openvino量化自己训练的yolov3模型至int8(有成功验证截图)

FP16 or FP32 to int8:

接着上一篇博客将darknet训练得到的model转换为IR模型后,该IR模型为Inter 加速推理后的模型
IR模型为FP16 or FP32
接下来将IR模型转换为INT8,得到一个速度较快准确率略低的模型
下述为实现步骤:

ubuntu18.04: openvino2020.4 :

pip list :
Package Version


accuracy-checker 0.7.7
addict 2.2.1
apturl 0.5.2
asn1crypto 0.24.0
Brlapi 0.6.6
certifi 2018.1.18
cffi 1.15.0
chardet 3.0.4
click 6.7
colorama 0.3.7
command-not-found 0.3
cryptography 2.1.4
cupshelpers 1.0
defer 1.0.6
defusedxml 0.7.1
distro-info 0.18ubuntu0.18.04.1
future 0.18.2
httplib2 0.9.2
hyperopt 0.1.2
idna 2.6
jstyleson 0.0.2
keyring 10.6.0
keyrings.alt 3.0
language-selector 0.1
launchpadlib 1.10.6
lazr.restfulclient 0.13.5
lazr.uri 1.0.3
louis 3.5.0
macaroonbakery 1.1.3
Mako 1.0.7
MarkupSafe 1.0
netifaces 0.10.4
networ

Guess you like

Origin blog.csdn.net/QQ_778132974/article/details/122219510