You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, thank you for sharing your code. However, I encountered an issue where the results are not reproducible. I ran the model with the exact same environment, parameters, and even fixed the random seed, but the results keep getting worse, and I can't achieve reproducibility. Have you ever faced this issue? How did you solve it?
This is the result of two training sessions with fixed random seeds.
[07/26 20:33:37] opendet2.evaluation.pascal_voc_evaluation INFO:
mAP
WI
AOSE
AP@K
P@K
R@K
AP@U
P@U
R@U
15.93
21.48
28328.00
63.69
9.21
91.35
16.36
28.97
29.44
[07/26 20:33:37] opendet2.engine.defaults INFO: Evaluation results for voc_coco_20000_test in csv format:
Hi, thank you for sharing your code. However, I encountered an issue where the results are not reproducible. I ran the model with the exact same environment, parameters, and even fixed the random seed, but the results keep getting worse, and I can't achieve reproducibility. Have you ever faced this issue? How did you solve it?
This is the result of two training sessions with fixed random seeds.
[07/26 20:33:37] opendet2.evaluation.pascal_voc_evaluation INFO:
[08/15 06:26:48] opendet2.evaluation.pascal_voc_evaluation INFO:
The text was updated successfully, but these errors were encountered: