diff options
| author | thecha01 <theo.charalambous@arm.com> | 2020-09-02 16:18:25 +0100 |
|---|---|---|
| committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-09-04 15:54:27 +0000 |
| commit | be058f4b5fb2383e0fa4666a92c43ca994486c86 (patch) | |
| tree | 3748a2665d4254a44f3ac905e6935c2829095d2a /src/graph/backends/NEON/NEFunctionFactory.cpp | |
| parent | b3182b19251cd010baad8252e7607de7059ac986 (diff) | |
| download | ComputeLibrary-be058f4b5fb2383e0fa4666a92c43ca994486c86.tar.gz | |
Use shape broadcast for Mult inputs in EDSR graph
We no longer have to explicitly create a tensor with the correct
dimensions for the Const nodes, instead we use the graph API shape
propogation logic in EltwiseLayerNode to broadcast the shapes
Signed-off-by: thecha01 <theo.charalambous@arm.com>
Change-Id: Ifb62b572d6391850d3357cd3307cef7cd9645ee3
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3898
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/graph/backends/NEON/NEFunctionFactory.cpp')
0 files changed, 0 insertions, 0 deletions