MLCE-753 Expand Tensorshape for relevent layers before verifying support
Previously we were adding a reshape layer to "broadcast" tensors
for elementwise operations. This broadcast was happening too late
and was really just an expand dims. This was breaking the constant
attributes of tensors and layer support of certain backends.
* Remove addition of reshape layer when expanding dimensions
* Replace broadcast function with expand dims to equal rank function
* Fix some error status checks in various layers
* Add new TensorUtil function that expands dims to a defined rank
* Add unit tests to new TensorUtil function
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I31aca47c98075fef4f86864a15470f5faa55ab8d
diff --git a/include/armnnUtils/TensorUtils.hpp b/include/armnnUtils/TensorUtils.hpp
index a2aa9b0..7bf41c1 100644
--- a/include/armnnUtils/TensorUtils.hpp
+++ b/include/armnnUtils/TensorUtils.hpp
@@ -48,6 +48,8 @@
armnn::TensorShape ExpandDims(const armnn::TensorShape& tensorShape, int axis);
+armnn::TensorShape ExpandDimsToRank(const armnn::TensorShape& tensorShape, unsigned int rank);
+
std::vector<unsigned int> SqueezeDims(const armnn::TensorShape& tensorShape);
unsigned int GetNumElementsBetween(const armnn::TensorShape& shape,