junjieliuswu / lanedetection_end2end Goto Github PK

<class 'torch.Tensor'>
torch.Size([1, 3, 256, 512])
Traceback (most recent call last):
File "test.py", line 165, in
main()
File "test.py", line 155, in main
beta0, beta1, beta2, beta3, weightmap_zeros, M, output_net, outputs_line, outputs_hozizon = model(image, True)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(input, **kwargs)
File "C:\Users\Sharvari\Desktop\SCIENTISTS\24-11-2021\LaneDetection_End2End\Birds_Eye_View_Loss\LaneDetection_End2End\Birds_Eye_View_Loss\Networks\LSQ_layer.py", line 295, in forward
shared_encoder, output = self.net(input, end_to_endself.pretrained)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "C:\Users\Sharvari\Desktop\SCIENTISTS\24-11-2021\LaneDetection_End2End\Birds_Eye_View_Loss\LaneDetection_End2End\Birds_Eye_View_Loss\Networks\ERFNet.py", line 155, in forward
encoder_output = self.encoder(input)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "C:\Users\Sharvari\Desktop\SCIENTISTS\24-11-2021\LaneDetection_End2End\Birds_Eye_View_Loss\LaneDetection_End2End\Birds_Eye_View_Loss\Networks\ERFNet.py", line 87, in forward
output = self.initial_block(input)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "C:\Users\Sharvari\Desktop\SCIENTISTS\24-11-2021\LaneDetection_End2End\Birds_Eye_View_Loss\LaneDetection_End2End\Birds_Eye_View_Loss\Networks\ERFNet.py", line 20, in forward
output = torch.cat([self.conv(input), self.pool(input)], 1)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "C:\Users\Sharvari\anaconda3\envs\Lanebirds\lib\site-packages\torch\nn\modules\conv.py", line 338, in forward
self.padding, self.dilation, self.groups)
RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED

哈喽，请问你有自己训练过本算法或者有一些改进的方法尝试过吗，就是这个算法在一些其他的数据表现比较糟糕，如下图


求交流！谢谢！