missbook520

Thanks for your great work Have you finished Cascade train?

Modelnet40 eval

2

Hello, thank you very much for your work！ **But I used the pospool_sin_cos_avg pre-trained model you provided to predict modelnet40, and the results are as follows:** [11/29 13:58:30] modelnet40_eval INFO:...

Network structure

7

您好，我看您给出的pytorch-unet链接中，编码器中最后一层的down层（down4）为： self.down4 = Down(512, 1024 // factor)，输出的维度的为1024通道，而您写的tensorrt unet网络中最后一层的down层为： 为auto x5 = down(network,weightMap,*x4->getOutput(0),512,1,"down4");好像输出的维度是512，请问是需要做相应的修改吗？

Can use python?

5

Thank you very much for your work, can I use python to call it?

HOW to get desired prediction?

7

i run the main.py, parameter is default,then run the test_cenet.py,but the prediction is not good,could you give me some suggestion?thank you very much!