Windows上的TensorFlow版本1.0.0-rc2：“OpKernel（'op：”BestSplits“device_type：”CPU“'）for unknown op：BestSplits”with test code
我在Windows 7 SP1 x64 Ultimate（Python 3.5.2 | Anaconda自定义（64位））上安装了TensorFlow 1.0.0-rc2版本，使用：
pip install --upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-1.0.0rc2-cp35-cp35m-win_amd64.whl 当我尝试在Eclipse 4.5或控制台中从https://web.archive.org/web/20170214034751/https://www.tensorflow.org/get_started/os_setup#test_the_tensorflow_installation运行testing脚本时：
import tensorflow as tf print('TensorFlow version: {0}'.format(tf.__version__)) hello = tf.constant('Hello, TensorFlow!') sess = tf.Session() print(sess.run(hello))
我得到一些错误信息：
TensorFlow version: 1.0.0-rc2 'Hello, TensorFlow!' E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflob w\core\framework\op_kernel.cc:943] OpKernel ('op: "BestSplits" device_type: "CPU"') for unknown op: BestSplits E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "CountExtremelyRandomStats" device_type: "CPU"') for unknown op: CountExtremelyRandomStats E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "FinishedNodes" device_type: "CPU"') for unknown op: FinishedNodes E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "GrowTree" device_type: "CPU"') for unknown op: GrowTree E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ReinterpretStringToFloat" device_type: "CPU"') for unknown op: ReinterpretStringToFloat E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "SampleInputs" device_type: "CPU"') for unknown op: SampleInputs E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ScatterAddNdim" device_type: "CPU"') for unknown op: ScatterAddNdim E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNInsert" device_type: "CPU"') for unknown op: TopNInsert E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNRemove" device_type: "CPU"') for unknown op: TopNRemove E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TreePredictions" device_type: "CPU"') for unknown op: TreePredictions E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "UpdateFertileSlots" device_type: "CPU"') for unknown op: UpdateFertileSlots
为什么？
我没有这样的问题与TensorFlow 0.12.1（安装pip install tensorflow==0.12.1 ）：
TensorFlow version: 0.12.1 b'Hello, TensorFlow!'
安装今天晚上构build （CPU版本）：
pip install --upgrade http://ci.tensorflow.org/view/Nightly/job/nightly-win/85/DEVICE=cpu,OS=windows/artifact/cmake_build/tf_python/dist/tensorflow-1.0.0rc2-cp35-cp35m-win_amd64.whl
修复了这个问题（没有更多的“OpKernel ('op: ”BestSplits“ device_type: ”CPU“') for unknown op: BestSplits”等）。
现在有一些SSE警告：
TensorFlow version: 1.0.0-rc2 b'Hello, TensorFlow!' 2017-02-15 19:56:22.688266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE instructions, but these are available on your machine and could speed up CPU computations. 2017-02-15 19:56:22.688266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE2 instructions, but these are available on your machine and could speed up CPU computations. 2017-02-15 19:56:22.689266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations. 2017-02-15 19:56:22.689266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. 2017-02-15 19:56:22.689266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations. 2017-02-15 19:56:22.689266: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
在这种情况下，您可以尝试如何使用SSE4.2和AVX指令编译Tensorflow？
TensorFlow 1.0.0是在几天前发布的。 但是，它也有同样的问题。 最近的一个夜晚版本有不同的警告 ：
sess = tf.Session() 2017-02-17 13:01:59.790943: W c:\tf_jenkins\home\workspace\nightly-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE instructions, but these are available on your machine and could speed up CPU computations.
FYI： Tensorflow macOS二进制文件，使用SSE4.1，SSE4.2和AVX优化编译。
要隐藏警告/错误，可以使用 os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' ，例如：
import tensorflow as tf import os os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' print('TensorFlow version: {0}'.format(tf.__version__)) hello = tf.constant('Hello, TensorFlow!') sess = tf.Session() print(sess.run(hello))
TF_CPP_MIN_LOG_LEVEL ：
-
0：显示所有日志（这是默认设置） -
1：过滤掉INFO日志 -
2：另外过滤掉WARNING日志 -
3：另外过滤掉ERROR日志。
参考上面的build议，我认为做2个步骤是有帮助的：
1，升级tensorflow：
pip install --upgrade tensorflow==1.1.0rc1
然后，错误日志变成警告日志：
W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
第二，你可能会抑制2级的警告filter。
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
我认为没有过滤“错误”日志效果很好。
你也许能够抑制级别2的警告filter。在virtualenv安装中，这对于TensorFlow 1.0.1来说是有效的。
os.environ ['TF_CPP_MIN_LOG_LEVEL'] ='2'
Sry有关额外的答案，但我不值得评论。
似乎问题在版本1.1.0rc0和更高版本中得到解决。
查找tensorflow的最新版本：
pip search --version tensorflow
升级tensorflow：
pip install --upgrade tensorflow==1.1.0rc1
