简体   繁体   English

在virtualenv中使用CUDA为Theano服务

[英]Use CUDA in virtualenv to serve Theano

Lacking root permissions I installed Theano and Lasagne in a python3 virtualenv at Ubuntu 14.04. 缺乏root权限我在Ubuntu 14.04的python3 virtualenv中安装了Theano和Lasagne。

Running some code, I get an ImportError: dnn not available , which as far as I see results from missing CUDA header, although it is installed at /usr/local/cuda-6.5/targets/x86_64-linux/include/cudnn.h . 运行一些代码,我得到一个ImportError: dnn not available ,虽然它安装在/usr/local/cuda-6.5/targets/x86_64-linux/include/cudnn.h ,但我看到丢失的CUDA头的结果。 。

Following the theano documentation I set environment variable CUDA_ROOT=/usr/local/cuda-6.5 , which yields an additional error: theano文档之后,我设置了环境变量CUDA_ROOT=/usr/local/cuda-6.5 ,这会产生一个额外的错误:

ERROR (theano.sandbox.cuda): Failed to compile cuda_ndarray.cu: libcublas.so.6.5: cannot open shared object file: No such file or directory

Is there any way to include this CUDA library into the virtualenv? 有没有办法将这个CUDA库包含在virtualenv中?

Finally found the answer. 终于找到了答案。 In addition to environment variables I forgot to configure the .theanorc in my home directory like the theano documentation suggested: 除了环境变量,我忘了在我的主目录中配置.theanorc ,如theano文档建议:

$ cat /home/user/.theanorc
[global]
floatX = float32
device = gpu0

[cuda]
root = /usr/local/cuda-6.5

Thanks anyway! 不管怎么说,还是要谢谢你! Learned the lesson to read docs carefully ... 学到了很多东西,仔细阅读文档......

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM