Issues
- 0
在训练了几个epoch之后报错:CUDA out of memory
#30 opened by zhuxz0299 - 3
Why do you use segmentation labels instead of bounding boxes in the COCO dataset?
#28 opened by gongyan1 - 0
Continue training model on new data
#27 opened by mansi-aggarwal-2504 - 2
Evaluation metrics explanation
#26 opened by ojasvijain - 1
Early stopping or using epoch_xx.pt
#25 opened by mansi-aggarwal-2504 - 3
Config for custom data
#24 opened by mansi-aggarwal-2504 - 7
Which model path should be provided for dynamic thresholds for variable-speed inference step?
#23 opened by mansi-aggarwal-2504 - 2
您好,想请问一下原文中公式13的问题
#21 opened by CZY-Code - 3
不好意思,您能帮我解决一下这个问题吗?
#13 opened by 1324039468 - 1
您好,没有在您分享的代码中找到关于Resnet网络的代码,可以提供这一部分的代码吗?
#20 opened by chenxlin222 - 2
冒昧问一下大佬这个有什么前景/应用价值吗
#19 opened by feidufuzhou - 26
- 0
代码在test和detect时预测的标签中的类别不显示
#17 opened by yangyahu-1994 - 2
关于模型FLOPs和Parameters的计算
#18 opened by yangyahu-1994 - 1
关于Adaptive router计算的difficulty score
#16 opened by yangyahu-1994 - 7
关于Variable-speed inference的疑问
#15 opened by yangyahu-1994 - 4
关于训练adaptive router
#14 opened by yangyahu-1994 - 4
关于Fig 7的疑问
#12 opened by yangyahu-1994 - 2
- 2
The implementation of Dy-YOLOv7 is developed by the YOLOv7 [45] framework, with two identical detectors. The implementation of dynamic two-stage detectors is developed by the open-source CBNet [24] framework, with two identical backbones and a shared neck and head.
#10 opened by yangyahu-1994 - 4
为什么奖励和惩罚项是训练损失差的一半?
#9 opened by yangyahu-1994 - 4
- 2
运行python train_step2.py 报错,
#6 opened by LeonNerd - 1
Whether onnx or trt is supported
#5 opened by LeonNerd - 2
请问有提供权重文件吗?
#4 opened by SSZ1 - 1
想请教下论文中优化策略的自适应偏差问题
#3 opened by 781458112 - 1
- 1