commit	645af97a4c6af150cf70db3dd473ebfeda28ff09	[log] [tgz]
author	Marat Dukhan <maratek@google.com>	Sun Jan 09 22:50:27 2022 -0800
committer	XNNPACK Team <xnnpack-github-robot@google.com>	Sun Jan 09 22:51:44 2022 -0800
tree	6873d0de17ec713ef790f80161e0f48a992acff9
parent	84598220eb0eed791711f93780d4ce83b0e8d4ce [diff]

FMA3 implementation of F16 DWCONV/VCLAMP/VMULCADDC microkernels

PiperOrigin-RevId: 420676523

76 files changed

tree: 6873d0de17ec713ef790f80161e0f48a992acff9

README.md

XNNPACK

XNNPACK is a highly optimized library of floating-point neural network inference operators for ARM, WebAssembly, and x86 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 TensorFlow Lite, TensorFlow.js, PyTorch, and MediaPipe.

Supported Architectures

ARM64 on Android, Linux, macOS, and iOS (including WatchOS and tvOS)
ARMv7 (with NEON) on Android
ARMv6 (with VFPv2) on Linux
x86 and x86-64 (up to AVX512) on Windows, Linux, macOS, Android, and iOS simulator
WebAssembly MVP
WebAssembly SIMD
RISC-V (RV32GV and RV64GC)

Operator Coverage

XNNPACK implements the following neural network operators:

2D Convolution (including grouped and depthwise)
2D Deconvolution (AKA Transposed Convolution)
2D Average Pooling
2D Max Pooling
2D ArgMax Pooling (Max Pooling + indices)
2D Unpooling
2D Bilinear Resize
2D Depth-to-Space (AKA Pixel Shuffle)
Add (including broadcasting, two inputs only)
Subtract (including broadcasting)
Divide (including broadcasting)
Maximum (including broadcasting)
Minimum (including broadcasting)
Multiply (including broadcasting)
Squared Difference (including broadcasting)
Global Average Pooling
Channel Shuffle
Fully Connected
Abs (absolute value)
Bankers' Rounding (rounding to nearest, ties to even)
Ceiling (rounding to integer above)
Clamp (includes ReLU and ReLU6)
Convert (includes fixed-point and half-precision quantization and dequantization)
Copy
ELU
Floor (rounding to integer below)
HardSwish
Leaky ReLU
Negate
Sigmoid
Softmax
Square
Truncation (rounding to integer towards zero)
PReLU

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.

Performance

Mobile phones

The table below presents single-threaded performance of XNNPACK library on three generations of MobileNet models and three generations of Pixel phones.

Model	Pixel, ms	Pixel 2, ms	Pixel 3a, ms
FP32 MobileNet v1 1.0X	82	86	88
FP32 MobileNet v2 1.0X	49	53	55
FP32 MobileNet v3 Large	39	42	44
FP32 MobileNet v3 Small	12	14	14

The following table presents multi-threaded (using as many threads as there are big cores) performance of XNNPACK library on three generations of MobileNet models and three generations of Pixel phones.

Model	Pixel, ms	Pixel 2, ms	Pixel 3a, ms
FP32 MobileNet v1 1.0X	43	27	46
FP32 MobileNet v2 1.0X	26	18	28
FP32 MobileNet v3 Large	22	16	24
FP32 MobileNet v3 Small	7	6	8

Benchmarked on March 27, 2020 with end2end_bench --benchmark_min_time=5 on an Android/ARM64 build with Android NDK r21 (bazel build -c opt --config android_arm64 :end2end_bench) and neural network models with randomized weights and inputs.

Raspberry Pi

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 Zero W (BCM2835), ms	RPi 2 (BCM2836), ms	RPi 3+ (BCM2837B0), ms	RPi 4 (BCM2711), ms	RPi 4 (BCM2711, ARM64), ms
FP32 MobileNet v1 1.0X	3937	299	114	72	76
FP32 MobileNet v2 1.0X	1987	187	79	41	44
FP32 MobileNet v3 Large	1658	158	67	38	41
FP32 MobileNet v3 Small	487	50	23	13	14
INT8 MobileNet v1 1.0X	2598	169	61	29	24
INT8 MobileNet v2 1.0X	1487	109	40	20	17

Benchmarked on Oct 15, 2021 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. INT8 inference was evaluated on per-channel quantization schema.

Publications

Marat Dukhan "The Indirect Convolution Algorithm". Presented on Efficient Deep Learning for Compute Vision (ECV) 2019 workshop (slides, paper on ArXiv).
Erich Elsen, Marat Dukhan, Trevor Gale, Karen Simonyan "Fast Sparse ConvNets". Paper on ArXiv, pre-trained sparse models.
Marat Dukhan, Artsiom Ablavatski "The Two-Pass Softmax Algorithm". Paper on ArXiv.
Yury Pisarchyk, Juhyun Lee "Efficient Memory Management for Deep Neural Net Inference". Paper on ArXiv.

Ecosystem

Machine Learning Frameworks

Acknowledgements

XNNPACK is a based on QNNPACK library. Over time its codebase diverged a lot, and XNNPACK API is no longer compatible with QNNPACK.