MyGit

v1.4

intel/neural-compressor

版本发布时间: 2021-05-31 02:21:13

intel/neural-compressor最新发布版本:v2.6(2024-06-14 21:55:11)

Intel® Low Precision Optimization Tool v1.4 release is featured by:

Quantization

  1. PyTorch FX-based quantization support
  2. TensorFlow & ONNX RT quantization enhancement

Pruning

  1. Pruning/sparsity API refinement
  2. Magnitude-based pruning on PyTorch

Model Zoo

  1. INT8 key models updated (BERT on TensorFlow, DLRM on PyTorch, etc.)
  2. 20+ HuggingFace model quantization

User Experience

  1. More comprehensive logging message
  2. UI enhancement with FP32 optimization, auto-mixed precision (BF16/FP32), and graph visualization
  3. Online document: https://intel.github.io/lpot

Extended Capabilities

  1. Model conversion from QAT to Intel Optimized TensorFlow model

Validated Configurations:

Distribution:

  Channel Links Install Command
Source Github https://github.com/intel/lpot.git $ git clone https://github.com/intel/lpot.git
Binary Pip https://pypi.org/project/lpot $ pip install lpot
Binary Conda https://anaconda.org/intel/lpot $ conda install lpot -c conda-forge -c intel

Contact:

Please feel free to contact lpot.maintainers@intel.com, if you get any questions.

相关地址:原始地址 下载(tar) 下载(zip)

查看:2021-05-31发行的版本