search

TRI-ML / packnet-sfm

TRI-ML Monocular Depth Estimation Repository
https://tri-ml.github.io/packnet-sfm/
MIT License
1.23k stars 242 forks source link

MinkowskiEngine cuda error #179

Open mengjiaos opened 2 years ago

mengjiaos commented 2 years ago

Hello! When I was training DDAD datasets, the following problems occurred after running for a period of time: RuntimeError: CUDA error encountered at: /workspace/MinkowskiEngine/src/3rdparty/concurrent_unordered_map.cuh:595: 209 cudaErrorNoKernelImageForDevice no kernel image is available for execution on the device. I have no idea to solve this problem.

VitorGuizilini-TRI commented 2 years ago

There are different versions of the Minkowski Engine, are you using our dockerfile?

EphChem commented 2 years ago

Hello! Thank you for such great work! @VitorGuizilini , I am also getting same error as @mengjiaos:

File "/workspace/packnet-sfm/packnet_sfm/networks/layers/resnet/resnet_encoder.py", line 88, in forward x = (input_image - 0.45) / 0.225 RuntimeError: CUDA error: no kernel image is available for execution on the device

Using 1 RTX 3090 GPU with NVIDIA CUDA version 11.2 installed on my workstation. Could this be causing this error?

shaiSchneider commented 2 years ago

same problem. using your dockerfile

ArduinoHocam commented 2 years ago

Hi, I've already created a new issue: https://github.com/TRI-ML/packnet-sfm/issues/191 Having the similar issue on my side when using your docker image as well. Please can you give a hand? Thanks.