进程实例化神经网络类报错：self = reduction.pickle.load(from_parent) EOFError: Ran out of input

在一个进程中实例化并调用接口处理图像时报错如下：

THCudaCheck FAIL file=C:\w\1\s\windows\pytorch\torch/csrc/generic/StorageSharing.cpp line=245 error=71 : operation not supported
Traceback (most recent call last):
  File "", line 1, in 
  File "G:\venvs\py364\lib\multiprocessing\spawn.py", line 105, in spawn_main
    exitcode = _main(fd)
  File "G:\venvs\py364\lib\multiprocessing\spawn.py", line 115, in _main
    self = reduction.pickle.load(from_parent)
EOFError: Ran out of input

提供接口的类：

class DetectPerson(object):
    def __init__(self):

        self.det_model = Darknet(current_dir + "/data/yolov3.cfg")
        self.det_model.load_weights(current_dir + '/data/yolov3.weights')
        self.det_model.net_info['height'] = opt.inp_dim
        self.det_inp_dim = int(self.det_model.net_info['height'])
        assert self.det_inp_dim % 32 == 0
        assert self.det_inp_dim > 32
        self.det_model.cuda()
        self.det_model.eval()
   ......
   def detect_person(self, frame):
        """
        将传入的原始图像经过神经网络算出图像中的人并标记返回
        :param frame: 原始图像
        :return: 标记人体的图
        """
         ......
         return frame

调用的进程代码：

class VideoDetect(Process):
    def __init__(self, queue_list):
        super(VideoDetect, self).__init__()
        self.queue_list = queue_list
        **self.detect_person = DetectPerson()**  # 此处实例化报错如上
    ......
    # 运行进程
    def run(self):
		**self.detect_person = DetectPerson()**  # 此处实例化可以正常运行
        while True:
            cam_id, frame = self.queue_list.get()
       		frame = self.detect_person.detect_person(frame)
    ......

先解决然后在这记录下这个问题，有空还需再研究下，有了解原因的小伙伴还望不吝赐教哈哈