search

cwq159 / PyTorch-Spiking-YOLOv3

A PyTorch implementation of Spiking-YOLOv3. Two branches are provided, based on two common PyTorch implementation of YOLOv3(ultralytics/yolov3 & eriklindernoren/PyTorch-YOLOv3), with support for Spiking-YOLOv3-Tiny at present.
GNU General Public License v3.0
207 stars 58 forks source link

got a cudnn error when running ann_to_snn.py #53

Closed xuhelene closed 1 year ago

xuhelene commented 1 year ago

_Traceback (most recent call last): File "ann_to_snn.py", line 182, in opt.augment) File "/root/code/PyTorch-Spiking-YOLOv3-ultralytics/spiking_utils/snn_evaluate.py", line 276, in snn_evaluate output_snn1, output_snn2 = snn(data) # two branches File "/root/.local/lib/python3.7/site-packages/torch/nn/modules/module.py", line 541, in call result = self.forward(*input, *kwargs) File "/root/code/PyTorch-Spiking-YOLOv3-ultralytics/spiking_utils/spike_dag.py", line 102, in forward print(" op['op'](op_inputs) = ", op'op') File "/root/.local/lib/python3.7/site-packages/torch/nn/modules/module.py", line 541, in call result = self.forward(*input, **kwargs) File "/root/code/PyTorch-Spiking-YOLOv3-ultralytics/spiking_utils/spike_layer.py", line 75, in forward self.groups) RuntimeError: cuDNN error: CUDNN_STATUS_NOTSUPPORTED. This error may appear if you passed in a non-contiguous input.

my environment: python==3.7.6 pytorch==1.3.0 cuda==10.1

xuhelene commented 1 year ago

I searched the internet and found that someone said it might be the memory issue, I changed the timesteps and solved the problem