TensorFlow library was not compiled to use SSE3, SSE4.1, SSE4.2, AVX in the Google Cloud Platform console

Question

I have a TensorFlow model for testing a neural network with a wide range of n-deep, but I cannot get it to work on my Windows machine due to an error in the TensorFlow library. Now I need to resort to the Google Cloud Platform. I created everything where my python files are loaded, but when I run the code through the console, I get the following messages:

$ python -m widendeep.py -h W 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. 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. 

These messages do not appear on startup in Windows. Is it because I cannot run it through the cloud this way? Should I use gcloud ml-engine local train or gcloud ml-engine jobs submit training my_job? Any guidance on the right way to do this would be great.

Answer 1

These messages do not appear on startup in Windows. Is it because I cannot run it through the cloud this way?

You can run your code as is, or at least if you cannot have anything to do with the warnings you specify.

These warnings suggest that you can get better performance from TensorFlow with the machine architecture in which you use it, since it supports a more complex set of commands than the one that was compiled.

Do I need to use gcloud ml-engine for local trains or gbloud ml-engine? send my_job training?

I am not very familiar with the Google cloud (using Amazon at the moment), but I can say that if you need to use the commands above, this has nothing to do with the warnings listed above.

These warnings relate to the processor's instruction set and how it interacts with the GPU, so the performance improvement may or may not be significant (or even exist) depending on your specific application.

If you want to make sure that you are using the full potential of the equipment on which your program is running, you will need to compile TensorFlow on the platform you are working on (check How to compile Tensorflow with SSE4.2 and AVX instructions? ).