diff options
| author | Manuel Bottini <manuel.bottini@arm.com> | 2021-05-19 16:15:36 +0100 |
|---|---|---|
| committer | Manuel Bottini <manuel.bottini@arm.com> | 2021-05-26 13:59:44 +0000 |
| commit | f733e0375f62bbecbbf87045e4e40afad858b318 (patch) | |
| tree | 272873e8a5c0381f0ff6b1ead7892d65d7a7f354 /tests/validation/Helpers.cpp | |
| parent | d89e2faa60d148f3c04e57032a28f1065a1be0e8 (diff) | |
| download | ComputeLibrary-f733e0375f62bbecbbf87045e4e40afad858b318.tar.gz | |
DirectConvolutionLayer create image failure
Add implicit padding test on weights before the configure
Fix problem of considering left padding of weights when using cl image
Resolves: COMPMID-4493
Change-Id: I141d2de68e8bdfcbd6f18209db4f29fcc05305a1
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5689
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'tests/validation/Helpers.cpp')
| -rw-r--r-- | tests/validation/Helpers.cpp | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/tests/validation/Helpers.cpp b/tests/validation/Helpers.cpp index b53d46fc76..0f5d5c5101 100644 --- a/tests/validation/Helpers.cpp +++ b/tests/validation/Helpers.cpp @@ -325,7 +325,7 @@ std::pair<int, int> get_symm_quantized_per_channel_bounds(const QuantizationInfo return std::pair<int, int> { min_bound, max_bound }; } -void add_padding_x(std::initializer_list<ITensor *> tensors, const DataLayout &data_layout) +void add_padding_x(std::initializer_list<ITensor *> tensors, const DataLayout &data_layout, bool only_right_pad) { if(data_layout == DataLayout::NHWC) { @@ -342,7 +342,7 @@ void add_padding_x(std::initializer_list<ITensor *> tensors, const DataLayout &d std::mt19937 gen(library->seed() + seed_offset++); const unsigned int right = distribution(gen); - const unsigned int left = distribution(gen); + const unsigned int left = only_right_pad ? 0 : distribution(gen); tensor->info()->extend_padding(PaddingSize(0U, right, 0U, left)); } |