Linux下安装theano

0.install theano                                                                                                                                                        

 1.sudo apt-get install python-pip

 2.sudo apt-get install python-dev python-nose g++ libopenblas-dev git 

 3.sudo pip install theano

(if step 3 not work, change source url of pip(eg. douban),link url is  
http://ju.outofmemory.cn/entry/108631
)

 

1.user theano with gpu 

 

在用户的～目录下新建配置文件sudo vi ~/.theanorc

添加如下内容：

[global]

floatX=float32

device=gpu

 

新建文件test.py,将以下code贴到该文件中。

  from theano import function, config, shared, sandbox

import theano.tensor as T

import numpy

import time

 

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 xrange(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')

 

运行python test.py

 

result中显示use gpu就成功了。^_^

reference link:http://www.myexception.cn/cuda/2017261.html