Closed hhaAndroid closed 3 years ago
Check here. yolof_r50_c5_1x.log
Thank you!
BTW, when we reimplement the model, we observe some variance (about 0.2~0.3 AP lower than the performance reported in the paper).
BTW, when we reimplement the model, we observe some variance (about 0.2~0.3 AP lower than the performance reported in the paper).
OK
Hi, Can you provide the distributed training log corresponding to the
yolof_R_50_C5_1x.yaml
configuration