python Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA

缺乏、安全感 2021-11-23 05:12 102阅读 0赞

wu@wu-X555LF:~$ python  
Python 2.7.12 (default, Dec  4 2017, 14:50:18)  
\[GCC 5.4.0 20160609\] on linux2  
Type "help", "copyright", "credits" or "license" for more information.  
>>> import tensorflow as tf  
>>> hello = tf.constant('hello,tensorflow!')  
>>> sess = tf.Session()  
2019-07-25 21:42:45.261089: I tensorflow/core/platform/cpu\_feature\_guard.cc:140\] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA  
2019-07-25 21:42:45.403829: I tensorflow/stream\_executor/cuda/cuda\_gpu\_executor.cc:898\] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero  
2019-07-25 21:42:45.404392: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1212\] Found device 0 with properties:  
name: GeForce 930M major: 5 minor: 0 memoryClockRate(GHz): 0.941  
pciBusID: 0000:04:00.0  
totalMemory: 1.96GiB freeMemory: 1.74GiB  
2019-07-25 21:42:45.404441: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1312\] Adding visible gpu devices: 0  
2019-07-25 21:42:47.122397: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:993\] Creating TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 1499 MB memory) -> physical GPU (device: 0, name: GeForce 930M, pci bus id: 0000:04:00.0, compute capability: 5.0)  
>>>

I tensorflow/core/platform/cpu\_feature\_guard.cc:140\] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA

遇到了这个问题，意思是你的 CPU 支持AVX AVX2 （可以加速CPU计算），但你安装的 TensorFlow 版本不支持

解决：  
1. 如果是初学者 或者 没有太大计算速度的需求，在开头加上这两行忽略这个提示即可

import os  
os.environ\['TF\_CPP\_MIN\_LOG\_LEVEL'\] = '2'

说明：

os.environ\["TF\_CPP\_MIN\_LOG\_LEVEL"\] = '1' \# 默认，显示所有信息  
os.environ\["TF\_CPP\_MIN\_LOG\_LEVEL"\] = '2' \# 只显示 warning 和 Error  
os.environ\["TF\_CPP\_MIN\_LOG\_LEVEL"\] = '3' \# 只显示 Error