Question

我正在尝试从源安装Tensorflow，因为当我使用pip和virtualenv安装时没有安装CPU指令（SSE，AVX ......）的警告：

INFO [PO-thread-0] 2017-04-02 19:26:43,176 DbInfoRollupPlugin.java:196 - Error retrieving node level db summary java.util.concurrent.TimeoutException: null at java.util.concurrent.FutureTask.get(FutureTask.java:205) [na:1.8.0_112] at com.datastax.bdp.plugin.DeferringScheduler$DeferringTask.get(DeferringScheduler.java:115) ~[dse-core-5.0.5.jar:5.0.5] at com.datastax.bdp.reporting.snapshots.db.DbInfoRollupPlugin$DbInfoRollupTask.doRollup(DbInfoRollupPlugin.java:192) [dse-core-5.0.5.jar:5.0.5] at com.datastax.bdp.reporting.snapshots.db.DbInfoRollupPlugin$DbInfoRollupTask.run(DbInfoRollupPlugin.java:173) [dse-core-5.0.5.jar:5.0.5] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [na:1.8.0_112] at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) [na:1.8.0_112] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180) [na:1.8.0_112] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294) [na:1.8.0_112] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) [na:1.8.0_112] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) [na:1.8.0_112] at java.lang.Thread.run(Thread.java:745) [na:1.8.0_112]

所以我试图从源代码安装Tensorflow，以便可以使用Tensorflow编译这些指令。

除了Tensorflow网站上的说明外，我是否需要执行任何操作以使用这些说明进行编译？（W 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. W 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. W 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. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.？）
如果是这样，我需要做什么以及我该如何做？
在我必须安装pip包的安装中，Tensorflow网站说对于Linux，命令是： ./configure它说这个命令会因机器而异，MacOS的等效命令是什么？如何进行Bazel sudo pip install /tmp/tensorflow_pkg/tensorflow-1.0.1-py2-none-any.whl步骤？

旁注：我已经被告知有几种方法可以用Bazel做到这一点，但由于对Bazel缺乏经验并且在终端上没有太多经验，我需要明确的步骤来做什么？

OS Sierra（10.12.2）

Answer 1

对于您的第一个问题（和第二个问题），。/ configure的默认设置将为您提供针对您的机器优化的指令集。这是通过默认编译器选项-march=native完成的。只需按照here

步骤操作即可

对于第三个，一旦在whl（或其中任何地方）生成了/tmp/tensorflow_pkg文件，请查看该目录中文件的名称。它将与说明中给出的linux示例类似但不同。然后，您应该可以使用pip install /tmp/tensorflow_pkg/YOUR_FILENAME.whl安装它。如果您遇到任何问题，可以按照/tmp/tensorflow_pkg/YOUR_FILENAME.whl步骤操作，但使用tensorflow代替tensorflow-gpu或didReceiveRemoteNotification

祝你好运。

Answer 2

在OSX上你可以做到

brew install tensor-flow

我相信。

从哪里添加CPU指令从源安装Tensorflow？

2 个答案: