Pytorch Quantization

PyTorch-Quantization is a toolkit for training and evaluating PyTorch models with simulated quantization. Quantization can be added to the model automatically, or manually, allowing the model to be tuned for accuracy and performance. Quantization is compatible with NVIDIAs high performance integer kernels which leverage integer Tensor Cores. The quantized model can be exported to ONNX and imported by TensorRT 8.0 and later.

Install

Binaries

pip install pytorch-quantization --extra-index-url https://pypi.ngc.nvidia.com

From Source

git clone https://github.com/NVIDIA/TensorRT.git
cd tools/pytorch-quantization

Install PyTorch and prerequisites

pip install -r requirements.txt
# for CUDA 10.2 users
pip install torch>=1.8.0
# for CUDA 11.1 users
pip install torch>=1.8.0+cu111

Build and install pytorch-quantization

# Python version >= 3.7, GCC version >= 5.4 required
python setup.py install

NGC Container

pytorch-quantization is preinstalled in NVIDIA NGC PyTorch container since 20.12, e.g. nvcr.io/nvidian/pytorch:20.12-py3

Resources

Pytorch Quantization Toolkit userguide
Quantization Basics whitepaper