diff options
| author | Sadik Armagan <sadik.armagan@arm.com> | 2020-05-19 14:10:30 +0100 |
|---|---|---|
| committer | Sadik Armagan <sadik.armagan@arm.com> | 2020-05-19 14:10:30 +0100 |
| commit | 813f23049d73177edfc1f1cff71147c39f4b695e (patch) | |
| tree | cef15224e1e7a4bd8117bdc9f57a71be84e4325d /ConversionUtils_1_2.hpp | |
| parent | abc95d04dfb2462ffb42bc1facde4f45ecc65319 (diff) | |
| download | android-nn-driver-813f23049d73177edfc1f1cff71147c39f4b695e.tar.gz | |
IVGCVSW-4453 Add Support for ANEURALNETWORKS_QLSTM to HAL 1.3 Driver
* Add QLSTM support for Android NN Driver
* Add overrideOutputInfo parameter to SetupAndTrackLayerOutputSlot
* Add optional condition to GetInputScalar
* Refactor Quantized 16 Bit LSTM impl
Change-Id: Ie8fa98ad5ee4a62174ef91ca80f1df62b7fde937
Signed-off-by: Keith Davis <keith.davis@arm.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Diffstat (limited to 'ConversionUtils_1_2.hpp')
| -rw-r--r-- | ConversionUtils_1_2.hpp | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/ConversionUtils_1_2.hpp b/ConversionUtils_1_2.hpp index c87cb030..29367f2f 100644 --- a/ConversionUtils_1_2.hpp +++ b/ConversionUtils_1_2.hpp @@ -1497,11 +1497,11 @@ bool ConvertQuantize(const HalOperation& operation, const HalModel& model, Conve template<typename HalPolicy, typename HalOperation = typename HalPolicy::Operation, typename HalModel = typename HalPolicy::Model> -bool ConvertQuantizedLstm(const HalOperation& operation, const HalModel& model, ConversionData& data) +bool ConvertQuantized16BitLstm(const HalOperation& operation, const HalModel& model, ConversionData& data) { using HalOperand = typename HalPolicy::Operand; - ALOGV("HalPolicy::ConvertQuantizedLstm()"); + ALOGV("HalPolicy::ConvertQuantized16BitLstm()"); //Inputs: // 0: The input: A 2-D tensor of type ANEURALNETWORKS_TENSOR_QUANT8_ASYMM and shape [numBatches, inputSize] |