Error: TensorFlow binary was not compiled to use: AVX2 FMA
Following error has been reported when trying to run the script step1_launch_segment.py:
In [2]: run step1_launch_segment.py
2019-11-06 21:59:42.072589: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-11-06 21:59:42.078146: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcuda.so.1
2019-11-06 21:59:42.106865: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-11-06 21:59:42.107535: I tensorflow/compiler/xla/service/platform_util.cc:197] StreamExecutor cuda device (0) is of insufficient compute capability: 3.5 required, device is 2.1
2019-11-06 21:59:42.107683: F tensorflow/stream_executor/lib/statusor.cc:34] Attempting to fetch value instead of handling error Internal: no supported devices found for platform CUDA
Aborted (core dumped)
This error occurs on computers with recent CPUs (even though computation is carried out on GPU). To solve this issue, please follow instructions on following link: https://stackoverflow.com/questions/47068709/your-cpu-supports-instructions-that-this-tensorflow-binary-was-not-compiled-to-u