DMGNN icon indicating copy to clipboard operation
DMGNN copied to clipboard

human3.6数据集提问

Open helloful opened this issue 3 years ago • 4 comments

您好,在您的paper中,讲到human3.6有32个关节点,但是您代码中的数据并不是表示关节点的三维坐标,而是表示旋转向量;其次,您的代码中,您训练后的模型并没有完整的提供99个数据,训练模型仅仅输出了30个数据,请问您是如何可视化最后的预测结果的?得到您paper中的图的呢?

helloful avatar Apr 12 '21 13:04 helloful

您好,请问您现在代码复现成功了吗?复现的时候有没有遇到这个问题呢,求教谢谢! Traceback (most recent call last): File "main.py", line 23, in p.start() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\processor.py", line 126, in start self.test() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\recognition.py", line 255, in test self.MAE_tensor[iter_time, :, 9])/4) AttributeError: 'REC_Processor' object has no attribute 'mean_MAE_select'

Charles-9802 avatar Jul 01 '21 12:07 Charles-9802

我复现了啊,其实作者的代码中,提供了torchlight,照着安装一下就好了

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月1日(星期四) 晚上8:55 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [limaosen0/DMGNN] human3.6数据集提问 (#9)

您好,请问您现在代码复现成功了吗?复现的时候有没有遇到这个问题呢,求教谢谢! Traceback (most recent call last): File "main.py", line 23, in p.start() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\processor.py", line 126, in start self.test() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\recognition.py", line 255, in test self.MAE_tensor[iter_time, :, 9])/4) AttributeError: 'REC_Processor' object has no attribute 'mean_MAE_select'

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

helloful avatar Jul 01 '21 12:07 helloful

好的谢谢

Charles-9802 avatar Jul 01 '21 12:07 Charles-9802

我运行这个的时候,也出现了跟你同样的问题。请问你解决了吗?我用Google colab 按照作者的指令安装,在第一步GPU那里就出错了。然后我直接尝试运行已保存的model, 就出现了跟你一样的运行报错结果。

您好,请问您现在代码复现成功了吗?复现的时候有没有遇到这个问题呢,求教谢谢! Traceback (most recent call last): File "main.py", line 23, in p.start() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\processor.py", line 126, in start self.test() File "D:\DMGNN\dwgnn-master\DMGNN-master\h36m-short\processor\recognition.py", line 255, in test self.MAE_tensor[iter_time, :, 9])/4) AttributeError: 'REC_Processor' object has no attribute 'mean_MAE_select'

Cindy0532 avatar Aug 26 '21 19:08 Cindy0532