search

anishathalye / neural-style

Neural style in TensorFlow! 🎨
https://anishathalye.com/an-ai-that-can-mimic-any-artist/
GNU General Public License v3.0
5.54k stars 1.52k forks source link

The TensorFlow library wasn't compiled to use SSE、AVX 、FMA？ #112

Closed tss12 closed 6 years ago

tss12 commented 6 years ago

2018-02-05 17:40:45.026070: 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. 2018-02-05 17:40:45.026107: 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. 2018-02-05 17:40:45.026115: 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. 2018-02-05 17:40:45.026123: 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. 2018-02-05 17:40:45.026130: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX512F instructions, but these are available on your machine and could speed up CPU computations. 2018-02-05 17:40:45.026137: 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. killed

———————— when I run the code python neural_style.py --content 'examples/1-content.jpg' --styles 'examples/1-style.jpg' --output 'examples/my1.jpg' ,then give me some warring and killed,it is may be tensorflow`s trouble,but I do not should I do.

anishathalye commented 6 years ago

looks like a memory issue.

tss12 commented 6 years ago

@anishathalye thanks, maybe I should try tensorflow with gpu.

xuzhe0628 commented 6 years ago

This is not an error, just warnings. TF can run faster if you compile it from source. See details in this question.

anishathalye commented 6 years ago

I think he's talking about the "killed" at the end, which is probably from the oom killer. I don't think the question is about the warnings.

xuzhe0628 commented 6 years ago

I see, thanks buddy

ghost commented 6 years ago

I have the same issue @tss12

kthordarson commented 6 years ago

I get this error when using style images that are too large, try smaller or resizing.

tss12 commented 6 years ago

@maitreepasad hello,it may be your mashine memory is too small.

tss12 commented 6 years ago

@kthordarson and how much time does the code take to run?

ghost commented 6 years ago

Yes I think

I ran it on Floyd hub.. ran successfully! Thanks

On Wed, 18 Apr 2018 11:25 tss12, notifications@github.com wrote:

@maitreepasad https://github.com/maitreepasad hello,it may be your mashine memory is too small.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/anishathalye/neural-style/issues/112#issuecomment-382270737, or mute the thread https://github.com/notifications/unsubscribe-auth/Ab6oIpPKC29yt9Mt5mrnn4UcuVm5JLmZks5tptU5gaJpZM4R5OQK .

tss12 commented 6 years ago

@maitreepasad congratulation,only with cpu?

ghost commented 6 years ago

No with GPU Image size approx 500*400 1000 iterations - 7 minutes

On Wed, 18 Apr 2018 11:36 tss12, notifications@github.com wrote:

@maitreepasad https://github.com/maitreepasad congratulation,only with cpu?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/anishathalye/neural-style/issues/112#issuecomment-382272855, or mute the thread https://github.com/notifications/unsubscribe-auth/Ab6oInuHZCunZXyC5D0wyIIIXqSPZ6E1ks5tptfLgaJpZM4R5OQK .