IVGCVSW-5842 Remove cross-wiring in depthwise
* Reading tensor infos won't allow a permutation vector anymore.
The permutation only changed the quantization dimension not the
shape and was therefore misleading
* The permutation of the full tensor info is now performed in
armnnUtils::Permuted
* Changed TfLite Parser depthwise parsing function
* Added unit tests to TfLite Parser with more random data
* Changed TfLite Delegate depthwise parsing function
* Added unit test to the delegate with per channel quantization
!android-nn-driver:5412
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I1f985ee69547bcaf16a72201e00a6b6fe1ef9a97
diff --git a/delegate/src/Convolution.hpp b/delegate/src/Convolution.hpp
index 153f449..6566fff 100644
--- a/delegate/src/Convolution.hpp
+++ b/delegate/src/Convolution.hpp
@@ -291,7 +291,7 @@
// Mappings from TensorflowLite filter tensors to the ArmNN filter tensors (ArmNN weights have to be [M, I, H, W])
armnn::PermutationVector permutationVector{ 2, 3, 1, 0 }; // [H, W, I, M] -> [M, I, H, W]
- armnn::TensorInfo filterTensorInfo = GetTensorInfoForTfLiteTensor(tfLiteFilterTensor, permutationVector);
+ armnn::TensorInfo filterTensorInfo = GetTensorInfoForTfLiteTensor(tfLiteFilterTensor);
// Assuming input is NHWC
unsigned int inputHeight = inputTensorInfo.GetShape()[1];