COMPMID-3316: NEDeconvolutionLayer failing for a big input

- Using NEDirectConvolution for big shapes since the memory
  required explodes for 9x9 kernel
- Adding test cases
- Fix enables only the NEON Deconvolution for NHWC

Change-Id: I8a541346428e5686818f8ecb7f69e2a9106cbceb
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3135
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com>

1 files changed, 0 insertions, 7 deletions

diff --git a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h
index c4c1664f20..e2ed0e0abc 100644
--- a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h
+++ b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h
@@ -120,16 +120,9 @@ private:
     NEConvolutionLayer _conv_f;
     CPPUpsample        _upsample_f;
     NEReverse          _flip_weights;
-    NEPermute          _permute_input;
-    NEPermute          _permute_weights;
-    NEPermute          _permute_output;
     Tensor             _scaled_output;
     Tensor             _weights_flipped;
-    Tensor             _permuted_input;
-    Tensor             _permuted_weights;
-    Tensor             _permuted_output;
     Tensor             _flip_axis;
-    bool               _is_nchw;
     const ITensor     *_original_weights;
     ITensor           *_input;
     PadStrideInfo      _info;