- 3144159 IVGCVSW-7214 Disable BF16-Turbo-Mode and remove conversion layers by Ryan OShea · 1 year, 10 months ago
- c013bc8 Fix some compile warnings & errors by Rob Hughes · 3 years, 1 month ago
- 53ef795 IVGCVSW-5826 Change weights layout for depthwise to [1,H,W,I*M] by Jan Eilers · 3 years, 3 months ago
- aba90cd IVGCVSW-5091 Add Logical ops frontend and ref impl by James Conroy · 3 years, 10 months ago
- ea8ce70 IVGCVSW-5325 Fix non-channel per axis quantization by Finn Williams · 3 years, 11 months ago
- b9dcfe6 IVGCVSW-5325 Speed up the reference backend by Finn Williams · 4 years ago
- cbd2c23 IVGCVSW-5007 Implement an Int32 reference Elementwise workload by Finn Williams · 4 years, 2 months ago
- ac2770a IVGCVSW-4485 Remove Boost assert by Narumol Prangnawarat · 4 years, 5 months ago
- b60dd24 IVGCVSW-4565 TENSOR_BOOL8 data type not supported in AndroidNN Driver by Sadik Armagan · 4 years, 5 months ago
- 8eb2560 IVGCVSW-4482 Remove boost::ignore_unused by Jan Eilers · 4 years, 6 months ago
- 8832522 IVGCVSW-4517 Implement BFloat16 Encoder and Decoder by Narumol Prangnawarat · 4 years, 6 months ago
- 9add120 IVGCVSW-4386 Add ArmNN reference support for QAsymmS8 by Ryan OShea · 4 years, 7 months ago
- 246bd46 Remove inclusion of ArmNN.hpp where it is unnecessary. by Matthew Bentham · 4 years, 7 months ago
- fd27106 IVGCVSW-4211 Add Signed 8 bit Quantisation support into the Reference backend by Finn Williams · 4 years, 9 months ago
- e011d20 IVGCVSW-4209 Create a public API for the ArmNN Utils by Matteo Martincigh · 4 years, 9 months ago
- 5edc881 IVGCVSW-3837 Add support for per-axis quantization to reference Convolution2d workload by Aron Virginas-Tar · 4 years, 10 months ago
- b67f957 IVGCVSW-3836 Add support for Int32 per-axis scales by Aron Virginas-Tar · 4 years, 10 months ago
- 5236e1d IVGCVSW-3835 Create Encoder and Decoder for QSymm8PerAxis by Keith Davis · 4 years, 10 months ago
- e69c399 IVGCVSW-3824 Implement Float 16 Encoder and Decoder by Matthew Jackson · 5 years ago
- 198ee40 IVGCVSW-3609 Fix decoding and encoding of INT32 tensors by Aron Virginas-Tar · 5 years ago
- c394a6d IVGCVSW-3307 Don't assume TensorInfo::Map() can be called before Execute() by Matthew Bentham · 5 years ago
- 43aec58 IVGCVSW-3134 Refactor FullyConnected workloads into single workload by Francis Murtagh · 5 years ago
- 9b39832 IVGCVSW-3025: Refactor reference Convolution2d workload by Mike Kelly · 5 years ago
- eb2b329 IVGCVSW-2997 Refactor reference LSTM workload by Nattapat Chaimanowong · 5 years ago
- d4f0fea IVGCVSW-2947 Remove boost dependency from include/TypesUtils.hpp by Aron Virginas-Tar · 5 years ago
- f30f7d3 IVGCVSW-2946 RefElementwiseWorkload configures prior to first execute by Derek Lamberti · 5 years ago
- 2999a02 IVGCVSW-2862 Extend the Elementwise Workload to support QSymm16 Data Type by Sadik Armagan · 5 years ago
- 2e6dc3a IVGCVSW-2861 Refactor the Reference Elementwise workload by Sadik Armagan · 5 years ago