. Advertisement .
..3..
. Advertisement .
..4..
I don’t know what I’m doing wrong, but I’ve already lost a couple of days struggling with this. Here is my command line:
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
sess.close()
import tensorflow as tf
hello = tf.constant('Hello, TensorFlow!')
sess = tf.Session()
print(sess.run(hello))
This returns:
2017-11-02 01:56:21.698935: I C:\tf_jenkins\home\workspace\rel-win\M\windows\PY\36\tensorflow\core\platform\cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2
I don’t have any experience with your cpu supports instructions that this tensorflow binary was not compiled to use: avx avx2 fma. In this case, how should I change?
The cause: You installed tensorflow default distribution without CPU extensions, such as SSE4.1, SSE4.2, AVX, AVX2, FMA,…Another argument is that even with these extensions CPU is much more slower than a GPU, and a medium- and large-scale machine-learning training is expected to be performed on a GPU.
Solution: If your GPU has AVX support enabled, most of the more expensive operations will be sent to a GPU device (unless you specifically set it not). You can ignore this warning in this instance.
Or set
export TF_CPP_MIN_LOG_LEVEL=2
on Unix. These annoying warnings won’t appear if Tensorflow is running fine. You don’t need a GPU but want to make the most of your CPU . Use the source optimized for . Your CPU should have AVX, FMA, and AVX2 enabled. This topic has been covered in this question , and this issue . Tensorflow uses an ad hoc build system called Bazel. Although it’s not easy to create, it is definitely possible. This will make the warning disappear and tensorflow should perform better.This command will update the tensorflow binary on your CPU & OS.
You can find the download URL for the whl file here
https://github.com/lakshayg/tensorflow-build