- dda6914 Update documentation for OpenCL kernel padding removal by Sheri Zhang · 3 years, 7 months ago
- 8b555e3 Fix tolerance for NEON RNNLayer by Manuel Bottini · 3 years, 7 months ago
- cf87f50 Tweak scheduling use of SQDMULH in quantized AVG pooling by Michele Di Giorgio · 3 years, 7 months ago
- 89de118 Fix unused variable compilation error by Giorgio Arena · 3 years, 7 months ago
- 76cb751 Fix CL Compiler frontend failure for ArgMinMax by Giorgio Arena · 3 years, 7 months ago
- 0841ca0 Fix OpenCL direct convolution by Gian Marco Iodice · 3 years, 7 months ago
- 8d8a1c5 Fix segfault in fsrcnn.tflite in GpuAcc by Teresa Charlin · 3 years, 7 months ago
- e131466 Make CL Pooling kernels and functions state-less by Michele Di Giorgio · 3 years, 7 months ago
- be9f9f9 Add WBSM tuning to CLTuner by Manuel Bottini · 3 years, 7 months ago
- 7061eb2 Implement MLGO module by SiCong Li · 3 years, 8 months ago
- 74a142c Account for eventual padding in Pooling assembly kernels by Michele Di Giorgio · 3 years, 7 months ago
- ea85727 Compress OpenCL kernel files using zlib for Android by Georgios Pinitas · 3 years, 7 months ago
- 655e8c6 Make data_layout an attribute of the Scale function by Michele Di Giorgio · 3 years, 7 months ago
- cc438f2 Remove redundant header from NEElementwiseUnaryLayer by Sang-Hoon Park · 3 years, 7 months ago
- c9c8905 Make CL Elementwise Unary kernels and functions state-less by Michele Di Giorgio · 3 years, 7 months ago
- c727d52 Add broadcasting support in NEPixelWiseMultiplicationKernel for FP16 by Pablo Marquez Tello · 3 years, 7 months ago
- ef28340 Force early exit of threads when output_height=1 and channels=1 in assembly pooling by Michele Di Giorgio · 3 years, 7 months ago
- 237be03 Make Gpu Elementwise Operations kernels and functions state-less by Michele Di Giorgio · 3 years, 7 months ago
- 26cf798 Fix NEON/CL CustomConvolution failures by Manuel Bottini · 3 years, 7 months ago
- 4cfab18 Make CLArithmeticSubtraction kernel and function state-less by Michele Di Giorgio · 3 years, 7 months ago
- 1cf1a7f Remove iostream header file when not needed by Gian Marco Iodice · 3 years, 7 months ago
- 1e0208a Make CLArithmeticAddition kernel and function state-less by Michele Di Giorgio · 3 years, 7 months ago
- 7249f15 Rename functions/classes for elementwise unary operations by Sang-Hoon Park · 3 years, 7 months ago
- 6a4ebe1 CTS Failures in Android CpuAcc in Add by Teresa Charlin · 3 years, 7 months ago
- 63001ac Rename functions/classes for elementwise operations by Sang-Hoon Park · 3 years, 7 months ago
- 360f576 Fix Scale CTS failures by Sheri Zhang · 3 years, 7 months ago
- f9943c8 Check for unsupported armv7a optimized pooling kernels by Michele Di Giorgio · 3 years, 7 months ago
- 2a0939d Fixed segfault NEDeconvLayer due to overflow. by Pablo Marquez Tello · 3 years, 7 months ago
- 2344189 Update Fp16 heuristic for Mali-G76 and Mali-G77 by Gian Marco Iodice · 3 years, 7 months ago
- e36208c CTS failures in Android R and Q in GpuAcc in ArgMinMax by Giorgio Arena · 3 years, 7 months ago
- 0094c02 Add window parameter to scheduler by Sang-Hoon Park · 3 years, 7 months ago
- f47f718 Make CLFloor and CLActivation kernels and functions state-less by Georgios Pinitas · 3 years, 7 months ago
- ae182f2 Check for FP16 CPU support in optimized pooling kernels by Michele Di Giorgio · 3 years, 7 months ago
- 098e94e Fix softmax SVE compiling failure without SVE2 support by Sheri Zhang · 3 years, 7 months ago
- fc6744a Make Sub kernel and operator stateless by Sheri Zhang · 3 years, 7 months ago
- d556d7b Integrate improved pooling layer on NEON by Michele Di Giorgio · 3 years, 10 months ago
- 7d61ff0 Make all CL Concatenate kernels and functions state-less by Michele Di Giorgio · 3 years, 7 months ago
- da6a6eb Update documentation for SVE/SVE2 support by Sheri Zhang · 3 years, 8 months ago
- d95c3e8 Direct convolution fix for quantized data type by Gian Marco Iodice · 3 years, 7 months ago
- bd2c8e1 Fix doxygen references to new kernels by Michele Di Giorgio · 3 years, 7 months ago
- 0f7ef8a Make memset/copy functions state-less by Georgios Pinitas · 3 years, 7 months ago
- ff1fe3e Remove padding from direct convolution - OpenCL by Gian Marco Iodice · 3 years, 8 months ago
- 6124390 Make Add kernel and operator stateless by Sheri Zhang · 3 years, 7 months ago
- 5bf4412 LOGICAL_NOT fail by Teresa Charlin · 3 years, 7 months ago
- 61ba069 Make Concatenate kernels and operator stateless by Georgios Pinitas · 3 years, 7 months ago
- b5a450a Add SVE for Softmax by Michalis Spyrou · 3 years, 8 months ago
- d2447bb Decouple data types of elementwise kernels by Sang-Hoon Park · 3 years, 7 months ago
- 33e0307 Cycle estimate-based kernel selection for dot product quantized s8/u8 kernels by Georgios Pinitas · 3 years, 7 months ago
- 8d5337e Add explicit cast to fix gcc10 build error by Sang-Hoon Park · 3 years, 7 months ago
- 7229182 Add missing std headers by Sang-Hoon Park · 3 years, 7 months ago
- 6c25aad Introduce threshold for NEON floating-point SOFT_RELU activation by Michele Di Giorgio · 3 years, 7 months ago
- ddd79f5 Release winograd transformed weights if not needed on NEON by Georgios Pinitas · 3 years, 7 months ago
- 18fbb92 Fix build error for gcc 10.2+ by Sang-Hoon Park · 3 years, 7 months ago
- d05d56d [Nightly Failure] Fix DeconvolutionLayer OpenCL kernel compilation by Giorgio Arena · 3 years, 7 months ago
- 50e98bb Add support for Armv8-R AArch64 build by Sang-Hoon Park · 3 years, 7 months ago
- 0f3d597 Remove OpenCL padding CLTransposeKernel by Manuel Bottini · 3 years, 8 months ago
- e9aaacd Add missing INTEGER type in nn driver by Michalis Spyrou · 3 years, 7 months ago
- 4551403 Add support for macOS by Georgios Pinitas · 3 years, 8 months ago
- fefd16a Fix wrong file inclusion in QLSTMLayerNormalization reference by Georgios Pinitas · 3 years, 7 months ago
- 7e5a865 Add tolerance for quantized activations computed in float by Michele Di Giorgio · 3 years, 7 months ago
- d30405a Alter tolerance in QuantizationLayer to use int8_t for asymmetric signed by Georgios Pinitas · 3 years, 7 months ago
- 3d3a01c Remove padding for CLArgMinMaxLayerKernel and fix CLRange mismatches by Giorgio Arena · 3 years, 7 months ago
- d23a251 Add SVE2 kernels for quantized elementwise operations by Sang-Hoon Park · 3 years, 7 months ago
- 8bcb65f [Nightly Failure] Fix CLDepthwiseConvolutionLayer 3x3 QASYMM8 on Midgard by Giorgio Arena · 3 years, 7 months ago
- e78ef6f Remove OpenCL padding CLFloorKernel by Manuel Bottini · 3 years, 8 months ago
- 5db75c3 Add SVE support for non-quantized elementwise operations by Sang-Hoon Park · 3 years, 8 months ago
- 23adc4c Add SVE support and decouple data type for NEScaleKernel by Sheri Zhang · 3 years, 8 months ago
- b14a0f0 Add meta-data to express dynamic shapes in ITensorInfo by Georgios Pinitas · 3 years, 8 months ago
- f8f0442 Make CpuActivation stateless by Georgios Pinitas · 3 years, 8 months ago
- 4f77ba9 Remove OpenCL padding: CLROIAlignLayerKernel by Manuel Bottini · 3 years, 8 months ago
- 4f8ce13 [Nightly Failure] Fix NEIm2Col argument validation errors by Giorgio Arena · 3 years, 8 months ago
- 70eb53b Make CpuFloorKernel kernel stateless by Georgios Pinitas · 3 years, 8 months ago
- c216545 [Nightly Failure] Fix OpenCL kernel compilation for CLRange by Giorgio Arena · 3 years, 8 months ago
- 33b103b Resolve merge conflicts between new fills and new fp16 generator by Giorgio Arena · 3 years, 8 months ago
- b56c175 Generalization of CLTuner by Manuel Bottini · 3 years, 9 months ago
- 805145d Add an SVE implementation of the Leaky ReLU activation function for qasymm8 and signed qasymm8 data. by arngra01 · 3 years, 8 months ago
- 4bdd177 Fix fill() for FP data type in fixtures - Part 2 by Giorgio Arena · 3 years, 8 months ago
- 4d9383e [Nightly Failure] Fix ColorConvert empty coordinates error by Giorgio Arena · 3 years, 8 months ago
- af1870b Add SVE support to elementwise unary kernels by Sang-Hoon Park · 3 years, 9 months ago
- 7e5b7bf Review all shapes in datasets to account for padding removal Part 3 by SiCong Li · 3 years, 9 months ago
- a8e2aeb Generalize custom uniform generator for floating point data types with 16 bits by Giorgio Arena · 3 years, 8 months ago
- c5a6139 Clean up macro definitions in arm_compute headers by Giorgio Arena · 3 years, 8 months ago
- ee82d34 Fix Doxygen references by Michele Di Giorgio · 3 years, 8 months ago
- a3c9a3b COMPMID-3874: Create ArithmeticAddition SVE/SVE2 by Michalis Spyrou · 3 years, 9 months ago
- b309fc2 Fix uniform_real_distribution_fp16 warning and single Dimension increment by Giorgio Arena · 3 years, 8 months ago
- 97b3f11 COMPMID-4076: ArmNN unittest failure with memory access voilation in FuseReLUIntoBatchNormFloat32CpuAccTest by Sheri Zhang · 3 years, 8 months ago
- dff2935 Improve NEIm2Col validation for invalid shapes by Georgios Pinitas · 3 years, 8 months ago
- 9324022 Remove OpenCL padding: CLPadLayerKernel by Giorgio Arena · 3 years, 8 months ago
- 0870db4 Add utility functions for SVE by Sang-Hoon Park · 3 years, 9 months ago
- 8a8f4fa Fix FFT FP16 failures for some OpenCL compilers by Giorgio Arena · 3 years, 8 months ago
- b49988c Remove unused tolerance value by Michalis Spyrou · 3 years, 8 months ago
- 8d5d78b COMPMID-3871: Create BatchNormalization SVE/SVE2 by Sheri Zhang · 3 years, 8 months ago
- 410e21e Fix baremetal arm_compute_validation build errors by SiCongLi · 3 years, 8 months ago
- 2dcffde Fix Select errors at OpenCL kernel compile time by Giorgio Arena · 3 years, 8 months ago
- 82c0d7f Create fill functions available everywhere for easy debugging by Giorgio Arena · 3 years, 8 months ago
- 00e86cd Add new shapes in WinogradOutputTransform dataset by Giorgio Arena · 3 years, 8 months ago
- 0c58f99 Remove OpenCL padding CLScaleKernel by Manuel Bottini · 3 years, 9 months ago
- 2567adf Add new shapes to WinogradInputTransform dataset and fix border size for NCHW data layout by Giorgio Arena · 3 years, 8 months ago
- b6869dd Adding no padding check asserts to specific CL Kernels by Manuel Bottini · 3 years, 8 months ago
- 6aeb217 Fix fill() for FP data type in fixtures - Part 1 by Giorgio Arena · 3 years, 8 months ago