Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

yolo-tiny.cfg vs. yolo-baby.cfg #11

Closed
zayfod opened this issue Nov 21, 2016 · 5 comments
Closed

yolo-tiny.cfg vs. yolo-baby.cfg #11

zayfod opened this issue Nov 21, 2016 · 5 comments

Comments

@zayfod
Copy link
Contributor

zayfod commented Nov 21, 2016

Not really an issue but, I wonder what is the difference between the yolo-tiny.cfg vs. yolo-baby.cfg and whether yolo-baby.cfg is supposed to work in darktf revision c963245 ?

Is yolo-baby.cfg an older style Darknet configuration?

@thtrieu
Copy link
Owner

thtrieu commented Nov 21, 2016

They're not the same. yolo-tiny.cfg is old and yolo-baby is actually the new tiny-yolo.cfg. Except that I am still working on batch norm layer so it is not going to work for tiny-yolo soon.

I am trying a lot to refactor codes to see if people can help me out on the coding :D

@zayfod
Copy link
Contributor Author

zayfod commented Nov 21, 2016

Hey, thanks for the prompt response!

My experience is that yolo-tiny.cfg works today, and yolo-baby does not. Both load, but only yolo-tiny produces actual output. I do not understand what is wrong with yolo-baby.cfg , would you please elaborate a bit?

@thtrieu
Copy link
Owner

thtrieu commented Nov 21, 2016

@zayfod It's a bit confusing, I'll make it clear:
yolo-tiny is the old yolo-tiny (183MB)
yolo-baby is the new tiny-yolo (103MB)
And anything new is not yet implemented, hence your situation :)

@zayfod
Copy link
Contributor Author

zayfod commented Nov 21, 2016

OK. Got it. Is there an older version of Darknet, that I can temporarily revert to, that uses the old yolo-tiny (183 MB)?

E.g. revision e6b3006d ?

@thtrieu
Copy link
Owner

thtrieu commented Nov 21, 2016

New code is compatible to old code. You just have to download the old weights file and old config files and put them in the right places. Read #1

@thtrieu thtrieu closed this as completed Nov 22, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants