Training YOLOv4 CNN's Using darknet

preview_player
Показать описание

Revision 05/13/20, 12:00 PM: Debugging: I was able to build the "Debug x64" version of darknet (YOLOv3 and YOLOv4) by changing Project Properties -- C/C++ -- Code Generation -- Runtime Library from "Multi-threaded Debug DLL (/MDd)" to "Multi-threaded DLL (/MD)". I wonder what all the differences are, and if there will be any complications?
Answer 05/17/20, 12:47 AM: Reading images into OpenCV Mat objects fails when launching a Debug build and using OpenCV Release libraries.

Рекомендации по теме
Комментарии
Автор

Thank you Tom, you really guided me through it with your tutorial!

mrblonde
Автор

Really great training tutorial Tom :) Well done!

Augmented_AI
Автор

Trying to run the power shell command in an anaconda prompt as I don’t have access to my powershell (IT security issue). It’s not recognising the command and I don’t have the darknet executable in the specified location?

callumbrown
Автор

Hello Tom, thanks for sharing your knowledge in this video. At 29:08, you said we need to set classes=1. If we have multiple classes, do we still need to have this set to classes=1, or classes=n (n number of classes)? thanks!

RyanMartinRAM
join shbcf.ru