commit | 00bf68eb515038d63e66d29f4e2cb7c1cd791539 | [log] [tgz] |
---|---|---|
author | Frank Barchard <fbarchard@google.com> | Sun Oct 27 03:00:09 2019 -0700 |
committer | XNNPACK Team <xnnpack-github-robot@google.com> | Sun Oct 27 03:00:32 2019 -0700 |
tree | c5c1d054d5427b9d57de90df59a299d554021f12 | |
parent | c452eb13f60169bbb4f60b7b56cfb42c695e9adf [diff] |
A53 6x8 GEMM unrolled Unroll main loop to 48 FMA and use 2 sets of registers Mobile net V2 on P20l (a53) Was sgemm_6x8__aarch64_neonfma_cortex_a53 54095647 21 Now sgemm_6x8__aarch64_neonfma_cortex_a53 52253291 21 sgemm_4x12__aarch64_neonfma_cortex_a53 48738926 21 A55r1 Was sgemm_6x8__aarch64_neonfma_cortex_a53 52324249 21 Now sgemm_6x8__aarch64_neonfma_cortex_a53 50994408 21 sgemm_4x12__aarch64_neonfma_cortex_a53 47558932 21 ruy_st 54597647 21 sgemm_4x8__aarch64_neonfma_cortex_a53 57665554 21 PiperOrigin-RevId: 276926494
XNNPACK is a highly optimized library of floating-point neural network inference operators for ARM, WebAssembly, and x86 (SSE2 level) platforms. XNNPACK is not intended for direct use by deep learning practitioners researchers; instead it provides low-level performance primitives for accelerating high-level machine learning frameworks, such as MediaPipe, TensorFlow Lite, and TensorFlow.js.
XNNPACK implements the following neural network operators:
All operators in XNNPACK support NHWC layout, but additionally allow custom stride along the Channel dimension. Thus, operators can consume a subset of channels in the input tensor, and produce a subset of channels in the output tensor, providing a zero-cost Channel Split and Channel Concatenation operations.
The table below presents single-threaded performance of XNNPACK library on two generations of MobileNet models and three generations of Pixel phones.
Model | Pixel, ms | Pixel 2, ms | Pixel 3a, ms |
---|---|---|---|
MobileNet v1 1.0X | 81 | 93 | 88 |
MobileNet v2 1.0X | 48 | 58 | 54 |
Benchmarked on October 9, 2019 with end2end_bench --benchmark_min_time=5
on an Android/ARM64 build (bazel build -c opt --config android_arm64 :end2end_bench
) and neural network models with randomized weights and inputs.
XNNPACK is a based on QNNPACK library. However, unlike QNNPACK, XNNPACK focuses entirely on floating-point operators, and its API is no longer compatible with QNNPACK.