commit | 8fe54e48c25596598f79379b747da1608e949063 | [log] [tgz] |
---|---|---|
author | Marat Dukhan <maratek@google.com> | Thu Oct 10 14:12:59 2019 -0700 |
committer | XNNPACK Team <xnnpack-github-robot@google.com> | Thu Oct 10 14:13:27 2019 -0700 |
tree | 56a31a0ac24b97efec02097adfe5ecade871d096 | |
parent | 810171ddd29de6e621268f3f1383752efde47f0d [diff] |
Extra :xnnpack_operators_nhwc_f32 target with only F32 operators in NHWC layout Reduce build size for typical use-cases. size_test reduction: - Android/ARMv7: 185K -> 167K - Android/ARM64: 188K -> 156K - Android/x86: 147K -> 121K - WAsm: 50K -> 40K PiperOrigin-RevId: 274036079
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.