chj1933
chj1933
@ahangchen同学，你好！请问我直接用你的孪生网络在1501上训练一个模型，应该用那部分代码？
@ahangchen ,你好，请问你训练时后面的-acc是什么的准确率？
@wzjiang ,请问你是怎样处理 ValueError: expected 2D or 3D input (got 1D input)？
@huanghoujing ，Hello， if I want to save the final model_weight.pth, how should I do?
@seathiefwang ，if I want to use my own dataset , how to train it?
@seathiefwang , If i use my own dataset , It's appropriate what parameters should I usually adjust?
@seathiefwang ，thanks! It works when I use my own dataset , but the mAP and rank1 are not very high , if I want to improve the mAP and rank1,...
@seathiefwang ，please tell me how are your training parameters set？I always lower 1% than your, it can't achieve rank1 95.8% and mAP 93.5%
@seathiefwang , why is lower when I use 4 GPUS than 2 GPUS, other parameters remain the same
@seathiefwang ，你好，若果我用自己的数据训练，我在解析自己的数据时候除了修改（market1501.py）把ID信息和摄像头信息导入,还要修改哪里嘛？不然会在dist = cdist(qf,gf)显示XA must be a 2-dimensional array