers.h"
 
-#include "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
+#include "src/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLGEMMLowpOutputStage.cpp b/src/runtime/CL/functions/CLGEMMLowpOutputStage.cpp
index 94d4c33fa2..6feed0d713 100644
--- a/src/runtime/CL/functions/CLGEMMLowpOutputStage.cpp
+++ b/src/runtime/CL/functions/CLGEMMLowpOutputStage.cpp
@@ -32,7 +32,7 @@
 #include "arm_compute/core/Types.h"
 
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClGemmLowpOutputStage.h"
+#include "src/gpu/cl/operators/ClGemmLowpOutputStage.h"
 
 #include <algorithm>
 
diff --git a/src/runtime/CL/functions/CLLSTMLayer.cpp b/src/runtime/CL/functions/CLLSTMLayer.cpp
index 9754bdcb82..0122162073 100644
--- a/src/runtime/CL/functions/CLLSTMLayer.cpp
+++ b/src/runtime/CL/functions/CLLSTMLayer.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/core/gpu/cl/kernels/ClTransposeKernel.h"
+#include "src/gpu/cl/kernels/ClTransposeKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLLogicalAnd.cpp b/src/runtime/CL/functions/CLLogicalAnd.cpp
index 98c98abed5..306957a8d1 100644
--- a/src/runtime/CL/functions/CLLogicalAnd.cpp
+++ b/src/runtime/CL/functions/CLLogicalAnd.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/runtime/CL/functions/CLLogicalAnd.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLLogicalNot.cpp b/src/runtime/CL/functions/CLLogicalNot.cpp
index 388d2bce86..a0504d7852 100644
--- a/src/runtime/CL/functions/CLLogicalNot.cpp
+++ b/src/runtime/CL/functions/CLLogicalNot.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClLogicalNot.h"
+#include "src/gpu/cl/operators/ClLogicalNot.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLLogicalOr.cpp b/src/runtime/CL/functions/CLLogicalOr.cpp
index 897963ab50..63524213f7 100644
--- a/src/runtime/CL/functions/CLLogicalOr.cpp
+++ b/src/runtime/CL/functions/CLLogicalOr.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/runtime/CL/functions/CLLogicalOr.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "src/core/gpu/cl/kernels/ClElementwiseKernel.h"
+#include "src/gpu/cl/kernels/ClElementwiseKernel.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLPReluLayer.cpp b/src/runtime/CL/functions/CLPReluLayer.cpp
index bb7aff218d..186e7b4ba2 100644
--- a/src/runtime/CL/functions/CLPReluLayer.cpp
+++ b/src/runtime/CL/functions/CLPReluLayer.cpp
@@ -24,8 +24,8 @@
 #include "arm_compute/runtime/CL/functions/CLPReluLayer.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "src/core/gpu/cl/IClKernel.h"
-#include "src/runtime/gpu/cl/operators/ClPRelu.h"
+#include "src/gpu/cl/IClKernel.h"
+#include "src/gpu/cl/operators/ClPRelu.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLPermute.cpp b/src/runtime/CL/functions/CLPermute.cpp
index c1da2a9eca..556e943152 100644
--- a/src/runtime/CL/functions/CLPermute.cpp
+++ b/src/runtime/CL/functions/CLPermute.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClPermute.h"
+#include "src/gpu/cl/operators/ClPermute.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLPixelWiseMultiplication.cpp b/src/runtime/CL/functions/CLPixelWiseMultiplication.cpp
index 932659268d..9d91e58367 100644
--- a/src/runtime/CL/functions/CLPixelWiseMultiplication.cpp
+++ b/src/runtime/CL/functions/CLPixelWiseMultiplication.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClMul.h"
+#include "src/gpu/cl/operators/ClMul.h"
 
 #include <utility>
 
diff --git a/src/runtime/CL/functions/CLPoolingLayer.cpp b/src/runtime/CL/functions/CLPoolingLayer.cpp
index 7ba911c342..0ebce318fa 100644
--- a/src/runtime/CL/functions/CLPoolingLayer.cpp
+++ b/src/runtime/CL/functions/CLPoolingLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClPool2d.h"
+#include "src/gpu/cl/operators/ClPool2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLQLSTMLayer.cpp b/src/runtime/CL/functions/CLQLSTMLayer.cpp
index 5df895a91c..6ddf555b5c 100644
--- a/src/runtime/CL/functions/CLQLSTMLayer.cpp
+++ b/src/runtime/CL/functions/CLQLSTMLayer.cpp
@@ -32,8 +32,8 @@
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
 #include "src/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/gpu/cl/kernels/ClGemmLowpReductionKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLQuantizationLayer.cpp b/src/runtime/CL/functions/CLQuantizationLayer.cpp
index e6451b2eb4..b249bdd1db 100644
--- a/src/runtime/CL/functions/CLQuantizationLayer.cpp
+++ b/src/runtime/CL/functions/CLQuantizationLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClQuantize.h"
+#include "src/gpu/cl/operators/ClQuantize.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLReshapeLayer.cpp b/src/runtime/CL/functions/CLReshapeLayer.cpp
index 060eddb96c..c51a3298c1 100644
--- a/src/runtime/CL/functions/CLReshapeLayer.cpp
+++ b/src/runtime/CL/functions/CLReshapeLayer.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClReshape.h"
+#include "src/gpu/cl/operators/ClReshape.h"
 
 /** [CLReshapeLayer snippet] **/
 namespace arm_compute
diff --git a/src/runtime/CL/functions/CLScale.cpp b/src/runtime/CL/functions/CLScale.cpp
index cbd93c1086..5b78989bfa 100644
--- a/src/runtime/CL/functions/CLScale.cpp
+++ b/src/runtime/CL/functions/CLScale.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClScale.h"
+#include "src/gpu/cl/operators/ClScale.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLSoftmaxLayer.cpp b/src/runtime/CL/functions/CLSoftmaxLayer.cpp
index de58bf1b02..d52352fc8d 100644
--- a/src/runtime/CL/functions/CLSoftmaxLayer.cpp
+++ b/src/runtime/CL/functions/CLSoftmaxLayer.cpp
@@ -28,10 +28,10 @@
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "src/core/gpu/cl/kernels/ClSoftmaxKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClPermute.h"
-#include "src/runtime/gpu/cl/operators/ClSoftmax.h"
+#include "src/gpu/cl/kernels/ClSoftmaxKernel.h"
+#include "src/gpu/cl/operators/ClPermute.h"
+#include "src/gpu/cl/operators/ClSoftmax.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLTranspose.cpp b/src/runtime/CL/functions/CLTranspose.cpp
index 142cf73259..e63c92eeb4 100644
--- a/src/runtime/CL/functions/CLTranspose.cpp
+++ b/src/runtime/CL/functions/CLTranspose.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/CL/ICLKernel.h"
-#include "src/runtime/gpu/cl/operators/ClTranspose.h"
+#include "src/gpu/cl/operators/ClTranspose.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp b/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
index fa01c914c5..b416d0fcf1 100644
--- a/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/KernelDescriptors.h"
 #include "src/core/CL/ICLKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/gpu/cl/operators/ClWinogradConv2d.h"
+#include "src/gpu/cl/operators/ClWinogradConv2d.h"
 #include "support/Cast.h"
 
 namespace arm_compute
diff --git a/src/runtime/CL/gemm/CLGEMMDefaultTypeBifrost.cpp b/src/runtime/CL/gemm/CLGEMMDefaultTypeBifrost.cpp
index 390bb97665..67253c7277 100644
--- a/src/runtime/CL/gemm/CLGEMMDefaultTypeBifrost.cpp
+++ b/src/runtime/CL/gemm/CLGEMMDefaultTypeBifrost.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/src/runtime/CL/gemm/CLGEMMDefaultTypeMidgard.cpp b/src/runtime/CL/gemm/CLGEMMDefaultTypeMidgard.cpp
index b799de6967..a64de9952e 100644
--- a/src/runtime/CL/gemm/CLGEMMDefaultTypeMidgard.cpp
+++ b/src/runtime/CL/gemm/CLGEMMDefaultTypeMidgard.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/GPUTarget.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/src/runtime/CL/gemm/CLGEMMDefaultTypeValhall.cpp b/src/runtime/CL/gemm/CLGEMMDefaultTypeValhall.cpp
index 982748810d..b3403b2aaf 100644
--- a/src/runtime/CL/gemm/CLGEMMDefaultTypeValhall.cpp
+++ b/src/runtime/CL/gemm/CLGEMMDefaultTypeValhall.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.cpp b/src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.cpp
index b8437487f8..b06c3b0f8e 100644
--- a/src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.cpp
+++ b/src/runtime/CL/gemm_auto_heuristics/CLGEMMAutoHeuristics.cpp
@@ -27,11 +27,11 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/ICLGEMMKernelSelection.h"
-#include "src/core/gpu/cl/kernels/gemm/ClGemmHelpers.h"
-#include "src/core/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h"
-#include "src/core/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/ClGemmHelpers.h"
+#include "src/gpu/cl/kernels/gemm/IClGemmKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/native/ClGemmNativeKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/reshaped/ClGemmReshapedKernelConfig.h"
+#include "src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmReshapedOnlyRhsKernelConfig.h"
 #include "src/runtime/CL/gemm/CLGEMMKernelSelection.h"
 #include "src/runtime/CL/mlgo/MLGOHeuristics.h"
 #include "src/runtime/CL/mlgo/Utils.h"
diff --git a/src/runtime/NEON/functions/NEActivationLayer.cpp b/src/runtime/NEON/functions/NEActivationLayer.cpp
index 2b5c51fa5a..e48aede590 100644
--- a/src/runtime/NEON/functions/NEActivationLayer.cpp
+++ b/src/runtime/NEON/functions/NEActivationLayer.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuActivation.h"
+#include "src/cpu/operators/CpuActivation.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEArithmeticAddition.cpp b/src/runtime/NEON/functions/NEArithmeticAddition.cpp
index 2e4755b949..a7581ca9f4 100644
--- a/src/runtime/NEON/functions/NEArithmeticAddition.cpp
+++ b/src/runtime/NEON/functions/NEArithmeticAddition.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuAdd.h"
+#include "src/cpu/operators/CpuAdd.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEArithmeticSubtraction.cpp b/src/runtime/NEON/functions/NEArithmeticSubtraction.cpp
index 0263d4cbb6..6fdd4267bf 100644
--- a/src/runtime/NEON/functions/NEArithmeticSubtraction.cpp
+++ b/src/runtime/NEON/functions/NEArithmeticSubtraction.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h"
 
 #include "arm_compute/core/ITensor.h"
-#include "src/runtime/cpu/operators/CpuSub.h"
+#include "src/cpu/operators/CpuSub.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NECast.cpp b/src/runtime/NEON/functions/NECast.cpp
index b519576ad5..a39e639ad4 100644
--- a/src/runtime/NEON/functions/NECast.cpp
+++ b/src/runtime/NEON/functions/NECast.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NECast.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuCast.h"
+#include "src/cpu/operators/CpuCast.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEConcatenateLayer.cpp b/src/runtime/NEON/functions/NEConcatenateLayer.cpp
index dcc5cd3a64..ceb697aad6 100644
--- a/src/runtime/NEON/functions/NEConcatenateLayer.cpp
+++ b/src/runtime/NEON/functions/NEConcatenateLayer.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEConcatenateLayer.h"
 
-#include "src/runtime/cpu/operators/CpuConcatenate.h"
+#include "src/cpu/operators/CpuConcatenate.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
diff --git a/src/runtime/NEON/functions/NEConvertFullyConnectedWeights.cpp b/src/runtime/NEON/functions/NEConvertFullyConnectedWeights.cpp
index 1f6b3c94e2..535ac99001 100644
--- a/src/runtime/NEON/functions/NEConvertFullyConnectedWeights.cpp
+++ b/src/runtime/NEON/functions/NEConvertFullyConnectedWeights.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuConvertFullyConnectedWeights.h"
+#include "src/cpu/operators/CpuConvertFullyConnectedWeights.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEConvolutionLayer.cpp b/src/runtime/NEON/functions/NEConvolutionLayer.cpp
index 8bd1119a69..ca62a40cc8 100644
--- a/src/runtime/NEON/functions/NEConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEConvolutionLayer.cpp
@@ -28,11 +28,11 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuConv2d.h"
-#include "src/runtime/cpu/operators/CpuDirectConv2d.h"
-#include "src/runtime/cpu/operators/CpuGemmConv2d.h"
-#include "src/runtime/cpu/operators/CpuGemmDirectConv2d.h"
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/operators/CpuConv2d.h"
+#include "src/cpu/operators/CpuDirectConv2d.h"
+#include "src/cpu/operators/CpuGemmConv2d.h"
+#include "src/cpu/operators/CpuGemmDirectConv2d.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NECopy.cpp b/src/runtime/NEON/functions/NECopy.cpp
index 20642b5eed..c2059e8e98 100644
--- a/src/runtime/NEON/functions/NECopy.cpp
+++ b/src/runtime/NEON/functions/NECopy.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NECopy.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuCopy.h"
+#include "src/cpu/operators/CpuCopy.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEDepthConvertLayer.cpp b/src/runtime/NEON/functions/NEDepthConvertLayer.cpp
index 07e985c25e..1ec32074a5 100644
--- a/src/runtime/NEON/functions/NEDepthConvertLayer.cpp
+++ b/src/runtime/NEON/functions/NEDepthConvertLayer.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEDepthConvertLayer.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuCast.h"
+#include "src/cpu/operators/CpuCast.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEDepthwiseConvolutionLayer.cpp b/src/runtime/NEON/functions/NEDepthwiseConvolutionLayer.cpp
index daa5fd5ab9..ed6dec3850 100644
--- a/src/runtime/NEON/functions/NEDepthwiseConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEDepthwiseConvolutionLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/runtime/cpu/operators/CpuDepthwiseConv2d.h"
+#include "src/cpu/operators/CpuDepthwiseConv2d.h"
 
 using namespace arm_compute::misc;
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/runtime/NEON/functions/NEDequantizationLayer.cpp b/src/runtime/NEON/functions/NEDequantizationLayer.cpp
index 91e37594af..83e0131c83 100644
--- a/src/runtime/NEON/functions/NEDequantizationLayer.cpp
+++ b/src/runtime/NEON/functions/NEDequantizationLayer.cpp
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/Tensor.h"
-#include "src/runtime/cpu/operators/CpuDequantize.h"
+#include "src/cpu/operators/CpuDequantize.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEDirectConvolutionLayer.cpp b/src/runtime/NEON/functions/NEDirectConvolutionLayer.cpp
index 58530e4a8f..ef3d3d6055 100644
--- a/src/runtime/NEON/functions/NEDirectConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEDirectConvolutionLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "src/runtime/cpu/operators/CpuDirectConv2d.h"
+#include "src/cpu/operators/CpuDirectConv2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEElementwiseOperations.cpp b/src/runtime/NEON/functions/NEElementwiseOperations.cpp
index 946bbb24b8..c958adf97c 100644
--- a/src/runtime/NEON/functions/NEElementwiseOperations.cpp
+++ b/src/runtime/NEON/functions/NEElementwiseOperations.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEElementwiseOperations.h"
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuElementwise.h"
+#include "src/cpu/operators/CpuElementwise.h"
 
 #include "arm_compute/core/ITensor.h"
 
diff --git a/src/runtime/NEON/functions/NEElementwiseUnaryLayer.cpp b/src/runtime/NEON/functions/NEElementwiseUnaryLayer.cpp
index 1a9e8839ca..a0674ec320 100644
--- a/src/runtime/NEON/functions/NEElementwiseUnaryLayer.cpp
+++ b/src/runtime/NEON/functions/NEElementwiseUnaryLayer.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/runtime/NEON/functions/NEElementwiseUnaryLayer.h"
-#include "src/runtime/cpu/operators/CpuElementwiseUnary.h"
+#include "src/cpu/operators/CpuElementwiseUnary.h"
 #include <utility>
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEFill.cpp b/src/runtime/NEON/functions/NEFill.cpp
index ee539fdfc8..43667783bf 100644
--- a/src/runtime/NEON/functions/NEFill.cpp
+++ b/src/runtime/NEON/functions/NEFill.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEFill.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuFill.h"
+#include "src/cpu/operators/CpuFill.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEFlattenLayer.cpp b/src/runtime/NEON/functions/NEFlattenLayer.cpp
index 4d1054ad25..f435842634 100644
--- a/src/runtime/NEON/functions/NEFlattenLayer.cpp
+++ b/src/runtime/NEON/functions/NEFlattenLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/helpers/AutoConfiguration.h"
-#include "src/runtime/cpu/operators/CpuFlatten.h"
+#include "src/cpu/operators/CpuFlatten.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEFloor.cpp b/src/runtime/NEON/functions/NEFloor.cpp
index f8a3c13d6d..d2dc48a159 100644
--- a/src/runtime/NEON/functions/NEFloor.cpp
+++ b/src/runtime/NEON/functions/NEFloor.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEFloor.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuFloor.h"
+#include "src/cpu/operators/CpuFloor.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp b/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
index cb7e2dc7ec..3f55a1f34e 100644
--- a/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
+++ b/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/runtime/MemoryGroup.h"
 #include "arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuFullyConnected.h"
+#include "src/cpu/operators/CpuFullyConnected.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEGEMM.cpp b/src/runtime/NEON/functions/NEGEMM.cpp
index b470afe1c6..58ade9fb3a 100644
--- a/src/runtime/NEON/functions/NEGEMM.cpp
+++ b/src/runtime/NEON/functions/NEGEMM.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/CPP/Validate.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
+#include "src/cpu/operators/CpuGemm.h"
 
 using namespace arm_compute::experimental;
 
diff --git a/src/runtime/NEON/functions/NEGEMMConv2d.cpp b/src/runtime/NEON/functions/NEGEMMConv2d.cpp
index 2230e80e4b..42b8b70405 100644
--- a/src/runtime/NEON/functions/NEGEMMConv2d.cpp
+++ b/src/runtime/NEON/functions/NEGEMMConv2d.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemmDirectConv2d.h"
+#include "src/cpu/operators/CpuGemmDirectConv2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEGEMMConvolutionLayer.cpp b/src/runtime/NEON/functions/NEGEMMConvolutionLayer.cpp
index 47ab16816a..c780d63763 100644
--- a/src/runtime/NEON/functions/NEGEMMConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEGEMMConvolutionLayer.cpp
@@ -28,7 +28,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemmConv2d.h"
+#include "src/cpu/operators/CpuGemmConv2d.h"
 
 using namespace arm_compute::experimental;
 
diff --git a/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp b/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
index b85530c70f..6c179f8387 100644
--- a/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
+++ b/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
@@ -31,7 +31,7 @@
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/helpers/MemoryHelpers.h"
 
-#include "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
+#include "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
 
 using namespace arm_compute::experimental;
 
diff --git a/src/runtime/NEON/functions/NEGEMMLowpOutputStage.cpp b/src/runtime/NEON/functions/NEGEMMLowpOutputStage.cpp
index 8351cc66d0..7e1de3c257 100644
--- a/src/runtime/NEON/functions/NEGEMMLowpOutputStage.cpp
+++ b/src/runtime/NEON/functions/NEGEMMLowpOutputStage.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuGemmLowpOutputStage.h"
+#include "src/cpu/operators/CpuGemmLowpOutputStage.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEPReluLayer.cpp b/src/runtime/NEON/functions/NEPReluLayer.cpp
index a05b545e9a..80c5690a4e 100644
--- a/src/runtime/NEON/functions/NEPReluLayer.cpp
+++ b/src/runtime/NEON/functions/NEPReluLayer.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEPReluLayer.h"
 
 #include "arm_compute/core/ITensor.h"
-#include "src/runtime/cpu/operators/CpuPRelu.h"
+#include "src/cpu/operators/CpuPRelu.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEPermute.cpp b/src/runtime/NEON/functions/NEPermute.cpp
index f707fad757..517b86a1cb 100644
--- a/src/runtime/NEON/functions/NEPermute.cpp
+++ b/src/runtime/NEON/functions/NEPermute.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEPermute.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuPermute.h"
+#include "src/cpu/operators/CpuPermute.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEPixelWiseMultiplication.cpp b/src/runtime/NEON/functions/NEPixelWiseMultiplication.cpp
index 3a2f1984b4..ad83a26beb 100644
--- a/src/runtime/NEON/functions/NEPixelWiseMultiplication.cpp
+++ b/src/runtime/NEON/functions/NEPixelWiseMultiplication.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h"
 
 #include "arm_compute/core/ITensor.h"
-#include "src/runtime/cpu/operators/CpuMul.h"
+#include "src/cpu/operators/CpuMul.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEPoolingLayer.cpp b/src/runtime/NEON/functions/NEPoolingLayer.cpp
index 8d267a32c0..5a3b9c5e7e 100644
--- a/src/runtime/NEON/functions/NEPoolingLayer.cpp
+++ b/src/runtime/NEON/functions/NEPoolingLayer.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuPool2d.h"
+#include "src/cpu/operators/CpuPool2d.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEQLSTMLayer.cpp b/src/runtime/NEON/functions/NEQLSTMLayer.cpp
index 946791a104..565c5531c4 100644
--- a/src/runtime/NEON/functions/NEQLSTMLayer.cpp
+++ b/src/runtime/NEON/functions/NEQLSTMLayer.cpp
@@ -32,8 +32,8 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h"
-#include "src/core/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
 #include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/CpuGemmLowpMatrixReductionKernel.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEQuantizationLayer.cpp b/src/runtime/NEON/functions/NEQuantizationLayer.cpp
index e607917615..dad246ac89 100644
--- a/src/runtime/NEON/functions/NEQuantizationLayer.cpp
+++ b/src/runtime/NEON/functions/NEQuantizationLayer.cpp
@@ -26,7 +26,7 @@
 
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/Tensor.h"
-#include "src/runtime/cpu/operators/CpuQuantize.h"
+#include "src/cpu/operators/CpuQuantize.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEReshapeLayer.cpp b/src/runtime/NEON/functions/NEReshapeLayer.cpp
index c0c78ea652..3ccb42361e 100644
--- a/src/runtime/NEON/functions/NEReshapeLayer.cpp
+++ b/src/runtime/NEON/functions/NEReshapeLayer.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NEReshapeLayer.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuReshape.h"
+#include "src/cpu/operators/CpuReshape.h"
 
 #include <utility>
 
diff --git a/src/runtime/NEON/functions/NEScale.cpp b/src/runtime/NEON/functions/NEScale.cpp
index 0fbad07d0f..b952858181 100644
--- a/src/runtime/NEON/functions/NEScale.cpp
+++ b/src/runtime/NEON/functions/NEScale.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "src/core/utils/ScaleUtils.h"
-#include "src/runtime/cpu/operators/CpuScale.h"
+#include "src/cpu/operators/CpuScale.h"
 #include "support/Rounding.h"
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/NESoftmaxLayer.cpp b/src/runtime/NEON/functions/NESoftmaxLayer.cpp
index bee692c08b..0947ff94a6 100644
--- a/src/runtime/NEON/functions/NESoftmaxLayer.cpp
+++ b/src/runtime/NEON/functions/NESoftmaxLayer.cpp
@@ -25,10 +25,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/MemoryGroup.h"
 #include "arm_compute/runtime/Tensor.h"
-#include "src/core/cpu/kernels/CpuSoftmaxKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/core/helpers/SoftmaxHelpers.h"
-#include "src/runtime/cpu/operators/CpuSoftmax.h"
+#include "src/cpu/kernels/CpuSoftmaxKernel.h"
+#include "src/cpu/operators/CpuSoftmax.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NETranspose.cpp b/src/runtime/NEON/functions/NETranspose.cpp
index 3b3023f3b3..b6bf15e428 100644
--- a/src/runtime/NEON/functions/NETranspose.cpp
+++ b/src/runtime/NEON/functions/NETranspose.cpp
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/functions/NETranspose.h"
 
 #include "arm_compute/core/Validate.h"
-#include "src/runtime/cpu/operators/CpuTranspose.h"
+#include "src/cpu/operators/CpuTranspose.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
index 98ff12590b..f0c153d4f4 100644
--- a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
@@ -29,9 +29,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/CPP/Validate.h"
-#include "src/core/cpu/kernels/CpuWinogradConv2dKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/kernels/CpuWinogradConv2dKernel.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
 
 #include "src/core/NEON/kernels/convolution/common/utils.hpp"
 #include "src/core/NEON/kernels/convolution/winograd/winograd.hpp"
diff --git a/tests/CL/Helper.h b/tests/CL/Helper.h
index e3e64c9405..dd5e8647b0 100644
--- a/tests/CL/Helper.h
+++ b/tests/CL/Helper.h
@@ -29,8 +29,8 @@
 #include "arm_compute/runtime/CL/functions/CLFill.h"
 #include "arm_compute/runtime/IFunction.h"
 #include "src/core/CL/kernels/CLFillBorderKernel.h"
-#include "src/runtime/gpu/cl/IClOperator.h"
-#include "src/runtime/gpu/cl/operators/ClFill.h"
+#include "src/gpu/cl/IClOperator.h"
+#include "src/gpu/cl/operators/ClFill.h"
 
 #include "src/core/CL/ICLKernel.h"
 #include "support/Cast.h"
diff --git a/tests/NEON/Helper.h b/tests/NEON/Helper.h
index c9e53d11b0..fb0231b62a 100644
--- a/tests/NEON/Helper.h
+++ b/tests/NEON/Helper.h
@@ -29,7 +29,7 @@
 #include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "src/core/NEON/kernels/NEFillBorderKernel.h"
-#include "src/runtime/cpu/ICpuOperator.h"
+#include "src/cpu/ICpuOperator.h"
 #include "tests/Globals.h"
 
 #include <algorithm>
diff --git a/tests/SConscript b/tests/SConscript
index 29ed7394d3..7b05dcbc5f 100644
--- a/tests/SConscript
+++ b/tests/SConscript
@@ -134,7 +134,7 @@ if env['neon']:
     filter_pattern = test_env['test_filter']
     files_benchmark += Glob('benchmark/NEON/*/' + filter_pattern)
     files_benchmark += Glob('benchmark/NEON/' + filter_pattern)
-    test_env.Append(CPPPATH = ["#/src/core/cpu/kernels/assembly/"])
+    test_env.Append(CPPPATH = ["#/src/cpu/kernels/assembly/"])
     if env['external_tests_dir']:
         files_benchmark += Glob(env['external_tests_dir'] + '/tests/benchmark/NEON/' + filter_pattern)
 
diff --git a/tests/validation/CL/Col2Im.cpp b/tests/validation/CL/Col2Im.cpp
index 96096d314d..4b004e2472 100644
--- a/tests/validation/CL/Col2Im.cpp
+++ b/tests/validation/CL/Col2Im.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "src/core/gpu/cl/kernels/ClCol2ImKernel.h"
+#include "src/gpu/cl/kernels/ClCol2ImKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/framework/Asserts.h"
diff --git a/tests/validation/CL/GEMMLowpMatrixMultiplyNative.cpp b/tests/validation/CL/GEMMLowpMatrixMultiplyNative.cpp
index d733a00296..d0d06a8ddb 100644
--- a/tests/validation/CL/GEMMLowpMatrixMultiplyNative.cpp
+++ b/tests/validation/CL/GEMMLowpMatrixMultiplyNative.cpp
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyNativeKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/framework/Asserts.h"
diff --git a/tests/validation/CL/GEMMLowpMatrixMultiplyReshaped.cpp b/tests/validation/CL/GEMMLowpMatrixMultiplyReshaped.cpp
index 3baa39bffc..88455bdeb8 100644
--- a/tests/validation/CL/GEMMLowpMatrixMultiplyReshaped.cpp
+++ b/tests/validation/CL/GEMMLowpMatrixMultiplyReshaped.cpp
@@ -23,9 +23,9 @@
  */
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/framework/Asserts.h"
diff --git a/tests/validation/CL/GEMMLowpMatrixMultiplyReshapedOnlyRHS.cpp b/tests/validation/CL/GEMMLowpMatrixMultiplyReshapedOnlyRHS.cpp
index 1283713c4d..c56901effc 100644
--- a/tests/validation/CL/GEMMLowpMatrixMultiplyReshapedOnlyRHS.cpp
+++ b/tests/validation/CL/GEMMLowpMatrixMultiplyReshapedOnlyRHS.cpp
@@ -25,8 +25,8 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmLowpMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMMatrixMultiply.cpp b/tests/validation/CL/GEMMMatrixMultiply.cpp
index 21e085087d..faa2413489 100644
--- a/tests/validation/CL/GEMMMatrixMultiply.cpp
+++ b/tests/validation/CL/GEMMMatrixMultiply.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMMatrixMultiplyInterleavedTransposed.cpp b/tests/validation/CL/GEMMMatrixMultiplyInterleavedTransposed.cpp
index e47518ad7d..9313ae34d6 100644
--- a/tests/validation/CL/GEMMMatrixMultiplyInterleavedTransposed.cpp
+++ b/tests/validation/CL/GEMMMatrixMultiplyInterleavedTransposed.cpp
@@ -26,9 +26,9 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMMatrixMultiplyNative.cpp b/tests/validation/CL/GEMMMatrixMultiplyNative.cpp
index a737c687c4..dc5fbc36ba 100644
--- a/tests/validation/CL/GEMMMatrixMultiplyNative.cpp
+++ b/tests/validation/CL/GEMMMatrixMultiplyNative.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMMatrixMultiplyReshaped.cpp b/tests/validation/CL/GEMMMatrixMultiplyReshaped.cpp
index 6f368a9650..fd12dea4fe 100644
--- a/tests/validation/CL/GEMMMatrixMultiplyReshaped.cpp
+++ b/tests/validation/CL/GEMMMatrixMultiplyReshaped.cpp
@@ -26,9 +26,9 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMMatrixMultiplyReshapedOnlyRHS.cpp b/tests/validation/CL/GEMMMatrixMultiplyReshapedOnlyRHS.cpp
index 88e99bcfef..0f86a70e0f 100644
--- a/tests/validation/CL/GEMMMatrixMultiplyReshapedOnlyRHS.cpp
+++ b/tests/validation/CL/GEMMMatrixMultiplyReshapedOnlyRHS.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedOnlyRhsKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMReshapeLHSMatrix.cpp b/tests/validation/CL/GEMMReshapeLHSMatrix.cpp
index f995608308..894b83701f 100644
--- a/tests/validation/CL/GEMMReshapeLHSMatrix.cpp
+++ b/tests/validation/CL/GEMMReshapeLHSMatrix.cpp
@@ -25,7 +25,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeLhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/GEMMReshapeRHSMatrix.cpp b/tests/validation/CL/GEMMReshapeRHSMatrix.cpp
index ff1240ea2e..f8462058a6 100644
--- a/tests/validation/CL/GEMMReshapeRHSMatrix.cpp
+++ b/tests/validation/CL/GEMMReshapeRHSMatrix.cpp
@@ -25,7 +25,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/CL/CLTensorAllocator.h"
-#include "src/core/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
+#include "src/gpu/cl/kernels/ClGemmReshapeRhsMatrixKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/CL/Im2Col.cpp b/tests/validation/CL/Im2Col.cpp
index 041f549777..1f5b781690 100644
--- a/tests/validation/CL/Im2Col.cpp
+++ b/tests/validation/CL/Im2Col.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "src/core/gpu/cl/kernels/ClIm2ColKernel.h"
+#include "src/gpu/cl/kernels/ClIm2ColKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/framework/Asserts.h"
diff --git a/tests/validation/CL/WeightsReshape.cpp b/tests/validation/CL/WeightsReshape.cpp
index 93be75df98..4345c4b08a 100644
--- a/tests/validation/CL/WeightsReshape.cpp
+++ b/tests/validation/CL/WeightsReshape.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h"
+#include "src/gpu/cl/kernels/ClWeightsReshapeKernel.h"
 #include "tests/CL/CLAccessor.h"
 #include "tests/CL/Helper.h"
 #include "tests/datasets/ShapeDatasets.h"
diff --git a/tests/validation/NEON/Col2Im.cpp b/tests/validation/NEON/Col2Im.cpp
index 3ae784bd04..7eb8cbf0f6 100644
--- a/tests/validation/NEON/Col2Im.cpp
+++ b/tests/validation/NEON/Col2Im.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "src/core/cpu/kernels/CpuCol2ImKernel.h"
+#include "src/cpu/kernels/CpuCol2ImKernel.h"
 #include "tests/NEON/Helper.h"
 #include "tests/datasets/ShapeDatasets.h"
 #include "tests/framework/Asserts.h"
diff --git a/tests/validation/NEON/ConvolutionLayer.cpp b/tests/validation/NEON/ConvolutionLayer.cpp
index 2178b9b209..578921bddd 100644
--- a/tests/validation/NEON/ConvolutionLayer.cpp
+++ b/tests/validation/NEON/ConvolutionLayer.cpp
@@ -29,9 +29,9 @@
 #include "arm_compute/runtime/Tensor.h"
 #include "arm_compute/runtime/TensorAllocator.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemmConv2d.h"
-#include "src/runtime/cpu/operators/CpuGemmDirectConv2d.h"
-#include "src/runtime/cpu/operators/CpuWinogradConv2d.h"
+#include "src/cpu/operators/CpuGemmConv2d.h"
+#include "src/cpu/operators/CpuGemmDirectConv2d.h"
+#include "src/cpu/operators/CpuWinogradConv2d.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/PaddingCalculator.h"
 #include "tests/datasets/LargeConvolutionLayerDataset.h"
diff --git a/tests/validation/NEON/DepthwiseConvolutionLayerNative.cpp b/tests/validation/NEON/DepthwiseConvolutionLayerNative.cpp
index ddf3faacb6..a79987dd4a 100644
--- a/tests/validation/NEON/DepthwiseConvolutionLayerNative.cpp
+++ b/tests/validation/NEON/DepthwiseConvolutionLayerNative.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "src/core/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
+#include "src/cpu/kernels/CpuDepthwiseConv2dNativeKernel.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/NEON/Helper.h"
 #include "tests/framework/Macros.h"
diff --git a/tests/validation/NEON/DilatedConvolutionLayer.cpp b/tests/validation/NEON/DilatedConvolutionLayer.cpp
index 2dc6231425..30bf690456 100644
--- a/tests/validation/NEON/DilatedConvolutionLayer.cpp
+++ b/tests/validation/NEON/DilatedConvolutionLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "arm_compute/runtime/TensorAllocator.h"
-#include "src/runtime/cpu/operators/CpuConv2d.h"
+#include "src/cpu/operators/CpuConv2d.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/PaddingCalculator.h"
 #include "tests/datasets/DilatedConvolutionLayerDataset.h"
diff --git a/tests/validation/NEON/FullyConnectedLayer.cpp b/tests/validation/NEON/FullyConnectedLayer.cpp
index 8ba0f1f771..413250f755 100644
--- a/tests/validation/NEON/FullyConnectedLayer.cpp
+++ b/tests/validation/NEON/FullyConnectedLayer.cpp
@@ -26,7 +26,7 @@
 #include "arm_compute/runtime/Tensor.h"
 #include "arm_compute/runtime/TensorAllocator.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuFullyConnected.h"
+#include "src/cpu/operators/CpuFullyConnected.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/PaddingCalculator.h"
 #include "tests/datasets/FullyConnectedLayerDataset.h"
diff --git a/tests/validation/NEON/GEMM.cpp b/tests/validation/NEON/GEMM.cpp
index 7b207cc3f2..079047328a 100644
--- a/tests/validation/NEON/GEMM.cpp
+++ b/tests/validation/NEON/GEMM.cpp
@@ -25,11 +25,11 @@
 #include "arm_compute/runtime/NEON/functions/NEGEMM.h"
 #include "arm_compute/runtime/Tensor.h"
 #include "arm_compute/runtime/TensorAllocator.h"
-#include "src/core/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
-#include "src/core/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
-#include "src/core/cpu/kernels/CpuGemmTranspose1xWKernel.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemm.h"
+#include "src/cpu/kernels/CpuGemmInterleave4x4Kernel.h"
+#include "src/cpu/kernels/CpuGemmMatrixMultiplyKernel.h"
+#include "src/cpu/kernels/CpuGemmTranspose1xWKernel.h"
+#include "src/cpu/operators/CpuGemm.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/NEON/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/NEON/GEMMLowp.cpp b/tests/validation/NEON/GEMMLowp.cpp
index 1941586d5c..7dd1a479fe 100644
--- a/tests/validation/NEON/GEMMLowp.cpp
+++ b/tests/validation/NEON/GEMMLowp.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/runtime/Tensor.h"
 #include "arm_compute/runtime/TensorAllocator.h"
 #include "src/core/helpers/MemoryHelpers.h"
-#include "src/runtime/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
+#include "src/cpu/operators/CpuGemmLowpMatrixMultiplyCore.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/NEON/Helper.h"
 #include "tests/PaddingCalculator.h"
diff --git a/tests/validation/NEON/Im2Col.cpp b/tests/validation/NEON/Im2Col.cpp
index f338675346..ef5e75c5db 100644
--- a/tests/validation/NEON/Im2Col.cpp
+++ b/tests/validation/NEON/Im2Col.cpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "src/core/cpu/kernels/CpuIm2ColKernel.h"
+#include "src/cpu/kernels/CpuIm2ColKernel.h"
 #include "tests/NEON/Accessor.h"
 #include "tests/NEON/Helper.h"
 #include "tests/datasets/ShapeDatasets.h"