-1

I get the following error when I run the demo from https://github.com/rbgirshick/py-faster-rcnn/tree/master and all the other steps before demo has been done successfully:

mona@pascal:~/computer_vision/py-faster-rcnn$ ./tools/demo.py
Traceback (most recent call last):
  File "./tools/demo.py", line 18, in <module>
   m from fast_rcnn.test import im_detect
  File "/home/mona/computer_vision/py-faster-rcnn/tools/../lib/fast_rcnn/test.py", line 17, in <module>
    from fast_rcnn.nms_wrapper import nms
  File "/home/mona/computer_vision/py-faster-rcnn/tools/../lib/fast_rcnn/nms_wrapper.py", line 9, in <module>
    from nms.gpu_nms import gpu_nms
ImportError: libcudart.so.7.0: cannot open shared object file: No such file or directory

I have the following system settings:

CuDNN V4
$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17
$ uname -a
Linux pascal 3.13.0-62-generic #102-Ubuntu SMP Tue Aug 11 14:29:36 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
$ lspci | grep -i nvidia
03:00.0 3D controller: NVIDIA Corporation GK110BGL [Tesla K40c] (rev a1)
83:00.0 3D controller: NVIDIA Corporation GK110BGL [Tesla K40c] (rev a1)
$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 14.04.5 LTS
Release:    14.04
Codename:   trusty
$ gcc --version
gcc (Ubuntu 4.8.4-2ubuntu1~14.04.3) 4.8.4
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

What is the issue and how can it be solved?

Mona Jalal
  • 34,860
  • 64
  • 239
  • 408
  • You have CUDA 7.5 installed and the software you are trying to run is linked against CUDA 7.0. You can either install CUDA 7.0 on your machine and direct your software to use that via environment variables, or you can rebuild the software you are trying to run against CUDA 7.5 – Robert Crovella Aug 31 '16 at 23:50
  • I had built the software with CUDA7.5 too – Mona Jalal Sep 01 '16 at 00:02

3 Answers3

1

While not a neat solution, I ended up changing the paths to use CUDA 7.0. For whatever reason, seems Faster RCNN currently is not compatible with CUDA 7.5 on Ubuntu 14.04. On Ubuntu 15.10 I had it work with CUDA7.5 with the same exact settings!!!!

Mona Jalal
  • 34,860
  • 64
  • 239
  • 408
1

Try This: Worked for me.

export LD_LIBRARY_PATH=/usr/local/cuda/lib64/
Jayant Agrawal
  • 749
  • 8
  • 15
0

A solution may be changing the CUDA_PATH to be your true cuda path in the make.sh file.

Yexun Zhang
  • 1