back2future.pytorch icon indicating copy to clipboard operation
back2future.pytorch copied to clipboard

How can I fix 'back2future_kitti.pth.tar' error?

Open ahmetybilgin opened this issue 5 years ago • 3 comments

After making the necessary institutions, I ran demo.py. The error I encountered is this> FileNotFoundError: [Errno 2] No such file or directory: 'pretrained / back2future_kitti.pth.tar'. When I look at it with normal access it gives an error > an error occurred while loading the archive. Is the error caused by the file's corruption?

Secondly, the output I got when I ran the line you gave as test > python3 test_back2future.py --pretrained-flow path / to / pretrained / model --kitti-dir path / to / kitti / 2015 / root Traceback (most recent call last):    File "test_back2future.py", line 8, in      from path import Path ImportError: cannot import name 'Path'

What solution can I apply for this? Thank you from now. @anuragranj @JJanai

ahmetybilgin avatar May 20 '20 12:05 ahmetybilgin

pip3 install path.py

anuragranj avatar May 20 '20 20:05 anuragranj

Hi the first error is caused because we forgot to change the file name in the code. Please change the file path to 'pretrained/b2f_kitti.pth.tar' and then it should work.

In addition, please install the path package as suggested by Anurag. I hope this will solve your problems. Best Joel

JJanai avatar May 21 '20 11:05 JJanai

Thank you so much @JJanai @anuragranj . I didn't think the filename would be different. I did what you said and the mistakes are gone. But now I'm getting cuda error. I wonder what I need to solve this?

RuntimeError: cuda runtime error (48) : no kernel image is available for execution on the device at /opt/conda/conda-bld/pytorch_1518244421288/work/torch/lib/THC/generic/THCTensorMathPairwise.cu:40

Ubuntu:18.04 Nvidia GeForce 840M (compute capability: 5.0) Nvidia driver:390 cuda:9.0 pytorch:0.3.1
cudnn7.0

Thank you for your help.

ahmetybilgin avatar May 21 '20 15:05 ahmetybilgin