说明:系统是unbuntu14.04LTS,32位的操做系统,之前安装了python3.4,如今想要安装theano和keras。步骤以下:python
1,安装pipjson
sudo apt-get install python3-setuptools sudo easy_install3 pip
2,安装g++vim
sudo apt-get install g++
采用上述命令安装g++,安装完成后可用g++ -version查看是否安装完成。注意,若是没有安装g++,在import theano时会出现如下错误:dom
WARNING (theano.configdefaults): g++ not detected ! Theano will be unable to execute optimized C-implementations (for both CPU and GPU) and will default to Python implementations. Performance will be severely degraded. To remove this warning, set Theano flags cxx to an empty string.
搜了一下是由于theano采用g++编译的话速度比较快,在网上找到的大部分解决方案都是基于Anaconda安装的,解决方法是:测试
conda install mingw libpython
3,安装theanothis
sudo pip3 install theano
该命令会自动下载theano所须要的依赖,包括numpy,scipy等等。spa
4,安装keras操作系统
sudo pip3 install keras
最后须要注意的是,keras默认的backend是tensorflow,咱们须要的是theano,因此须要修改下设置。(并且tensorflow用pip3安装,在32位系统上没有对应的版本!用源文件安装又很复杂)code
vim ~/.keras/keras.json { "image_dim_ordering":"tf", "epsilon":1e-07, "floatx":"float32", "backend":"theano" }
5,测试theanoorm
import numpy as np import time import theano A = np.random.rand(1000,10000).astype(theano.config.floatX) B = np.random.rand(10000,1000).astype(theano.config.floatX) np_start = time.time() AB = A.dot(B) np_end = time.time() X,Y = theano.tensor.matrices('XY') mf = theano.function([X,Y],X.dot(Y)) t_start = time.time() tAB = mf(A,B) t_end = time.time() print("NP time: %f[s], theano time: %f[s] (times should be close when run on CPU!)" %( np_end-np_start, t_end-t_start)) print("Result difference: %f" % (np.abs(AB-tAB).max(), ))