CPU version of tensorflow threw a GPU error ! real strange !

[hora...@gmail.com]

Hey folks, this is really strange. I installed a CPU version of tensorflow (1.1.0) under a conda virtual environment. I ran a test script and it complains about a CUDA Invalid Device error !!! This error also appears in the GPU virtual environment (I installed both CPU and GPU version, also under conda env).

Since this is a CPU version, why on earth would it care about my GPU?

My config :

    ubuntu 14.04

    Anaconda2-4.4.0-Linux-x86_64

    python 2.7.13

    wheel : https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.1.0-cp27-none-linux_x86_64.whl

                                                                                                                                  

Error below :

2017-06-13 17:59:42.255163: 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.

2017-06-13 17:59:42.255210: 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.

2017-06-13 17:59:42.255218: 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.

2017-06-13 17:59:42.255225: 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.

2017-06-13 17:59:42.255231: 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.

2017-06-13 17:59:42.319914: I tensorflow/core/common_runtime/gpu/gpu_device.cc:887] Found device 0 with properties: 

name: GeForce GTX 660 Ti

major: 3 minor: 0 memoryClockRate (GHz) 1.0455

pciBusID 0000:04:00.0

Total memory: 2.95GiB

Free memory: 2.91GiB

2017-06-13 17:59:42.320002: W tensorflow/stream_executor/cuda/cuda_driver.cc:485] creating context when one is currently active; existing: 0x27aa360

2017-06-13 17:59:42.391534: E tensorflow/core/common_runtime/direct_session.cc:137] Internal: failed initializing StreamExecutor for CUDA device ordinal 1: Internal: failed call to cuDevicePrimaryCtxRetain: CUDA_ERROR_INVALID_DEVICE

Traceback (most recent call last):

  File "tensorflow6_session.py", line 18, in <module>

    sess = tf.Session()

  File "/home/rspace/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1193, in __init__

    super(Session, self).__init__(target, graph, config=config)

  File "/home/rspace/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 554, in __init__

    self._session = tf_session.TF_NewDeprecatedSession(opts, status)

  File "/home/rspace/anaconda2/lib/python2.7/contextlib.py", line 24, in __exit__

    self.gen.next()

  File "/home/rspace/anaconda2/lib/python2.7/site-packages/tensorflow/python/framework/errors_impl.py", line 466, in raise_exception_on_not_ok_status

    pywrap_tensorflow.TF_GetCode(status))

tensorflow.python.framework.errors_impl.InternalError: Failed to create session.

[Martin Wicke]

Do you have both installed? This does look like the loading sequence of a GPU version. Have you ever had a GPU version installed someplace else on the machine that may still be around?

--
You received this message because you are subscribed to the Google Groups "Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to discuss+unsubscribe@tensorflow.org.
To post to this group, send email to dis...@tensorflow.org.
To view this discussion on the web visit https://groups.google.com/a/tensorflow.org/d/msgid/discuss/0bbfa7ca-45ee-4bff-b18a-d13660e27c5a%40tensorflow.org.