This file will show you how to quantize your network with PPQ
You should prepare your model and calibration dataset as follow:
~/working/model.onnx<--your model
~/working/data/.npy or~/working/data/.bin<--your dataset
if you are using caffe model:
~/working/model.caffemdoel<--your model
~/working/model.prototext<--your model
###MAKE SURE YOUR INPUT LAYOUT IS[N,C,H,W]or[C,H,W]###
quantized model will be generated at:~/working/quantized.onnx
"""
from ppq import*
from ppq.api import*
import os
#modify configuration below:
WORKING_DIRECTORY='working'#choose your working directory
TARGET_PLATFORM=TargetPlatform.PPL_CUDA_INT8#choose your target platform
MODEL_TYPE=NetworkFramework.ONNX#or NetworkFramework.CAFFE
INPUT_LAYOUT='chw'#input data layout,chw or hwc
NETWORK_INPUTSHAPE=[1,3,224,224]#input shape of your network
CALIBRATION_BATCHSIZE=16#batchsize of calibration dataset
EXECUTING_DEVICE='cuda'#'cuda'or'cpu'.
REQUIRE_ANALYSE=False
DUMP_RESULT=False#是否需要Finetuning一下你的网络
#SETTING对象用于控制PPQ的量化逻辑
#当你的网络量化误差过高时，你需要修改SETTING对象中的参数进行特定的优化
SETTING=UnbelievableUserFriendlyQuantizationSetting(
platform=TARGET_PLATFORM,finetune_steps=2500,
finetune_lr=1e-3,calibration='kl',#【改】量化算法可选'kl','pecentile','mse'
equalization=True,non_quantable_op=None)
SETTING=SETTING.convert_to_daddy_setting()
print('正准备量化你的网络，检查下列设置:')
print(f'WORKING DIRECTORY:{WORKING_DIRECTORY}')
print(f'TARGET PLATFORM:{TARGET_PLATFORM.name}')
print(f'NETWORK INPUTSHAPE:{NETWORK_INPUTSHAPE}')
print(f'CALIBRATION BATCHSIZE:{CALIBRATION_BATCHSIZE}')
#此脚本针对单输入模型，输入数据必须是图像数据layout:[n,c,h,w]
#如果你的模型具有更复杂的输入格式，你可以重写下面的load_calibration_dataset函数
#请注意，任何可遍历对象都可以作为PPQ的数据集作为输入
dataloader=load_calibration_dataset(
directory=WORKING_DIRECTORY,
input_shape=NETWORK_INPUTSHAPE,
batchsize=CALIBRATION_BATCHSIZE,
input_format=INPUT_LAYOUT)
print('网络正量化中，根据你的量化配置，这将需要一段时间:')
quantized=quantize(
working_directory=WORKING_DIRECTORY,setting=SETTING,
model_type=MODEL_TYPE,executing_device=EXECUTING_DEVICE,
input_shape=NETWORK_INPUTSHAPE,target_platform=TARGET_PLATFORM,
dataloader=dataloader,calib_steps=32)
