Skip to content

pretrained file #2198

Answered by glenn-jocher
Phelan164 asked this question in Q&A
Feb 13, 2021 · 1 comments · 1 reply
Discussion options

You must be logged in to vote

@Phelan164 YOLOv5x6 has an additional P6 output layer (4 outputs total, P3-P6), vs 3 outputs (P3-P5) in the standard models. The P6 layer is for larger objects, so we've trained this model at --img 1280. It performs at about 54.0 AP on COCO.

Replies: 1 comment 1 reply

Comment options

You must be logged in to vote
1 reply
@Phelan164
Comment options

Answer selected by Phelan164
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants