TensorFlowGitHub地址
TensorFlow支持Python2.7和3.3以上版本,本文使用Python3.5,下载并将它添加到路径当中(在安装提示最下面选项打钩即可).
阅读MD文档,找到Installation->DownLoad and Setup->Pip installation on Windows
文中提示说这个TensorFlow需要一个MSVCP140.DLL文件,你当前系统可能没有安装,应该安装 Visual C++ 2015 redistributable (x64 version).按照提示一路Next就好.
开始安装:
打开命令行提示符,win+R输入cmd,打开即可,需要安装TensorFlow的CPU和GPU
CPU安装:pip install –upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-1.0.0-cp35-cp35m-win_amd64.whl 复制粘贴到命令提示符,稍等片刻,就会自动安装,安装过程中有很多类似进度条的格子,代表进度;
GPU安装: pip install –upgrade https://storage.googleapis.com/tensorflow/windows/gpu/tensorflow_gpu-1.0.0-cp35-cp35m-win_amd64.whl 同样复制粘贴,同样稍等片刻,同样自动安装,同样有类似进度条格子
Cpu和GPU安装好之后,下面会有Test the TensorFlow installation来检测TensorFlow的安装情况,还会在安装一个CUDA:下载地址
安装cudnn 下载地址,将其拷贝到CUDA的安装目录下,里面3个文件夹分别有一个文件,将其按照目录拷贝到CUDA的各个对应的文件中。
将CUDA的路径添加到环境变量中…
检测是否安装成功:
在命令行中打开python 环境
- 12345678910111213> python>> >>> import tensorflow as tf> >>> hello = tf.constant('Hello, TensorFlow!')> >>> sess = tf.Session()> >>> sess.run(hello)> Hello, TensorFlow!> >>> a = tf.constant(10)> >>> b = tf.constant(32)> >>> sess.run(a+b)> 42> >>>>
如果正确显示以上内容,就是成功了.
对抗网络学习指南,地址
安装 numpy +mkl,地址
找到numpy+mkl 下载文件的文件,shift+鼠标右键 点击在此处打开命令窗口,输入pip install+下载文件的全名.whl
安装Scipy.whl 地址
安装方法同上。
这个错误 是因为没有安装Pillow ,安装命令 pip3 install Pillow
- 1AttributeError: module 'scipy.misc' has no attribute 'imread'
可能出现的错误,基本上都能在这里找到,请耐心阅读英语,
import tensorflow as tf
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\dso_loader.cc:135] successfully opened CUDA library cublas64_80.dll
locally
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\dso_loader.cc:135] successfully opened CUDA library cudnn64_5.dll lo
cally
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\dso_loader.cc:135] successfully opened CUDA library cufft64_80.dll l
ocally
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\dso_loader.cc:135] successfully opened CUDA library nvcuda.dll local
ly
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\dso_loader.cc:135] successfully opened CUDA library curand64_80.dll
locally
hello = tf.constant(‘Hello,TensorFlow’)
sess = tf.Session()
E c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stre
am_executor\cuda\cuda_driver.cc:509] failed call to cuInit: CUDA_ERROR_NO_DEVICE
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_diagnostics.cc:158] retrieving CUDA diagnostic information for host: snow-PCI c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_diagnostics.cc:165] hostname: snow-PC
Win7-64位系统安装TensorFlow
真诚地希望能帮到你!