Created by: wusj18

按照教程训练fruit数据集，报错'SharedMemoryMgr' object has no attribute '_allocator'，但是机器上还有40G空间的。

2020-06-01 17:35:20,953-INFO: If regularizer of a Parameter has been set by 'fluid.ParamAttr' or 'fluid.WeightNormParamAttr' already. The Regularization[L2Decay, regularization_coeff=0.000500] in Optimizer will not take effect, and it will only be applied to other Parameters!
W0601 17:35:21.420003  8053 device_context.cc:252] Please NOTE: device: 0, CUDA Capability: 35, Driver API Version: 9.0, Runtime API Version: 9.0
W0601 17:35:21.422808  8053 device_context.cc:260] device: 0, cuDNN Version: 5.1.
W0601 17:35:22.815994  8053 device_context.h:155] WARNING: device: 0. The installed Paddle is compiled with CUDNN 7.3, but CUDNN version in your machine is 5.1, which may cause serious incompatible bug. Please recompile or reinstall Paddle with compatible CUDNN version.
2020-06-01 17:35:23,185-WARNING: ./paddle/weights/yolov3_mobilenet_v1.pdparams not found, try to load model file saved with [ save_params, save_persistables, save_vars ]
2020-06-01 17:35:23,618-WARNING: variable yolo_output.1.conv.bias not used
2020-06-01 17:35:23,618-WARNING: variable yolo_output.1.conv.weights not used
2020-06-01 17:35:23,618-WARNING: variable yolo_output.2.conv.bias not used
2020-06-01 17:35:23,618-WARNING: variable yolo_output.2.conv.weights not used
2020-06-01 17:35:23,619-WARNING: variable yolo_output.0.conv.bias not used
2020-06-01 17:35:23,619-WARNING: variable yolo_output.0.conv.weights not used
Traceback (most recent call last):
  File "tools/train.py", line 366, in <module>
    main()
  File "tools/train.py", line 203, in main
    devices_num=devices_num)
  File "/***/layout_detection/PaddleDetection/ppdet/data/reader.py", line 410, in create_reader
    reader = Reader(**cfg)()
  File "/***/layout_detection/PaddleDetection/ppdet/data/reader.py", line 263, in __init__
    use_process, memsize)
  File "/***/layout_detection/PaddleDetection/ppdet/data/parallel_map.py", line 86, in __init__
    self._setup()
  File "/***/layout_detection/PaddleDetection/ppdet/data/parallel_map.py", line 106, in _setup
    self._outq = Queue(bufsize, memsize=memsize)
  File "/***/layout_detection/PaddleDetection/ppdet/data/shared_queue/queue.py", line 63, in __init__
    capacity=memsize, pagesize=pagesize)
  File "/***/layout_detection/PaddleDetection/ppdet/data/shared_queue/sharedmemory.py", line 415, in __init__
    self._setup()
  File "/***/layout_detection/PaddleDetection/ppdet/data/shared_queue/sharedmemory.py", line 418, in _setup
    self._shared_mem = RawArray('c', self._cap)
  File "/***/python_paddle/lib/python3.6/multiprocessing/context.py", line 129, in RawArray
    return RawArray(typecode_or_type, size_or_initializer)
  File "/***/python_paddle/lib/python3.6/multiprocessing/sharedctypes.py", line 60, in RawArray
    obj = _new_value(type_)
  File "/***/python_paddle/lib/python3.6/multiprocessing/sharedctypes.py", line 40, in _new_value
    wrapper = heap.BufferWrapper(size)
  File "/***/python_paddle/lib/python3.6/multiprocessing/heap.py", line 248, in __init__
    block = BufferWrapper._heap.malloc(size)
  File "/***/python_paddle/lib/python3.6/multiprocessing/heap.py", line 230, in malloc
    (arena, start, stop) = self._malloc(size)
  File "/***/python_paddle/lib/python3.6/multiprocessing/heap.py", line 128, in _malloc
    arena = Arena(length)
  File "/***/python_paddle/lib/python3.6/multiprocessing/heap.py", line 77, in __init__
    f.write(zeros)
OSError: [Errno 28] No space left on device
Exception ignored in: <bound method SharedMemoryMgr.__del__ of <ppdet.data.shared_queue.sharedmemory.SharedMemoryMgr object at 0x7f1732470c50>>
Traceback (most recent call last):
  File "/***/layout_detection/PaddleDetection/ppdet/data/shared_queue/sharedmemory.py", line 524, in __del__
    if not self._released and not self._allocator.empty():
AttributeError: 'SharedMemoryMgr' object has no attribute '_allocator'