commit | b177732f75a0b095f6a5d489366b9a6bec644178 | [log] [tgz] |
---|---|---|
author | Frank Barchard <fbarchard@google.com> | Tue Jan 07 12:11:56 2020 -0800 |
committer | XNNPACK Team <xnnpack-github-robot@google.com> | Tue Jan 07 12:12:35 2020 -0800 |
tree | 2aa2db01c516db73dba935c8e18ce11d3b330c8e | |
parent | fa0a432189244bc0b2412455ea3348fdc8a049ec [diff] |
Remove prefetch of output buffer from A53 kernels. Prefetch on output buffer was intended to improve GEMMINC and IGEMM by prefetching from top to bottom, while the store may be reverse order without a penalty. Although gemm benchmarks show a small improvement, end to end doesnt show a real improvement with prefetches on a53 PiperOrigin-RevId: 288546265
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 and 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.
The table below presents multi-threaded performance of XNNPACK library on three generations of MobileNet models and three generations of Raspberry Pi boards.
Model | RPi 2 (BCM2836), ms | RPi 3+ (BCM2837B0), ms | RPi 4 (BCM2711), ms |
---|---|---|---|
MobileNet v1 1.0X | 342 | 122 | 79 |
MobileNet v2 1.0X | 199 | 82 | 47 |
MobileNet v3 Large | 166 | 71 | 42 |
MobileNet v3 Small | 53 | 24 | 15 |
Benchmarked on December 12, 2019 with end2end_bench --benchmark_min_time=5
on a Raspbian Buster build with CMake (./scripts/build-local.sh
) and neural network models with randomized weights and inputs.
XNNPACK is a based on QNNPACK library. Unlike QNNPACK, XNNPACK focuses entirely on floating-point operators, and its API is no longer compatible with QNNPACK.