compression

Model Compression

• DeepSparse

  • ResNet-50 on CPUs: Sparsifying for Better Performance

• GitHub - open-mmlab/mmrazor: OpenMMLab Model Compression Toolbox and Benchmark.

• GitHub - microsoft/VPTQ: VPTQ, A Flexible and Extreme low-bit quantization algorithm

Quantization

• A Visual Guide to Quantization - by Maarten Grootendorst

https://github.com/666DZY666/micronet

Lower Precision

GitHub - TimDettmers/bitsandbytes: 8-bit CUDA functions for PyTorch

Knowledge Distillation

For Knowledge Distillation see this note

Lectures

https://efficientml.ai/schedule/