HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

Default confidences predict 0.25, val 0.001 (#149)

Browse source
This commit is contained in:
Glenn Jocher 2023-01-05 15:56:52 +01:00 • committed by GitHub
parent 3cbf3ec455
commit 314da263c7
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 6 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ultralytics/yolo/configs/default.yaml
View file

@ -39,7 +39,7 @@ dropout: False # use dropout regularization
val: True # validate/test during training
save_json: False # save results to JSON file
save_hybrid: False # save hybrid version of labels (labels + additional predictions)
conf: 0.001 # object confidence threshold for detection
conf: null # object confidence threshold for detection (default 0.25 predict, 0.001 val)
iou: 0.7 # intersection over union (IoU) threshold for NMS
max_det: 300 # maximum number of detections per image
half: False # use half precision (FP16)