linux下theano/tensorflow安装cuda支持gpu

本人在安装过程中碰到很多问题，一一记录下来

1.theano运行gpu，测试代码如下

vlen = 10 * 30 * 768  # 10 x #cores x # threads per core
iters = 1000

rng = numpy.random.RandomState(22)
x = shared(numpy.asarray(rng.rand(vlen), config.floatX))
f = function([], T.exp(x))
print(f.maker.fgraph.toposort())
t0 = time.time()
for i in range(iters):
    r = f()
t1 = time.time()
print("Looping %d times took %f seconds" % (iters, t1 - t0))
print("Result is %s" % (r,))
if numpy.any([isinstance(x.op, T.Elemwise) for x in f.maker.fgraph.toposort()]):
    print('Used the cpu')
else:
    print('Used the gpu')

最新版 的theano 运行报错，错误内容如下：

nvcc fatal   : Value 'sm_61' is not defined for option 'gpu-architecture'

错误原因，sm_61需要cuda8.0的才能运行，安装cuda8.0

在安装目录运行如下命令：./cuda_8.0.27_linux.run

Do you accept the previously read EULA?
accept/decline/quit: accept

You are attempting to install on an unsupported configuration. Do you wish to continue?
(y)es/(n)o [ default is no ]: y

Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 361.77?
(y)es/(n)o/(q)uit: n

Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y

Enter Toolkit Location
 [ default is /usr/local/cuda-8.0 ]: 

/usr/local/cuda-8.0 is not writable.
Do you wish to run the installation with 'sudo'?
(y)es/(n)o: y

Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y

Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y

Enter CUDA Samples Location
 [ default is /home/homolo ]: 

Installing the CUDA Toolkit in /usr/local/cuda-8.0 ...

安装之后重新运行测试代码报错：

ERROR (theano.sandbox.cuda): Failed to compile cuda_ndarray.cu: libcublas.so.8.0: cannot open shared object file: No such file or directory
WARNING (theano.sandbox.cuda): CUDA is installed, but device gpu is not available  (error: cuda unavailable)
[Elemwise{exp,no_inplace}()]
Looping 1000 times took 2.026935 seconds
Result is [ 1.23178029  1.61879337  1.52278066 ...,  2.20771813  2.29967761
  1.62323284]
Used the cpu

错误原因是文件找不到，解决办法：

执行如下代码： ldconfig /usr/local/cuda/lib64

2.tensorflow运行gpu，测试代码如下：

hello = tf.constant("hello TensorFlow!")
sess=tf.Session()
print(sess.run(hello))

a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
c = tf.matmul(a, b)
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
print(sess.run(c))

在cuda8.0上，运行报错，错误信息如下：

ImportError: libcudart.so.7.5: cannot open shared object file: No such file or directory

就上面的信息而言，tensorflow不支持cuda8，重新安装cuda7.5

在安装目录执行 ./cuda_7.5.18_linux.run

Do you accept the previously read EULA? (accept/decline/quit): accept     
You are attempting to install on an unsupported configuration. Do you wish to continue? ((y)es/(n)o) [ default is no ]: y
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 352.39? ((y)es/(n)o/(q)uit): n
Install the CUDA 7.5 Toolkit? ((y)es/(n)o/(q)uit): y
Enter Toolkit Location [ default is /usr/local/cuda-7.5 ]: y
Toolkit location must be an absolute path.
Enter Toolkit Location [ default is /usr/local/cuda-7.5 ]: 
/usr/local/cuda-7.5 is not writable.
Do you wish to run the installation with 'sudo'? ((y)es/(n)o): y
Do you want to install a symbolic link at /usr/local/cuda? ((y)es/(n)o/(q)uit): y
Install the CUDA 7.5 Samples? ((y)es/(n)o/(q)uit): y
Enter CUDA Samples Location [ default is /home/homolo ]: 
Installing the CUDA Toolkit in /usr/local/cuda-7.5 ...

安装完成后继续出现如下错误：

ImportError: libcudart.so.7.5: cannot open shared object file: No such file or directory

执行如下代码：ldconfig /usr/local/cuda/lib64

重新运行：

I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcublas.so locally
I tensorflow/stream_executor/dso_loader.cc:102] Couldn't open CUDA library libcudnn.so. LD_LIBRARY_PATH: /data/work/pycharm-2016.1.4/bin:
I tensorflow/stream_executor/cuda/cuda_dnn.cc:2259] Unable to load cuDNN DSO
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcufft.so locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcurand.so locally
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:925] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:102] Found device 0 with properties: 
name: GeForce GTX 1060 6GB
major: 6 minor: 1 memoryClockRate (GHz) 1.7845
pciBusID 0000:01:00.0
Total memory: 5.93GiB
Free memory: 5.47GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:126] DMA: 0 
I tensorflow/core/common_runtime/gpu/gpu_init.cc:136] 0:   Y 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:838] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 1060 6GB, pci bus id: 0000:01:00.0)
b'hello TensorFlow!'
Device mapping:
/job:localhost/replica:0/task:0/gpu:0 -> device: 0, name: GeForce GTX 1060 6GB, pci bus id: 0000:01:00.0
I tensorflow/core/common_runtime/gpu/gpu_device.cc:838] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 1060 6GB, pci bus id: 0000:01:00.0)
I tensorflow/core/common_runtime/direct_session.cc:175] Device mapping:
/job:localhost/replica:0/task:0/gpu:0 -> device: 0, name: GeForce GTX 1060 6GB, pci bus id: 0000:01:00.0

I tensorflow/core/common_runtime/simple_placer.cc:818] MatMul: /job:localhost/replica:0/task:0/gpu:0
I tensorflow/core/common_runtime/simple_placer.cc:818] b: /job:localhost/replica:0/task:0/gpu:0
I tensorflow/core/common_runtime/simple_placer.cc:818] a: /job:localhost/replica:0/task:0/gpu:0
I tensorflow/core/common_runtime/simple_placer.cc:818] Const: /job:localhost/replica:0/task:0/cpu:0
MatMul: /job:localhost/replica:0/task:0/gpu:0
b: /job:localhost/replica:0/task:0/gpu:0
a: /job:localhost/replica:0/task:0/gpu:0
Const: /job:localhost/replica:0/task:0/cpu:0
[[ 22.  28.]
 [ 49.  64.]]

目前最新的theano/tensorflow不能共用一个cuda

如果要在cuda7.5/8.0切换，可以执行如下代码

# rm -rf cuda
# ln -s cuda-8.0 cuda
# ldconfig /usr/local/cuda/lib64