tangsipeng

about reid model

9

i test the reid model 'ckpt.t7' on Market1501 but only got mAP: 24.8% , can you share your test result ? @ZQPei

about Trainning on the other backbone

6

@ifzhang i read your paper and i decide to train the model on my own backbone. refer to your paper you just train 10 epochs and you can get prettey...

如题，谢谢~

how to deal with machine-generated labels of openimages ?

5

how to deal with machine-generated labels of openimages ?

VFocalLoss in yolov5

3

hi，i test VFocalLoss in yolov5，but not getting improvment. did you have done some test about yolov5 ? or any suggestion ? thank your ~ class VFocalLoss(nn.Module): def __init__(self, loss_fcn, gamma=2.0,...

请教下开源模型中的关键点顺序是否为：左下，右下，右上，左上

如果想训练更大的模型，应该用哪个版本的coco预训练模型？

2

我想训练一个更大的网络，例如yolov5l，请问是否可以直接用yolov5官方的权重，请教下应该用哪个版本的权重？

感谢你分享的代码，有个问题想和你讨论下。 我看到你推理的代码是 outputs_1,outputs_2,outputs_3,_,_,_,_,_,_= model(input_img) outputs = torch.cat((outputs_1,outputs_2,outputs_3),1) 这里 outputs_1,outputs_2,outputs_3 在训练的时候经过分类器分别进行了单独的监督。但是为什么你认为在推理的时候3个特征concat起来效果就好呢？类似的尝试我也试过，好像concat起来并没有什么显著效果。请问你这里的出发点或者依据是什么？

### 是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this? - [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions ### 该问题是否在FAQ中有解答？ | Is there an...