Implement FP GPU depthwise convolution 1x1 kernel for in-place computation

* Implement in-place graph node mutator for 1x1 depthwise convolution

* Add in-place to validation fixture except for
  DepthwiseConvolutionLayerNativeValidationFixture as it would be a
  duplicate test otherwise (DepthwiseConvolutionLayerNative test tests
  the underlying kernel)

Resolves: COMPMID-4432

Change-Id: Id7f10f5ebdce7d49f550c0b62dbaaab7f5b59d29
Signed-off-by: SiCongLi <sicong.li@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5874
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>

1 files changed, 19 insertions, 0 deletions

diff --git a/tests/datasets/DepthwiseConvolutionLayerDataset.h b/tests/datasets/DepthwiseConvolutionLayerDataset.h
index a19e7ee8cf..53b5248374 100644
--- a/tests/datasets/DepthwiseConvolutionLayerDataset.h
+++ b/tests/datasets/DepthwiseConvolutionLayerDataset.h
@@ -262,6 +262,25 @@ public:
         add_config(TensorShape(9U, 9U, 32U), Size2D(5U, 5U), PadStrideInfo(2, 2, 4, 4, 4, 4, DimensionRoundingType::CEIL), Size2D(2U, 2U));
     }
 };
+
+/** Dataset containing in-place 1x1 depthwise convolution shapes.
+ *
+ * For a depthwise convolution op to be in-place:
+ * * Output has the same shape as the input;
+ *      * 1x1 filter
+ *      * stride == 1
+ *      * dilations == 1
+ *      * No paddings
+*/
+class SmallInPlaceDepthwiseConvolutionLayerDataset final : public DepthwiseConvolutionLayerDataset
+{
+public:
+    SmallInPlaceDepthwiseConvolutionLayerDataset()
+    {
+        add_config(TensorShape(7U, 7U, 1U), Size2D(1U, 1U), PadStrideInfo(1, 1, 0, 0));
+        add_config(TensorShape(11U, 13U, 16U), Size2D(1U, 1U), PadStrideInfo(1, 1, 0, 0));
+    }
+};
 } // namespace datasets
 } // namespace test
 } // namespace arm_compute