GPU 版 TensorFlow failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILED

如果你是使用 GPU 版 TensorFlow 的话，并且你想在显卡高占用率的情况下（比如玩游戏）训练模型，那你要注意在初始化 Session 的时候为其分配固定数量的显存，否则可能会在开始训练的时候直接报错退出：
2017-06-27 20:39:21.955486: E c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\stream_executor\cuda\cuda_blas.cc:365] failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILED
Traceback (most recent call last):
  File "C:\Users\DYZ\Anaconda3\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py", line 1139, in _do_call
    return fn(*args)
  File "C:\Users\DYZ\Anaconda3\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py", line 1121, in _run_fn
    status, run_metadata)
  File "C:\Users\DYZ\Anaconda3\envs\tensorflow\lib\contextlib.py", line 66, in __exit__
    next(self.gen)
  File "C:\Users\DYZ\Anaconda3\envs\tensorflow\lib\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: Blas GEMV launch failed:  m=2, n=100
         [[Node: MatMul = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/gpu:0"](_arg_Placeholder_0_0/_3, Reshape)]]
这时你需要用下面的方法创建 Session：
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.333)

sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))