aboutsummaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
2021-07-27Update GEMM assembly performance parametersGeorgios Pinitas
2021-07-26Add missing limits includeFreddie Liardet
2021-07-26Fix allocation of prepare tensor on ClWinogradConv2dGeorgios Pinitas
2021-07-25Reorganize the kernels into nhwc, nchw and common foldersAdnan AlSinan
2021-07-23Avoid allocation of auxiliary memory in CpuGemmConvolutionGeorgios Pinitas
2021-07-23Fix vector_length identification mechanism for SVEGeorgios Pinitas
2021-07-23Port NEFullyConnectedLayer to memory injecting interfaceMichele Di Giorgio
2021-07-23Pass fast math flag for correct GEMM3D validation supportGeorgios Pinitas
2021-07-23Fix formatting issue with blank profiler headerFreddie Liardet
2021-07-23Fix NN driver build script for Android RMichele Di Giorgio
2021-07-23Fix bare metal build errorFreddie Liardet
2021-07-22Remove printing Android version detected by the build systemPablo Marquez Tello
2021-07-22Expose fast_math mode for GEMM through BFloat16Georgios Pinitas
2021-07-22Inject temporary tensors to pack in they don't exist in CpuSoftmaxGeorgios Pinitas
2021-07-22Print the Android version detected by the build systemPablo Marquez Tello
2021-07-22Port ClGemmLowp to new APIGeorgios Pinitas
2021-07-22Fix oclgrind int overflow warningFreddie Liardet
2021-07-22Update GEMM assembly kernelsGeorgios Pinitas
2021-07-19Add layer data to JSON outputFreddie Liardet
2021-07-16Update OpenCL Header file to version v.2020.03.13Sheri Zhang
2021-07-16Include limits header to fix errors reported by GCC11Michele Di Giorgio
2021-07-16Avoid multiple Rhs matrix transformation on ClGemmGeorgios Pinitas
2021-07-16Disabled armv8-2a on Android PPablo Marquez Tello
2021-07-16Port CLIm2ColKernel to ClIm2ColKernelManuel Bottini
2021-07-15Disabled DirectConv method for NCHW and kernel > 15Pablo Marquez Tello
2021-07-15Port NEGEMMConvolutionLayerManuel Bottini
2021-07-15Improve filelist for GPUGeorgios Pinitas
2021-07-15Port CLCol2ImKernel to ClCol2ImKernelManuel Bottini
2021-07-14Increase tolerance for NEMeanStdDevNormalizationGeorgios Pinitas
2021-07-14Fix CL kernel compilation failureMichalis Spyrou
2021-07-13Add in-place calculation support for CL elementwise arithmetic kernelsSheri Zhang
2021-07-13Override arch option to alow building for armv8-2aPablo Marquez Tello
2021-07-13Port NEWinogradConvolutionLayerMichalis Spyrou
2021-07-09Limit the LOOP_UNROLLING on the kernel heightGian Marco Iodice
2021-07-09Port NECol2ImKernelManuel Bottini
2021-07-08Remove redundant implementations of Add/Sub operatorsGeorgios Pinitas
2021-07-08Change CLConvolution selection to run Direct approach on large kernelsGeorgios Pinitas
2021-07-08Add LayerData to all nodesFreddie Liardet
2021-07-08Port NEGEMMLowp Part 2Manuel Bottini
2021-07-07Add basic Operator interfaceGeorgios Pinitas
2021-07-07Validate unsupported data types with runtime informationGeorgios Pinitas
2021-07-06Fix manual LOOP_UNROLLINGGian Marco Iodice
2021-07-06Port NEIm2ColKernelManuel Bottini
2021-07-06Rework README.md structureGian Marco Iodice
2021-07-05Improve implementation selection speed of CpuElementwiseUnaryGeorgios Pinitas
2021-07-02Rework OpenCL Depthwise ConvolutionGian Marco Iodice
2021-07-02Align kernel/operator header layoutGeorgios Pinitas
2021-07-02Port NEGEMM to memory injecting interface (Part 3)Michele Di Giorgio
2021-07-02Implement FP GPU depthwise convolution 1x1 kernel for in-place computationSiCongLi
2021-07-01Add quantization helper functions for OpenCLGeorgios Pinitas
generated by cgit v1.2.1 (git 2.23.0) at 2024-06-17 15:24:02 +0000