commit | 8d3c07e03a55862847e0a6a90f6f9177e87dba4a | [log] [tgz] |
---|---|---|
author | Marat Dukhan <maratek@google.com> | Thu Jan 02 01:20:59 2020 -0800 |
committer | XNNPACK Team <xnnpack-github-robot@google.com> | Thu Jan 02 01:21:29 2020 -0800 |
tree | e328046667bbb59bdd88ce320abcf29f8857cc9a | |
parent | 279908a1af406a1973069979906d9fae569719fa [diff] |
Additional Sigmoid micro-kernels and accuracy evaluation stub - PSIMD micro-kernels and accuracy evaluation stubs - ARM NEON micro-kernels using 2048-entry table lookups - ARM NEON micro-kernels with alternative division implementations - ARM NEON micro-kernels without FMA - x4..x24 version of all SIMD micro-kernels - Eliminated comparison with one_cutoff & corresponding blend in all micro-kernels PiperOrigin-RevId: 287804583
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.