This tutorial demonstrates how to migrate quantization pipeline written using the OpenVINO Post-Training Optimization Tool (POT) to NNCF Post-Training Quantization API. This tutorial is based on Ultralytics Yolov5 model and additionally it compares model accuracy between the FP32 precision and quantized INT8 precision models and runs a demo of model inference based on sample code from Ultralytics Yolov5 with the OpenVINO backend.
The tutorial consists from the following parts:
- Convert YOLOv5 model to OpenVINO IR.
- Prepare dataset for quantization.
- Configure quantization pipeline.
- Perform model optimization.
- Compare accuracy FP32 and INT8 models
- Run model inference demo
- Compare performance FP32 and INt8 models
If you have not installed all required dependencies, follow the Installation Guide.