2017-08-19 09:02:27.038166: 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.
2017-08-19 09:02:27.038428: 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.
2017-08-19 09:02:27.038597: 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.
These are just warnings. They just inform you that tensorflow can be faster on your machine if you build from source code. By default, these instructions are not enabled on available versions, which I think may be related to more CPU compatibility import os TF_ CPP_ MIN_ LOG_ Level is a tensorflow environment variable responsible for logging. It is required to set the forbidden info log to 1, the filter warning to 2, and the Forbidden error log (not recommended) to 3
os.environ[‘TF_ CPP_ MIN_ LOG_ LEVEL’]=’2′
Similar Posts:
- Compilation of SSE/AVX/FMA instruction set in tensorflow CPU environment
- [Solved] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
- PHP: using Zend to encrypt the source code, Zend guard installation and Zend guard run time support missing
- [Solved] ImportError: libcublas.so.9.0: cannot open shared object file: No such file
- [Solved] ImportError: No module named tensorflow
- Summary of WebKit compiling on Windows platform
- [Python Debug]Kernel Crash While Running Neural Network with Keras|Jupyter Notebook Run Keras Server Down
- ImportError: DLL load failed: The specified module could not be found
- [Solved] Error caused by correspondence between tensorflow GPU version number and CUDA
- Failed to resolve: junit:junit:4.12 [How to Solve]