fbsamples/pytorch-quantization-workshop

Code for a workshop hosted at the MLOps World Summit '22

pytorch-quantization-workshop

This repo holds the files for the PyTorch Quantization Workshop conducted by Suraj Subramanian at the MLOpsWorld Conference on June 8 2022.

Notebooks

Quant_101.ipynb

Learn the fundamentals of quantization in pure Python code.

Quant_API.ipynb

Learn about quantization schemes, when some are better than others, and using QConfigs in PyTorch

Quant_Workflow.ipynb

The number of available options can be overwhelming. Choosing the correct quantization technique and scheme is an empirical process; this notebook contains a workflow that aids choosing the most suitable option to quantize your FP32 model.

Requirements

• An x86 or ARM CPU
• PyTorch 1.10.0+

Further Reading

• Quantization — PyTorch 1.11.0 documentation
• Practical Quantization in PyTorch
• FX Graph Mode Quantization User Guide
• PyTorch Forum - Quantization
• PyTorch Github Issues

Issues/Requests

If you encounter a bug, please open an issue or a PR. See CONTRIBUTING.MD