一、定义

1. 分类
2. 模型量化接口进阶
3. 量化支持的模式/流/硬件

二、实现

1. 分类
   
   
2. 模型量化接口进阶
   https://pytorch.org/docs/stable/quantization-support.html
   2.1 算子融合
   定义： 将多个算子融合到一起，运算时可以加快运行速度。

import torch
# define a floating point model where some layers could benefit from QAT
class M(torch.nn.Module):
    def __init__(self):
        super().__init__()
        # QuantStub converts tensors from floating point to quantized
        self.quant = torch.ao.quantization.QuantStub()
        self.conv = torch.nn.Conv2d(1, 1, 1)
        self.bn = torch.nn.BatchNorm2d(1)
        self.relu = torch.nn.ReLU()
        # DeQuantStub converts tensors from quantized to floating point
        self.dequant = torch.ao.quantization.DeQuantStub()

    def forward(self, x):
        x = self.quant(x)
        x = self.conv(x)
        x = self.bn(x)
        x = self.relu(x)
        x = self.dequant(x)
        return x

model_fp32 = M()
# model must be set to eval for fusion to work
model_fp32.eval()
print(model_fp32)
model_fp32_fused = torch.ao.quantization.fuse_modules(model_fp32,
    [['conv', 'bn', 'relu']])
print(model_fp32_fused)

3. 量化支持的模式/流/硬件
选择条件： 模式支持、硬件支持、算子支持等。
3.1 量化模式选择

3.2 硬件模式选择

3.3 算子对模式的支持