[nnc] Support for non-1 batch in Conv2D and DepthwiseConv2D operations (#2902)
authorСергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 <s.barannikov@samsung.com>
Tue, 22 Jan 2019 13:02:14 +0000 (16:02 +0300)
committerРоман Михайлович Русяев/AI Tools Lab /SRR/Staff Engineer/삼성전자 <r.rusyaev@samsung.com>
Tue, 22 Jan 2019 13:02:14 +0000 (16:02 +0300)
commit4593a7296b99ff7ca6ece3a22828aff0bc7800b3
treeb49a60e32cf03808ab5a7b1566dde84b85d4c521
parentcb7771bb5d45b22f4478135837f88ec26fb21c9f
[nnc] Support for non-1 batch in Conv2D and DepthwiseConv2D operations (#2902)

* Add support for non-1 batch size in Conv2D and DepthwiseConv2D in interpreter backend.
* Fix a bug in ShapeRange class that caused undefined behavior when it is constructed from a temporary.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
contrib/nnc/include/core/modelIR/ShapeRange.h
contrib/nnc/passes/interpreter/ops/Conv2D.cpp
contrib/nnc/passes/interpreter/ops/DepthwiseConv2D.cpp
contrib/nnc/unittests/soft_backend/CPPOperations.cpp