keras的backend 设置 tensorflow,theano

window 系统环境安装步骤： 1.首先是安装Python，建议安装anaconda 2.安装完anaconda后打开anaconda promp命令行promp，输入conda list. 可以看到已经安装的库以及版本等信息，注意此时没有keras. 3.通过 conda install keras 或  pip install keras 直接安装。（会默认的给你安装keras最新版本和所需要的theano） 4.安装完成之后，就可以打开notebook,输入import keras 检查是否成功。 5.因为windows版本的tensorflow刚刚才推出，所以目前支持性不太好。 但是keras的backend 同时支持tensorflow和theano. 并且默认是tensorflow,因此在win本上需要更改backend为theano才能运行。 这是官网的配置文档：http://keras-cn.readthedocs.io/en/latest/backend/点击打开链接 如果已经运行过一次Keras，你将在下面的目录下找到Keras的配置文件：~/.keras/keras.json 如果该目录下没有该文件，你可以手动创建一个 将文件的默认配置如下：

C:\Users\Administrator>python Python 2.7.12 |Anaconda 4.2.0 (64-bit)| (default, Jun 29 2016, 11:07:13) [MSC v.1500 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. Anaconda is brought to you by Continuum Analytics. Please check out: http://continuum.io/thanks and https://anaconda.org >>> import keras Using TensorFlow backend. Traceback (most recent call last): File "<stdin>", line 1, in <module> File "C:\Anaconda2\lib\site-packages\keras\__init__.py", line 2, in <module> from . import backend File "C:\Anaconda2\lib\site-packages\keras\backend\__init__.py", line 68, in <module> from .tensorflow_backend import * File "C:\Anaconda2\lib\site-packages\keras\backend\tensorflow_backend.py", line 1, in <module> import tensorflow as tf ImportError: No module named tensorflow >>> import keras Using Theano backend. 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.

C:\Users\Administrator>python Python 2.7.12 |Anaconda 4.2.0 (64-bit)| (default, Jun 29 2016, 11:07:13) [MSC v.1500 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. Anaconda is brought to you by Continuum Analytics. Please check out: http://continuum.io/thanks and https://anaconda.org >>> import keras Using TensorFlow backend. Traceback (most recent call last): File "<stdin>", line 1, in <module> File "C:\Anaconda2\lib\site-packages\keras\__init__.py", line 2, in <module> from . import backend File "C:\Anaconda2\lib\site-packages\keras\backend\__init__.py", line 68, in <module> from .tensorflow_backend import * File "C:\Anaconda2\lib\site-packages\keras\backend\tensorflow_backend.py", line 1, in <module> import tensorflow as tf ImportError: No module named tensorflow >>> import keras Using Theano backend. 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.

方法一：将C:\Anaconda2\Lib\site-packages\keras\backend\__init__.py的line 27修改

# Default backend: TensorFlow. #_BACKEND = 'tensorflow' _BACKEND = 'theano' 然后，python-> import keras

方法二： 出现 tensorflow提示错误的话，需要修改下面的位置的内容

[html] view plain copy print ? C:\Users\Administrator\.keras\keras.json  {  "image_dim_ordering":"tf",  "epsilon":1e-07,  "floatx":"float32",  "backend":"tensorflow"  }   C:\Users\Administrator\.keras\keras.json { "image_dim_ordering":"tf", "epsilon":1e-07, "floatx":"float32", "backend":"tensorflow" }将 [html] view plain copy print ? {      "image_dim_ordering": "tf",       "epsilon": 1e-07,       "floatx": "float32",       "backend": "theano"  }   { "image_dim_ordering": "tf", "epsilon": 1e-07, "floatx": "float32", "backend": "theano" }