How to install CUDA 8.0 in the latest version of Tensorflow (1.0) in AWS p2.xlarge instance, AMI ami-edb11e8d and nvidia drivers up to date (375.39)
可以将文章内容翻译成中文,广告屏蔽插件可能会导致该功能失效(如失效,请关闭广告屏蔽插件后再试): 问题: I have upgraded to Tensorflow version 1.0 and installed CUDA 8.0 with the cudnn 5.1 version and the nvidia drivers up to date 375.39. My NVIDIA hardware is the one that is on Amazon Web Services using the p2.xlarge instance, a Tesla K-80. My OS is Linux 64-bit. I get the next error message every time I use the command: tf.Session() [ec2-user@ip-172-31-7-96 CUDA]$ python Python 2.7.12 (default, Sep 1 2016, 22:14:00) [GCC 4.8.3 20140911 (Red Hat 4.8.3-9)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>>