Overview

This section describes the AMCT-based quantization scenarios and the functions in each scenario.

Based on whether retraining is required, quantization is classified into PTQ and QAT. For details about quantization concepts, see Compression Concepts. For details about quantization algorithms, see Compression Algorithms.

• PTQ

  Quantization takes two forms: Manual Quantization and Automatic Quantization based on whether the quantization configuration file is manually tuned after quantization.

  If the quantized model fails to meet the accuracy requirements, perform automatic quantization (recommended) or manual tuning.

• Quantization aware training

  Currently, QAT supports quantization only for float32 network models.

  Only manual quantization is supported at present. If the quantized model fails to meet the accuracy requirements, perform manual tuning.