Tensorflow：UnknownError (see above for traceback): Failed to get convolution algorithm.

UnknownError (see above for traceback): Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.

解决方案：

    #----------------------------------------------------------------------------------
    import os

    os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
    os.environ['CUDA_VISIBLE_DEVICES'] = "0,1"
    # 选择哪一块gpu,如果是-1，就是调用cpu
    config = tf.ConfigProto() # 对session进行参数配置
    config.allow_soft_placement = True
    # 如果你指定的设备不存在，允许TF自动分配设备
    config.gpu_options.per_process_gpu_memory_fraction = 0.7
    # 分配百分之七十的显存给程序使用，避免内存溢出，可以自己调整
    config.gpu_options.allow_growth = True
    # 按需分配显存，这个比较重要
    session = tf.Session(config=config)
    #----------------------------------------------------------------------------------