ubuntu14.04 installation Theano

Basically according to the official website to line:

First

sudo apt-get install python-numpy python-scipy python-dev python-pip python-nose g++ libopenblas-dev git

Again is

sudo pip install Theano

(In this step, if you are python3 by default, you will change to pip-2.7)

Under test:

Test the newly installed packages

Python-c "Import numpy; Numpy.test () "   about ~30spython-c" import scipy; Scipy.test () "   about ~1mpython-c" import Theano; Theano.test () "About ~30m

Speed Test Theano/blas

Python ' python-c ' import OS, Theano; Print Os.path.dirname (theano.__file__) "'/misc/check_blas.py

The test is successful and it's basically available.

The following are the GPU settings:

Check if your computer has CUDA-supported GPUs

Lspci | Grep-i nvidia

Generally get results similar to the following

01:00.0 VGA compatible Controller:nvidia ...

In addition, it is also available

Lspci | Grep-i VGA

Display all graphics cards, using

Lspci-v-S 01:00.0

Displays specific graphics card details.

If so, the installation will be OK:

sudo apt-get install nvidia-currentsudo apt-get Install Nvidia-cuda-toolkit

Then the main is to set the parameters, so that the priority to use the GPU

See the http://deeplearning.net/software/theano/install.html using the GPU section.

There are two main methods, a Setup theano_flags, an edit. theanorc file, for example,

Create a ". Theanorc" file under the $home directory (user home directory, typically/home/user name), edited as

[Global]device = Gpufloatx = Float32[cuda]root =/usr/lib/nvidia-cuda-toolkit    //cuda directory One is set to its own Cuda installation directory (contains bin, Lib,include Sub-folders)

Note that this file you created, will generally become invisible, can be used "ls-a" command to list all files so as to see, can be edited with Gedit.

Also, to set the Cuda Lib subfolder (which may also be set under 64-bit lib64 subfolder) to the environment variable $ld_library_path, see http://blog.csdn.net/xsc_c/article/details/ 23470565 (this blog Cuda a few introductory articles are also very good). Or this article http://www.linuxidc.com/Linux/2014-10/107501.htm and http://www.linuxidc.com/Linux/2012-04/58913.htm.
Test examples in Http://deeplearning.net/software/theano/tutorial/using_gpu.html#using-gpu

From Theano import function, config, shared, Sandboximport theano.tensor as timport numpyimport Timevlen = ten * 768
   # x #cores x # threads per coreiters = 1000RNG = Numpy.random.RandomState 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 '% iters, t1-t0, ' seconds ' print ' Result is ', Rif Numpy.any ([Isinstance ( X.op, T.elemwise) for x in F.maker.fgraph.toposort ()]):    print ' used the CPU ' else:    print ' used the GPU '

　　

ubuntu14.04 installation Theano