ilit 1.0b0

Repository of Intel® Low Precision Optimization Tool

Intel® Low Precision Optimization Tool

Intel® Low Precision Optimization Tool is an open-source python library which is intended to deliver a unified low-precision inference interface cross multiple Intel optimized DL frameworks on both CPU and GPU. It supports automatic accuracy-driven tuning strategies, along with additional objectives like performance, model size, or memory footprint. It also provides the easy extension capability for new backends, tuning strategies, metrics and objectives.

WARNING

GPU support is under development.

Currently supported Intel optimized DL frameworks are:

• Tensorflow*
• PyTorch*
• Apache* MXNet

Currently supported tuning strategies are:

• Basic
• Bayesian
• MSE
• Exhaustive
• Random

Introduction

Introduction explains Intel® Low Precision Optimization Tool infrastructure, design philosophy, supported functionality, details of tuning strategy implementations and tuning result on popular models.

Tutorials

• Hello World demonstrates the simple steps to utilize Intel® Low Precision Optimization Tool for quanitzation, which can help you quick start with the tool.
• Tutorials provides comprehensive instructions of how to utilize diffrennt features of Intel® Low Precision Optimization Tool.
• Examples is a tuning zoo to demonstrate the usage of Intel® Low Precision Optimization Tool in TensorFlow, PyTorch and MxNet for industry models of diffrent categories.

Install from source

git clone https://github.com/intel/lp-opt-tool.git
cd lp-opt-tool
python setup.py install

Install from binary

# install from pip
pip install ilit

# install from conda
conda config --add channels intel
conda install ilit

System Requirements

Hardware

Intel® Low Precision Optimization Tool supports systems based on Intel 64 architecture or compatible processors.

Software

Intel® Low Precision Optimization Tool requires to install Intel optimized framework version for TensorFlow, PyTorch, and MXNet.

Tuning Zoo

The followings are the examples integrated with Intel® Low Precision Optimization Tool for auto tuning.

TensorFlow Model	Category
ResNet50 V1	Image Recognition
ResNet50 V1.5	Image Recognition
ResNet101	Image Recognition
Inception V1	Image Recognition
Inception V2	Image Recognition
Inception V3	Image Recognition
Inception V4	Image Recognition
ResNetV2_50	Image Recognition
ResNetV2_101	Image Recognition
ResNetV2_152	Image Recognition
Inception ResNet V2	Image Recognition
SSD ResNet50 V1	Object Detection
Wide & Deep	Recommendation
VGG16	Image Recognition
VGG19	Image Recognition
Style_transfer	Style Transfer

PyTorch Model	Category
BERT-Large RTE	Language Translation
BERT-Large QNLI	Language Translation
BERT-Large CoLA	Language Translation
BERT-Base SST-2	Language Translation
BERT-Base RTE	Language Translation
BERT-Base STS-B	Language Translation
BERT-Base CoLA	Language Translation
BERT-Base MRPC	Language Translation
DLRM	Recommendation
BERT-Large MRPC	Language Translation
ResNext101_32x8d	Image Recognition
BERT-Large SQUAD	Language Translation
ResNet50 V1.5	Image Recognition
ResNet18	Image Recognition
Inception V3	Image Recognition
YOLO V3	Object Detection
Peleenet	Image Recognition
ResNest50	Image Recognition
SE_ResNext50_32x4d	Image Recognition
ResNet50 V1.5 QAT	Image Recognition
ResNet18 QAT	Image Recognition

MxNet Model	Category
ResNet50 V1	Image Recognition
MobileNet V1	Image Recognition
MobileNet V2	Image Recognition
SSD-ResNet50	Object Detection
SqueezeNet V1	Image Recognition
ResNet18	Image Recognition
Inception V3	Image Recognition

Known Issues

1. KL Divergence Algorithm is very slow at TensorFlow

   Due to TensorFlow not supporting tensor dump naturally, current solution of dumping the tensor content is adding print op and dumpping the value to stdout. So if the model to tune is a TensorFlow model, please restrict calibration.algorithm.activation and calibration.algorithm.weight in user yaml config file to minmax.

2. MSE tuning strategy doesn't work with PyTorch adaptor layer

   MSE tuning strategy requires to compare FP32 tensor and INT8 tensor to decide which op has impact on final quantization accuracy. PyTorch adaptor layer doesn't implement this inspect tensor interface. So if the model to tune is a PyTorch model, please not choose MSE tuning strategy.

Support

Please submit your questions, feature requests, and bug reports on the GitHub issues page. You may also reach out to [e-mail].

Contributing

We welcome community contributions to Intel® Low Precision Optimization Tool. If you have an idea on how to improve the library:

• For changes impacting the public API, submit an RFC pull request.
• Ensure that the changes are consistent with the code contribution guidelines and coding style.
• Ensure that you can run all the examples with your patch.
• Submit a pull request.

For additional details, see contribution guidelines.

This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.

License

Intel® Low Precision Optimization Tool is licensed under Apache License Version 2.0. This software includes components with separate copyright notices and license terms. Your use of the source code for these components is subject to the terms and conditions of the following licenses.

Apache License Version 2.0:

• Intel TensorFlow Quantization Tool

MIT License:

• bayesian-optimization

See accompanying LICENSE file for full license text and copyright notices.

---

Legal Information

Citing

If you use Intel® Low Precision Optimization Tool in your research or wish to refer to the tuning results published in the Tuning Zoo, please use the following BibTeX entry.

@misc{Intel® Low Precision Optimization Tool,
  author =       {Feng Tian, Chuanqi Wang, Guoming Zhang, Penghui Cheng, Pengxin Yuan, Haihao Shen, and Jiong Gong},
  title =        {Intel® Low Precision Optimization Tool},
  howpublished = {\url{https://github.com/intel/lp-opt-tool}},
  year =         {2020}
}