wulele
wulele
大佬您好，我训练HRSC2016数据集最终train loss和val loss维持在4左右就不变了，测试了demo发现预测效果也不好。能指点一二吗？谢谢了
Hi, I used your code training on the voc2007 train data which contains 5k images, and then run your demo finding the results is very bad. Could it be the...
I tried to reproduce the navie direct pose. Could you please help me check that my loss has become Nan?Could u please check it?thanks! class kptregression(torch.nn.Module): def __init__(self): super(kptregression, self).__init__()...
### 问题确认 Search before asking - [X] 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question >>> import paddle >>> paddle.utils.run_check() Running verify...
I compiled the operator, but the following error occurred during evaluation 4-scale-r50. Pytorch = 1.9.0. Looking forward to your reply.
 Thank you very much for your work, I usually need to do some lightweight work, the following two parts may not be friendly to our hardware support, so how...
Thank you very much for your outstanding work！I'm so sorry to bother you: I'm ready to fine-tune the data as well as the environment, and this is my directory structure：...
Hello, thanks for your code. i run your code use: bash main.py 8 -- train , after 1000 epoch, the reults is 69.1 and 48.2. The oks is slight worse...
Hello, I have tested your open-source weight pretrain/yoloe-v8s-seg.pt. My metrics on minival are as follows:  Why does this AP_r seem a little low? Please help me clarify.