diff options
| author | Teresa Charlin <teresa.charlinreyes@arm.com> | 2022-05-15 14:07:05 +0100 |
|---|---|---|
| committer | Ryan OShea <ryan.oshea3@arm.com> | 2022-05-19 11:06:53 +0100 |
| commit | 5841c740ba6bc6c8c3e96d24156dc47907af6430 (patch) | |
| tree | fef7e7068a613e34491a7c846dda9b19b61e3a8f /src/armnn/Network.cpp | |
| parent | 21fe06fad6760a0d453f2de9c8dd790983ae940c (diff) | |
| download | armnn-5841c740ba6bc6c8c3e96d24156dc47907af6430.tar.gz | |
IVGCVSW-6455 Support Const + Dequantize layer and optimize it.
* Support Float16 as input to Dequantize layer
* Add Optimization to substitute Const+Dequantize layers with Const layer
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I58bb7e3871ca480c7b6fca93c4efb2de84e09e64
Signed-off-by: David <david.monahan@arm.com>
Diffstat (limited to 'src/armnn/Network.cpp')
| -rw-r--r-- | src/armnn/Network.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/src/armnn/Network.cpp b/src/armnn/Network.cpp index 6a646d3cc8..9da28ceeea 100644 --- a/src/armnn/Network.cpp +++ b/src/armnn/Network.cpp @@ -1739,6 +1739,7 @@ IOptimizedNetworkPtr Optimize(const Graph& inGraph, FuseBatchNormIntoConvolution2DFloat16(), FuseBatchNormIntoDepthwiseConvolution2DFloat32(), FuseBatchNormIntoDepthwiseConvolution2DFloat16(), + ConvertConstDequantisationLayersToConstLayers(), RedirectMembersToConstantInputs())); // If Fp32 to Fp16 optimization is set convert Fp32 network to Fp16 |